TRUBNER'S
ORIENTAL SERIES.
"Sallantgne
HALLANTYNE, HANSON AND CO.
EDINBURGH AND LONDON
MISCELLANEOUS PAPERS
RELATING TO
INDO-CHINA
AND THE INDIAN ARCHIPELAGO.
REPRINTED FOR THE STRAITS BRANCH OF THE
ROYAL ASIATIC SOCIETY
/ THE "JOURNALS" OF THE ROYAL ASIATIC, BENGAL ASIATIC,
AND ROYAL GEOGRAPHICAL SOCIETIES; THE "TRANSACTIONS"
AND "JOURNAL" OF THE ASIATIC SOCIETY OF BATAVIA;
AND THE "MALAYAN MISCELLANIES."
SECOND SERIES.
VOL. II.
LONDON:
TRUBNER & CO., LUDGATE HILL.
-. 1887.
[All rights reserved.}
CONTENTS OF VOL. II.
VIII. Account of the Malay MSS. belonging to the Royal Asiatic
Society. By Dr. H. N. van der Tuuk .... 1-56
IX. Memorandum of a Journey to the Summit of Gunong Benko 57-69
X. Account of the Island of Bali. By Dr. R. Friederich . . 69-200
XL Notices on Zoological Subjects. By Messrs. Diard and
Duvancel . . ' 200-209
XII. Descriptions of Malayan Plants. By Dr. W. Jack . . 209-295
Notes to this Article. By Sir J. D. Hooker and the Hon.
D. F. A. Hervey 296-302
General and Geographical Index 303-306
Index of Latin Terms 307-309
Index of Malayan and other Oriental Terms 3 I o~3i3
MISCELLANEOUS ESSAYS.
VIII.
SHORT ACCOUNT OF THE MALAY
MANUSCRIPTS BELONGING TO THE ROYAL
ASIATIC SOCIETY.
By H. N. VAN DER TUUK.
["Journal of the Royal Asiatic Society," N.S., vol. ii. p. 85-135. ! ]
A. RAFFLES COLLECTION.
No. i (large folio of 460 pages) contains the *->
sy. About the hero see "Malayan Annals," translated
by Leyden, chapters xiv. and xvi. A small extract is
found in Crawfurd's " History of the Indian Archipelago,"
ii. p. 51. Manuscripts of this work, the text of which
might be available, are in the possession of Mr. J. Pijnappel,
at Leyden ; and of Mr. E. Netscher, at Riyow.* The
last chapters of this tale are found in No. 2,607 f the
manuscripts of the India Office, commencing with that
1 [This article has been revised from the Dutch translation in " Bijclragen,"
III. vol. i. pp. 409-74. In the references to other collections of Malay MSS.
the following abbreviations have been used : I.O. = India Office Library
("Tijdschrift van Nederlandsch Indie," 1849, v l- ' PP- 385-400); B.M.=
British Museum (" Bijdragen," III. vol. vi. pp. 96-101); L. = Leiden (ib., III.
vol. v. pp. 142-78) ; Bat. = Batavia (L. W. C. van den Berg, " Verslag van
eene verzameling Maleische etc. Handschriften." Batavia, 1877.]
* I shall make mention of other copies, as it is my opinion that no Malay
composition ought to be published without a supply of manuscripts bearing on
the same subject. Texts from one manuscript, such as those published by
Mr. J. J. de Hollander, in Holland, are not to be depended upon. Even
Quotations, found somewhere, I shall take notice of, as it may be useful to
the editor of a Malay text to consult them.
SECOND SERIES. VOL. II. B
2 THE ROYAL ASIATIC SOCIETY'S
part where the king of Malaka intends to make one of
his sons king on Mount Siguntang*
This composition is very interesting, as it exhibits a
faithful picture of Malay life, and is written in genuine
Malay, i
No. 2 (large folio of 288 pages; the last four pages
are filled up with doggrel rhymes by some transcriber).
This manuscript appears to be a transcript made by a
native of Java, for a great many words belonging to the
Malay dialect of Java occur in it ; as, for instance,
iiribajig, flower of the hibiscus rosa Sinensis ; bbpcug,
pock-marked ; kulbn, west,| &c. It also abounds with
Javanese titles, as dZmang, ngabc/ii, kandtiruivan, &c.
The manuscript is in many passages too corrupt to be of
use in editing the text. The transcriber has often changed
words he did not understand into such as resembled them
in sound, or nearly so.f But what is very strange, it has
now and then a form less corrupted than the Javanese ;
v.g. nantabdga (p.i88) instead of the Javanese antaboga (a
corruption of the Kavi anantabJiogd). As to the contents,
it follows the Javanese poem only to a certain extent,
whilst it oftens contains passages which are not explicable
otherwise than by supposing that a Javanese original has
been translated or imitated, which did not deviate so much
from the original Kavi poem, as the one published by
Mr. A. B. Cohen Stuart. Although it is evidently taken
from the Javanese, its first and last pages contain matter
not found either in the Kavi or Javanese work, whilst no
trace is found of the introduction, wherein the king, Jdya
Bdya, in whose reign Mpu SedaJi, the Javanese author,
* See No. 66 of my " Kort Verslag der Maleische Ilandschriften in het East
India House, London," where the reader will find a full account of those
closing chapters. As the numbers in that account havr been since changed, I
shall give here the present numbers in the India Office Collection.
1 [B.M. 9 ; L. 1762 ; Bat. 186. Copious extracts are given by G. K.
m in liis " Bloemlezing uit Maleische Geschriften" (1878), vol. i.
p. 1-104; vol. ii. p. 58-120.]
t Even Dutch words, as, for instance, j^J (blaauw, blue), occur in it.
So, for instance, we find passim A-&)^ (Ar.) instead of LpJ (alms,
largesses of a king to priests and religious mendicants).
MALA Y MANUSCRIPTS. 3
lived, is spoken of in laudatory terms ; moreover, the title
Barata yuda (Bharata-yuddha), which is given to the
Javanese version, is not known in Malay ; and the great
war between the Koravas and Pandavas, wherever it is
alluded to in Malay compositions, is always called ptirang
Panddiva Jdya, " the war of the victorious Pandavas."
Not until we reach p. I 34 do the contents of this manu-
script resemble the Kavi and Javanese composition. The
Malay author says in the opening that his work, although
containing the story of the Pandawa Pancha Kalima*
gives a great many beautiful tales in the beginning, and
afterwards the tale named Hikayat Pandawa Jaya. These
beautiful tales are, he says, a collection of Javanese
dramatic compositions (Itilakoii), to which he gives no
particular names. I shall, on another occasion, 1 make an
analysis of the whole and divide it into three parts. The
first will give a rapid view of the contents from pp. I
1 34, being what is not found either in the Kavi or Javanese
work. The second will be more circumstantial, as it may
illustrate the difficult passages of the Kavi original, and
will comprise what is found from pp. 134-208, being the
record of the great war. The third will give only a brief
account of the contents from p. 208 to the end, as it
deviates in this part almost in every respect from the
Javanese version,! which closes with a eulogy of king
Jdya Bdya, of which no trace is found in this manuscript.
The library of the India Office is possessed of two
manuscripts bearing on the same subject, but only con-
taining the description of the war. They are numbered
2,384 (small 4to, 234 pages), and 2,605 (8vo, 176 pages).
Both commence with introducing to the reader the chief
heroes who figure in it, and then speak of Kasna's%
mission to demand the half of the kingdom in behalf of
* Translation
1 [" Tijdschrift voor de T. L. en V.," vol. xxi. p. 1-90.]
f Of the Kavi version only twelve copies (!) have been lithographed by
order of the Dutch government. It is not complete, ending with the combat
of Arjuna and A'svatthama.
J Krisna,
B 2
4 THE ROYAL ASIATIC SOCIETY'S
the five sons of Pandn* To enable the reader to form a
judgment of the difference of the texts of the three
manuscripts, some specimens are here given.
The names of the four holy men (j'isi) that join Kasna
when setting out for Hastinapura as mediator are in No.
2,603 : jlj 1 * (*>>> J l^> < -^-^< a "d ^j*^', in No. 2,384,
j~>y> t^, .Ajw>l^ > ^ft 5 ^ 5 and ^^^^ J and in this
manuscript, j\ , ^ , ^^ , and ^y> f!>t
The passage where the Javanese version speaks of a
human sacrifice being performed by either of the contend-
ing parties runs in No. 2,384 as follows:
&c.
No. 2,603 nas :
&c.
This manuscript has (p. 147) :
&c.
* Lcyrlon ("As. Rcs. !r x. 1 78) mentions the following separate tales about
i, The tale about their gambling ; 2nd, that about their
borrowing a hall ; 3rd, that about their selling lime.
t In the K:ivi poem they are J\nasunlia, A'anva, Janaka, and Narad a
(the Javanese has the same, only differently spelt, according to the Javanese
pronunciation).
MALAY MANUSCRIPTS. 5
The Kavi (x. 6) has : tuwin pa/fa tlas wiakaryya
bhisuweng* tgal paprangan \ rawan ngaran i bang tawur
nripati paudava murwwani \ kuiiaug tawur i sang nripcug
kuru ya Jc aril nd brain nana j rikati sine rind pa sang dwija
sagotra viatiyalaga. " Then they all performed a sacrifice
on the field of battle, Rawan was the name of the victim
of the Pandawa king, commencing ; as to the victim of
the Kuru king, a brahmana was . . . . , thence he was
cursed by the twice born, to die with his [whole] family in
fighting." This remarkable passage will perhaps attract
the attention of some Sanskrit scholar, who may succeed
in explaining it. In No. 2 I (see below) I have not been
able to find it. 1
No. 3 (large folio of 244 pages, imperfect at the end)
contains the t-sU^vJ J^ ^^1 _ %*&) <&^- It is one of
the Panji tales, containing the adventures of Inu Kartapati,
prince of Kuripan. This manuscript commences with the
king of Kun'pan's getting a son, called at his birth
Asmara ning rat Onddkan Jay a. Then the birth is
* Instead of bhisitiva (bhismvang is bhisuwa + ing) a manuscript on palm
leaves in my possession has bhisu-cng (bhisna + ing). I should like to read here
bliisa-i<a (abJiisawd], The Malay text gives no explanation, as it is evidently
influenced by the Javanese version, where sagotra has become the name of a
person. Moreover it identifies Rawan with a son of Arjuna (of the name of
Irawan\ who is afterwards killed by a demon (xii. 17). The word w*wJ u3
in the Malay version is probably a substantive made from wvJU^, which
occurs in the Hikayat Kumala bahrin with the sense of to turn off the evil
influence of a ghost from a person who is supposed to have been visited by
a ghost, and in consequence of it has got some disease (compare the Ngaju-
Dayak palis). .j~J U*9 must then have the meaning of what is used to turn
off the evil influence of ghosts.
1 In the Malay Panchatandaran (ed. Van der Tuuk, p. 46, 1. 17), there
is the following allusion to this human sacrifice : (T
Dharmaraja is the name by which Yu-
dhishthira is known in Malay tales [I.O. 87 ; Bat. 143 ; J. J. dc Hollander,
" Handleiding bij de Beoefening der Maleische Taal- en Letterkunde," 5th
edition, p. 336, No. 2].
6 THE ROYAL ASIATIC SOCIETY'S
related of L&smining purl Cliandra-kirana, the princess
of Daha, also named Puspaning rat* and betrothed
to the above-named prince of Kuripan. This princess,
when yet a girl, was carried off by Batdra Kdla,
and placed with her attendants in a forest, where she
changed her name and that of her waiting- women. The
prince of Kuripan goes, attended by his followers, in
quest of his intended bride, and in his rambles for that
purpose takes the name of Rangga Ariya Kuda Nastapa,
his followers too changing each his name.
No. 4 (folio of 246 pages and ending abruptly) and
No. 73 (small 4to. of 420 pages) both contain the <&le*
^Ujjo ^jb, an imitation of the Persian <3>li J=^L. On
comparing the introduction, where the owner of the
parrot (^j*^ &jjjt*) is spoken of, I found the readings
to be nearly the same. In my possession is a copy
(folio of 90 pages) wherein the parrot tells thirteen tales.
In the library of the India Office there are two manu-
scripts of this composition (Nos. 2604 and 26O6). 1 The
former contains twenty-two tales, but the latter only ten,
whilst the introduction about Khojah Meymun is wanting
in it.f According to Abdu-llahJ the Moonshec this
composition also goes by the name of ^-^-y* ^^p* <&^^
after the parrot's owner. The two manuscripts of the
India Office seem to belong to one and the same version,
and only differ in the proper names, which have been
changed to Malay ones in No. 2606. The versions in
both differ from my manuscript.
No. 5 (folio of 315 pages) contains the ^^\ y\A &Ks>.
It is an imitation in prose of the Javanese poem, the com-
mencement of which has been published by Mr. J. J. de
Hollander in the Reader, p. 158 sqq., at the end of his
" Handleiding bij de Beoefening der Javaansche Taal- en
* Compare under No. 14. 1 [Now 285 and 327 ; L. p. 178 ; Bat. 173, 174.]
t Sec further " Kort Verslag der Maleische Handsclmftcn van hot E. 1.
House," p. 394. [For a Makassar version see B. F. Matlhes, " Kort Verslag,"
No. I, and for one in Bugi, ib. No. 90.
See his Journal, p. 95 of the Singapore edition. Of this Journal there
is also a reprint in the fourth volume of Meursinge's " Maleisch Leesboek ;"
and a French translation by Dulaurier.
MA LA Y MANUSCRIPTS. 7
Lcttcrkunde" (Breda, 1848). A translation, as it would
seem, of the Javanese poem is to be found in Roorda
van Eysinga's " Indiu " (Breda, 1843), p. 502 (3de boek,
eerste deel.). No. 1 1 (folio of 1 5 I pages, only written
half-way down, the open spaces being perhaps intended
for a translation) contains the same tale, but considerably
abridged.
No. 6. See No. 31.
No. 7 (folio) contains :
I. (7 1 pages) <^;L,J^J )*. This poem has been edited
with a Dutch translation and annotations by Mr. R. van
Hoevell, in vol. xix. of the " Transactions of the Batavian
Society of Arts and Sciences," but may be had separately.
A review of this edition is to be found in the " Indisch
Magazijn" and the "Gids" (1847), and quotations from
another manuscript in Roorda van Eysinga's "Maleisch-
Nederduitsch Woordenboek," under ^^ , /*s^ > uk*j|j ,
^>J , )\) , ^Jsx^, , ^JJsx~ , and f~u* . Another copy is
contained in No. 36 (folio of 130 pages, and ending
abruptly). Both manuscripts may serve to correct the
edited text. I subjoin here a specimen of the various
readings :
The printed edition, p. 3, line 9 from below, has ;
Satdlah (baginda sampey}* kapantey \ di Wiatiia pdrahu
(di atas lantey)*\ \ langkap (lah sakaliyaii)\ kajang dan
lantey \ (bdik) lah putdri duduk bdrjuntey \\
Page 5, line 5 from above :
Tidurlah anakku bulang hulu \ biyarlah ayahnda bdrjalan
ddhulu || (anakku pandang) IF hatiku pilu \ bagey di Juris
dangan sambilu \
Page 6, line 4 :
Sambilan bulan sdmbilan hari \ (kit kandung)** di dalani
(Jiutan duri)\\\
Page 6, line 2 from below :
* No. 7 has sampey baginda, and No. 36, sampey luwan turun.
t No. 36, tarlalu basey. J No. 36, dan.
Nos. 7 and 36, correctly ndik.
H No. 7, anak kupandang. ** No. 7, kukandung.
ft No. 7, rightly diri.
THE ROYAL ASIATIC SOCIETY'S
Bdrjalan lali baginda (laju mannliJi} * \ r as an a Jidndak
(bdrbalik}\ kambaliV
II. (69 pages) ^>y+ ^f f. This poem has been
twice edited by J. J. de Hollander, once in the Reader of
the first edition of his " Handleiding bij de Beoefening der
Maleische Taal- en Letterkunde," and once separately
(Leyden, 1856), from an untrustworthy transcript evi-
dently made in Java. The version of this manu-
script has hardly anything in common with that of the
one edited, but corresponds in many respects with that
recension of the poem, from which Marsden has given
extracts in the Reader at the end of his Grammar. The
king mentioned in the opening is called here Sari nara-
inddra di CJidmpaka Jajar, but in the one edited Ratu
Socha windu pura nagara. As proper names of females,
Kin Tddahan and Kin Pangalipnr occur here ; whilst the
name of the heroine is sometimes shortened into Kin
Tdbnh, for the sake of rhyme and metre. Wira Dandani,
Wira Pdndapa and Wira Kdrta are found as proper
names of males. The hero, the prince who fell in love
with the heroine, goes here by the names of Pnspa
Kdnchana, Raden Inn,\ Anak Manidri, Inu Bangsawan,
Radcn Inn Kartapati,^ and Anak Inn. The beauty of
the heroine is compared to that of Januwati^ the goddess
of love (yangyang kdsmna), and the celestial nymph Nila-
ntama. Instead of taman (garden), this manuscript makes
often use of the Kavi Idlangun** Padnka Mahddewi,
Paduka Matnr (?), and Paduka Liku are mentioned as
inferior wives of the old king. The pages of the king
employed on errands are called here pangdlasan, instead
* A correction by the editor instead of the words of the manuscript, sayans*
tarjalan. No. 7 has the true reading (sayang tarjali). t No. 7, balik.
1 [L. p. 178 ; Bat. 256 ; de Hollander, 1. 1. p. 309, No. 2. A Makassar prose
work of that name is described by Matthes, 1. 1. No. 9.]
$ Rhyming on iarmangii. See also the extracts in the Reader of Marsden's
Malay Grammar.
A name of I^anji.
' The name of Samba's sweetheart (see under No. 15).
** e.g., mari-lah dmas arinitt^sjtn, kita mandi kalalangitn, and tdngah hart
baginda ban^tn, fiirgi mandi kalalangun. The native tales speak always of
delightful gardens, where a bathing-place is one of the first requisites.
MALA Y MANUSCRIPTS. < }
of badnwanda. This version, moreover, docs not end so
tragically ; Indra bringing the two lovers back to life,
accosting the heroine with anakgaluh** One of the charac-
ters represented is Si-Tnguk,\ who is described as a kind
of Falstaff, big-bellied and fond of fun. There is also a
version in prose which goes by the name of ^l^ol <&ls>
c^s;^J. In this version, the heroine is the daughter of a
king of Wanggar, and the waiting-maid, who dies with
her, is called here, as in the printed edition, Kin Bayan.
One of her most beloved nurses has the name of Antamsmi,
and is addressed by her with kakak or the Javanese cmbbk
(elder sister). The principal attendants of the hero are
Panta Wira Jay a and Jaran Angsoka. The place where
the heroine is killed is here the wood (Jav. alas) Puchangan.
The lovers are brought to life by Bdtara Kala, who
changes them into lotus flowers, and then veils them in
a cloud of incense. The residence is called in the end
Sochawindn, but elsewhere or\\yPura nagara. The late Mr.
H. C. Millies, at Utrecht, had a manuscript of this version.
It is not worth publishing, but may be available for a new
edition of the poem, of which there is a manuscript also
in the library of King's College, if I recollect right. It
is beyond all doubt that the poem as well as the tale
belong to the widely spread cycle of the tales in which
the adventures of Panji are related. :
III. (26 pages) ^J^ Jo-L* j*&. This poem is known
on the west coast of Sumatra by the name of <^s\~> \*
c - j >rV */
itt^*. I possess two manuscripts of it (Svo of 36 pages,
and small 4to of 68 pages). A prose version of it is
contained in No. 2,715 of the manuscripts of the India
* In the Panji tales the princess of Daha, the intended wife of Chekel, is
commonly called Raden-galnh.
t One of the personating characters in the Panji tales.
See Raffles' "History of Java," ii. p. 88 sqq., i. 335 and 392 ; Cohen Stuart's
Djaja Lengkara, and Roorda's Lotgevallen van Raden Pandji, in the "Bijdragen
tot de taal- land- en volkenkunde van Nederlandsch Indie," vol. ii. p. 167 sqq.,
and vol. vii. nieuwe volgreeks, p. i sqq. [Bat. 247 ; L. p. 178 ; de Hollander,
1. 1. p. 308, No. i.]
In the end of this manuscript this proper name of the mother of the heroine
is spelt
io THE ROYAL ASIATIC SOCIETY'S
Office, and has been described in my " Kort Verslag der
Mai. Handschriften van het E. I. House." 1
IV. (io pages) l^J ^<jj j*,. This is a collection of
erotic verses put into the mouths of two fishes (a tambara
and a kakaf) who seem to be desperately in love. 2
No. 8. See under No. 17.
No. 9 (folio of 1 60 pages) contains, as do No. 37
(151 pages, and ending abruptly) and No. 55^ (small
4to of 262 pages) \jyj> I^Jj) <k\&*. This tale contains
the adventures of Indara Putara, son of Bakdrma Puspa^
king of Samanta-pura, and is replete with wonderful
narratives. The hero is carried off by a golden peacock ;
is sent by the king Shahsiyan to Barma Sdqti ; kills
a giant or demon on Mount Indara Kila ; \ finds the
wonderful sea in the midst of the world (tasik samu-
ddrd) ; meets with the princess Kumdla Rdtna Sari ;
contends with the prince Lela Mangdrna in exhibiting
supernatural feats ; is carried off by a genie (of the name
of ^j^J), whose son (called ^jJe^J) he kills ; meets
consecutively with mountains of gold and other precious
metals, the seas of wonder and love ; journeys in a
cavern during a month ; kills a serpent (of the name of
J^Joo), and a demon (called Ujfj>j) ; meets with Ddrma
Gangga, who instructs him in supernatural means of
conquering his enemies, and with Barma Sdqti, &c. At
last our hero comes home, and is made king of Samanta-
fjira with the title of Sultan Indara Manginddra. No. 5 5
terminates with a great many erotic verses not found in
the other copies. Many quotations from this work are to
be found in Werndly's " Maleische Spraakkunst " (pp .1 33,
157, 162, 170 (twice), 171, 174 (three times), 176, 185,
1 8 6, 191, 193, 194 and 195), in Roorda van Eysinga's
1 [Now No. 292 ; Bat. 251, 252; de Hollander, 1. 1. p. 317, No. 66.]
8 [Lithographed at Singapore, a. H. 1291.]
* Another copy is in the possession of Dr. Reinhold Rost (small 4to, 148
pages).
1* Of course most proper names occurring in this account are transliterated
guessingly, such as they would be pronounced by a Malay at first sight.
% "Where Arjuna performed penance to get supernatural arms. Such proper
names deserve being taken up in a Dictionary, as they occur very often.
MALA Y MANUSCRIPTS. 1 1
"Maleisch-Nederduitsch Woordenboek" (under pantas.pan-
dey, puji and gftmk\ and in the annotations of Mr. van
Hocvell on the Sair Bidasari (pp. 289, 305, 333, 335,
348352,375 and 399). Specimens of the reading of
the three manuscripts :
No. 9 :
eb
eb
No. 37 :
No. 55:
1 [L. 1690, and p. 178; Bat. 168; B. F. Matthes, "Kort verslag van
^lakassaarsche en Boeginesche Handschrifcen," Nos. 13 and 94.]
12 THE ROYAL ASIATIC SOCIETY'S
No. 10 (folio) contains a collection of transcripts of
treaties between the Dutch E. I. Company and several
native States in the Indian Archipelago. The first treaty
is that between Admiral Speelman and the king of Gowa
(Mangkasar), and the last that between the E. I. Com-
pany and the king of Johor and Pahang.
No. ii. See under No. 5.
No. 12 (folio of 444 pages) contains the l*^J <^j\&.
^Ijvg-i*. It is very seldom that tales are divided into
chapters (J.xsj), of which there are here sixteen. In the
commencement of the tale there is a kind of summary,
wherein the hero is said to be a great king, who visited
Mount Qaft China, and the land of the inferior gods
(dcwa), subjecting men and ghosts to his sovereignty.
On p. 2 a State Samanda-puri is mentioned. Its king
was called Sariyawan, and was sprung from Indara Dewa
Mahardma Rupa, whilst his queen was of mere mortal
extraction. He had two sons called Raja Ardan and
Raja Marsddan. The two princes went with a large
retinue to the forest Samanta Baranta, where a dewa of
the name Sardma Dewa was in the habit of enjoying
himself. This god hated the king, their father, who had
caused his residence to be destroyed in former times.
He changed himself into an old man and visited the
princes, saying that he wished to serve them. Contriv-
ing to separate them from their followers when engaged
in hunting, the god transformed himself into an elephant,
whom Ardan so hotly pursued, that he got the start of
his brother, and at last found himself entirely alone.
The god then flew away with the prince to the sky,
but was killed by the young hero. Ardan, having arrived
again on this sublunary orb, made the acquaintance of a
risi called Bdyu Rama, who told him that he was not to
revisit his country for many years. The prince remained
in the dwelling of the holy man, who instructed him in
all sorts of supernatural sciences. Marsddan goes in
quest of his brother, and in his rambles arrives at
Inddrapura, where he marries the king's only daughter,
MALA Y MANUSCRIPTS. 1 3
and succeeds his father-in-law.* Ardan has a great
many adventures of the same kind, delivering a princess
with her waiting women, &c. Banna SJiaJidtin, the hero
of the tale, is a son of Marsddan Shah, king of Kalingga
deiva^ and his eldest brother is called here Rdjadirdja.
This work is replete with pantuns, some of which are
worthy of notice. The late Mr. P. P. Roorda van
Eysinga possessed a manuscript (two volumes in folio),
which he would have published, but for want of a
sufficient number of subscribers : what has become of it
I cannot say. J. J. de Hollander (" Handleiding bij de
Beoefening der Mai. Taal- en Letterkunde," 3rd edition,
p. 332) says, I know not on what authority, that the
author was SJieikJi Ibn Abu Omar. 1
No. i 3 is a number I could not find. Dulaurier has
also omitted it in the list he gives (" Journal Asiatique,"
3rd series, x. 69) of titles of the manuscripts of this
collection.
No. 14 (folio of 456 pages : on the back of the cover,
Charang Kurina) contains the ^.jJ^ ;Us &le*. It is a
tale belonging to the Panji cyclus. The commencement
is about the king of Kuripan having two sons, the eldest
being Kdrta Buwdna, and the youngest Raden Asmara
Jaya, surnamed Ondakan Rawisarangga, who was betrothed
to the princess of Daha, called Raden Puspita-ning Rat.\
The name by which this tale goes is the assumed name
of the princess when she had fled from her father's
residence in order to follow the prince, in the garb of a man.
No. 15 (small folio of 180 pages) contains the <&l5C^
fV- cbW-* The plot of this tale is nearly the same as that
of the Bhaumakdvya,\ relating the adventures of Boma
* Called Bakdrma Dali raja. The proper name Bakdrma is very frequent
in Malay tales, and is a corruption of the Sanskrit vikrama; it is often con-
founded with Pakarma.
t The manuscript has^J >t ^ c^ji (p. 32).
1 [According to the same authority (5th edition, p. 353), the printing was
actually commenced in 1856, but discontinued in consequence of the editor's
death. See also L. p. 178.]
% Compare under No. 3. [B.M. 8.]
Edited by Friederich in the "Transactions of the Batavian Society/'
i 4 THE ROYAL ASIATIC SOCIETY'S
(the Sanskrit Bhaiuna, son of the earth). He was the son
of Bisnu ( Vismi) by the goddess Pdrtiwi (Sans, prithivl,
earth), and became a powerful king, whom even the gods
stood in dread of. As he, demon-fashion, annoyed the
penitents, Kdsna (Kris no) sends his son Samba against
him. Boma is at last killed by Hanoman, after having
himself killed Samba and Arjuna, who were, however,
called into life again by Nardda (Naradd) sent by
Batara Guru for the purpose. The celebrated episode*
of Ddrmadeiva and Ddrmadewi is here inserted in the
same way as in the Kavi poem, Ddrmadewa following
Bisnu when incarnating himself into Kdsna and becoming
Samba, whilst Ddrmadewi, after having burnt herself,
becomes Januwdtifi and so is reunited to her former love.
This tale is also named <~-^~> ~, &\&*. The R.A.S.'s
MS. (see also under No. 21) slightly differs from the one
in the India Office (No. 2905,* 4to, 120 pages). Raffles
("History of Java," i. p. 388, first edition) mentions the
Javanese version under the titles Buma Kalantaka and
Embatali. The first name is no doubt Bhaumakalantaka
(the death of the demon Bhauma, kala being used in
Javanese to denote demons and Titans), as may be
inferred from the Kavi poem, p. 233, where it is Bhau-
mdntaka (Bhauma's end, the hero dying by the hand
of Vismi). The Kavi version bears (ap. Raffles, 1. 1.) the
name Anrakasura, which is to be corrected into Nara-
kasura (the demon Naraka, another name of Bhauma). I
shall give on another occasion an analysis of this Malay
composition. 3
No. 1 6 (folio of 206 pages). A duplicate is No. 62, I.
(158 pages). The two manuscripts differ but slightly.
They contain the ^~q U~J dJ.Ke*. The work has been
edited by Mr. P. Roorda van Eysinga (Batavia, 1821), who
* This episode is often alluded to in Malay tales and poems (comp. under
No. 7, II.).
t Y<ij nawaii is her constant name in the Kavi poem, where she is never
called Ddrmadewi.
1 [Now 87.]
a [This account is to be found in the Batavian "Tijdschrift," vol. xxi. pp.
91-101. Cf. Bat. 142.]
MALAY MANUSCRIPTS. 15
has also given an analysis of it in the tenth volume of
the " Transactions of the Batavian Society." The episode
of the singing peacocks has been published from another
version by Meursinge in the third volume of his " Maleisch
Leesboek." In the library of the India Office there are
two manuscripts, Nos. 2429 and 2430 ( ? ). Mr. J.
Pijnappel has also a manuscript. A new edition of this
work is desirable, as that by Roorda van Eysinga has
long been out of print. Quotations from it are found in
Werndly's "Maleische Spraakkunst," pp. 142, 157, 170,
171, 172, 1 80, 182, and in the preface xl., xli. 1
No. 1 7 contains :
I. The ;th book of the ^J=>,LJ) ^l^j (drtina kdbon
sagala raja raja). No. 42 (folio of 440 pages : on the
back, Makota sagala raja raja) contains but five books of
this work, and No. 8 (folio of 367 pages) only four books
and a few pages of the fifth ; this copy is written with
vowel-signs." This excellent work, complete copies of
which are very rare, is divided into seven books, each
book containing a certain number of chapters. The
author calls himself Nuru-ddlu ibn 'All ibn Hasanji, son
of Muhammad, of the Hamid tribe, and a native of Ranlr
(see No. 78, IV.), and he composed it at Achih (Acheen)in
the year of the Muhammadan era 1040, by order of
Sultan Iskander II. Aliyu-ddm Mu'rayat* SJiah Johan
bar dawlat lillu-llahi fiTalam:\ The first book (many
chapters) treats of the creation of heaven and earth ; the
second (many chapters) is about prophets and kings ; the
third (six chapters) on just kings and clever ministers ;
the fourth (two chapters) on pious kings and holy men ; \
the fifth (two chapters) on unjust kings and foolish
ministers ; the sixth (two chapters) on honoured liberal
1 [I.O. 89, 90 ; B.M. 4; L. 1693 ,1737, 1747 ; Bat. 170-72 ; de Hollander,
1. 1. p. 352, No. 80. On the Buginese redaction, see Matthes, 1. 1. p. 32,
No. 98, and p. 95, No. 12366.]
3 [See also No. 70, I.]
t The shadow of God on the world
16 THE ROYAL ASIATIC SOCIETY'S
men and heroes ; the seventh (five chapters) on intelli-
gence, and on all sorts of sciences, medical, physiognomical,
historical, c. Everywhere a great many tales are given,
which might be used for a new Malay Reader. This
work gives more than it promises, which in Malay
literature may be called a miracle. The twelfth chapter
of the second book contains a summary of the history of
Malay States ; the thirteenth the history of Achih up to
the time of the author. In the first chapter of the fourth
book there are several tales about the celebrated Ibrahim
Ibn Adham, corresponding to some extent with the tale,
published by Mr. P. Roorda van Eysinga (Batavia, 1822)
and D. Lenting (Breda, 1 846) under the title " Geschie-
denis van Sultan Ibrahim vorst van Irak."* I have in my
possession a manuscript (4to, 194 pages) containing only
the first four chapters of the seventh book.i
II. An incomplete copy of ^^JsXJ! gto' (see under No.
42).
No. 1 8 (folio of 202 pages) jjJu* s^^s^. This col-
lection of historical tales has been published for the
greater part by Dulaurier in his " Chroniques Malayes,"
and translated by Leyden (" Malay Annals," edited by Sir
Stamford Raffles). This MS., however, contains chapters
not found in other copies, and not translated by Leyden.
The last chapter but one, for instance, is about Sang
Naya's conspiracy against the Portuguese at Malaka.
There are several versions of these chronicles, as the
reader will see from the various readings in Dulaurier's
edition. No. 35 (folio of 117 pages), No. 39 (folio 120
pages), and No. 68 (small 4to) end with the death of
Hang Kasturi, the last named number having besides an
entirely different introduction, and being properly but an
abridgment. In No. 76 (small 4to) only a part is found
commencing with the chapter on the depredations of a
Mangkasar prince (^^ -i^), and ending with the
conquest of Malaka by the Portuguese. No. 80 (4to of
* The Sundancse version has the title " Ilikayat Surtan Oliya He'nu Ibrahim
waliyullah," a copy of which is in my possession (small 4to, 90 pp-)
1 [L. 1694 and p. 178.]
MALA Y MANUSCRIPTS. 1 7
3 1 2 pages), and No. 5 of the Farquhar collection (small
4to, 259 pages) both end also with the conquest of Malaka
by the Portuguese. There are a great many copies of
this work* in Holland as well as in the Indian Archi-
pelago, in the Government offices, and in the possession
of individuals. 1
No. 19 (folio of 331 pages) and No. 20 (folio of 365
pages) contain the^U-^1 3j*J> Jb <uK^ . This is again
one of the Panji tales. The title is after a name by
which the prince of Kuripan was known when he was
changed by Batara Indara into a woman. The begin-
ning of the second volume is not connected with the last
words of the first, being <0cu* ^^
No. 43 (small folio of 142
pages ; on the back of the cover, Hikayat Pangeran Ke-
suina Agung) contains the same, but only its last part,t
the beginning words being fr U^
The name of the hero is in
this volume^U^J ^^ A)lj. No. 51 (small 4to of 149
pages ; on the back of the cover, Hikayat Dalang pudak
Asmara] is the same, but the usual commencement is
wanting, its first words being c ^
&C. c
No. 20. See under No. 19.
No. 2 i (small folio of 669 pages) contains the
* That it contains for the greater part but fabulous history is beyond all
doubt, as even the history of Malaka is tainted with the Panji tales ; see, for
instance, the chapter about the king of Malaka going to the court of Majapahit,
and marrying a princess of the name of Chandarakirana (compare under No. 3).
1 [L. 1703, 4, 1 6, 36 ; Bat. 188-90 ; de Hollander, 1. 1. p. 356, No. 2. The
work was lithographed by Keasberry at Singapore in 1830, and a reprint of
this appeared at Leiden in 1884. A French translation, by M. Devic, of part
of it (to page 123 of the Leiden edition) was published in Paris in 1878.]
"t This part is often found separately, as may be inferred from Bahru-ddln's
list (containing an account of Malay compositions found at Surabaya], whereia
we find a
SECOND SERIES. VOL. II.
iS THE ROYAL ASIATIC SOCIETY'S
,%- ) JoJ> . This is a collection of loosely connected tales,
the greater part of which relate to the persons involved in
the contest between the Kanrawas and Panda-was. To-
distinguish this composition from that which only relates
to the war, I propose to call the last <^U jboJ y <k\** ,
on account of its being so popular (see under No. 2), and
the first ^J^ gx3^|jx3 &Kfe * As to the contents of this
number, it is evident that it is an entirely different work,
and by no means to be identified either with No. 2, or the
two MSS. at the India Office (see under No. 2). The
commencement narrates the birth of Pardsu Rama and
Deiva Barata, sons of ^jj y*<~} by the celestial nymph
Manik* Then Santdnu is mentioned, and the birth of his
children, who had a peculiar fishy smell about them, as
they had been cut out of the belly of a fish, who had
swallowed the seed of Santdmt. On p. 2 Parasdra cures
the stinking princess Diirgandini, and calls her afterwards
Say oj ana Snganda (sweet-scented at the distance of a
yojana), taking her as his wife. She becomes the mother
of Biydsa. Another part of the narrative is about Basmaka,
king of Mandira-sapta, who had three daughters, called
Amba, Ambi, Ambalika. Amba becomes the wife of
Dewabrata, who kills her by inadvertence. He therefore
vows to surrender his life to a woman, burns his wife's
body, and then goes to his brother, Parasurama, who con-
soles him and changes his name into Bisma. On p. 34
\ve find mention made of the birth of Ddstardtaft Pandu
Dciva Nata>\ and Widura Saqma.^ Dastard ta was born
blind, because his mother, when visited by Biyasa, from
fear closed her eyes ; Pandits body was white as crystal
because his mother had covered herself with a white veil
when she conceived him ; Widura Saqma was born with one
lame leg, as his mother had pulled his leg (?). On p. 38
the birth of Kama is related : he was the son of Sangyang
* This title I derive from the first pages of No. 2, where the author calls the
part of his work not bearing directly on the war by this name (see under No. 2).
j [Probably Menaka.]
f- Dhritarastra. J In Malay the name of Pandit. I'ia'jira.
MALA Y MANUSCRIPTS. 1 9
Rawi* by Dewi Pdta.\ After this, the meeting is related
of Bisnu and the goddess Partiivi (see under No. 15),
and then the birth of Darmadeiva and Ddrmadcwi (see
under No. 15). On p. 91 we have the birth of Ktisna
(Krisnd) and Kakarsana (a surname of Baladewa]. The
last chapters relate the contest of Boma against Samba
(see under No. 15). Although this composition is but a
collection of narratives with no plot whatever to deserve
the name of Jukayat, it is very interesting, as it introduces
nearly all the persons acting in the hikay at parang Panddwa
Jay a, and the hikay at Maharaja Boma?
No. 22 (folio of 720 pages ; the commencement is want-
ing) contains the +\) ^j~ &^=>- It is a very elaborate
recension of the Malay Ramayana, from which Marsden
has given extracts in the Reader at the end of his
Grammar. A far shorter version has been published by
Mr. P. P. Roorda van Eysinga (Amsterdam, 1843). A
MS. in the Dutch India Office contains also a version of
it as elaborate as this ; it is in two small 4to volumes
(marked Ned. Kolonien. Handschriften C. No. i), the first
volume being of 475, and the second of 654 pages. 2
No. 2 3 (folio of 69 8 pages) and No. 4 5 (4to. of 2 7 8 pages).
Two copies of the c^>U A-^lj J^^ <^ql<s>. The first
number corresponds in version with a manuscript belonging
to the Dutch India Office (folio of 1 8 5 pages, and marked
Ned. Kolonien. Handschriften C. No. 2 1 ; it is not finished),
but is more elaborate. No. 45 seems to belong to the
same recension as the two copies of the India Office (No.
2,875 small folio, and No. 2,691 large 4to|). Another
version is contained in No. 27 (folio of 347 pages), and
No. 28 (folio of 348 pages: on the back of their cover,
Hikayat Dalang Indra Kcsuma). Both these volumes are
divided into chapters, each of which contains a tale,
* The god Sun. t Instead of Parta, Sansc. Pritha, i.e. Kunti.
1 [See above, No. 2.]
2 [This MS. is now the property of the Leiden University Library, and is
there marked No. 1689 ; Bat. 141 ; de Hollander, 1. 1. p. 336.]
See my " Kort Verslag der Mai. Handschriften van het East India House
te London," [p. 385, Nos. 50 and 51; L. 1699, 1709; Bat. 136, 137. On
the Buginese recension, see Matthes, 1. 1. No. 85, and p. 94.]
C 2
20 THE ROYAL ASIATIC SOCIETY'S
connected with the chief story ; the first volume contains
fifty-four tales and the beginning of the fifty-fifth, whilst
the second commences with the fifty-sixth tale. The
title, JsC^ &c., of this Panji tale is after a name which
the hero takes on his rambles in search of the princess
of Daha, disguising himself as a man of the lowest class.
This is one of the most interesting Malay compositions,
and has influenced almost every literary production of the
Malays ; on another occasion I shall give an analysis of
it. This cycle of stories has received by mistake also the
name of t>y~& <^^ &^=* from its commencement, where a
Batara Naya Kasuma, an inhabitant of IndrcCs heaven, is
spoken of as the grandfather of Kdrtapati.
No. 24 (two folio volumes of 446 and 450 pages)
contain the ^U> dallJ g)j dLJu- . The first volume
commences of course with Adam, whose son was Shlth
(ci_^wi>), whose son was Nurchaya t whose son was Sang-
yang Wcnaug, whose son was Sangyang Tunggal, whose
son was Guru, who had four sons and one daughter, being
Sangyang Sambu, Barahma, Mahddewa, Bisnu, and Dewi
Sari. Bisnu became king of Java with the title Prabu
Seta (? fcs^ft*-). Then a chapter treats of the Ratu Sela
Panvata of Citing Basi. It ends with Pangcran Dipati
taking the title of Susnnan ]\Iangku Rat Senapati, &c.
The second volume commences with Susunan Mangku
Rat being at Banu Mas, and ordering the Dipati of Tcgal
to be fetched, and terminates with Susunan Pakubuwamis
reign in Karlasura. Two quotations from this work have
been given by Dulaurier in the " Journal Asiatique " for
1846.
No. 25 (folio of 304 pages). This Panji tale goes by
the name of ~ c^JU cJol * cuKe* from a name the
heroine assumes when leading the life of a penitent. The
commencement treats of the prince of Kuripan, called
Kuda Jaya Asmara, surnamed Ktirtapati, who was
betrothed to the princess of Daha, Radcn Galuh Chandam-
kirana puspaning rat. A god falling in love with the
* From the Javanese endang (a female penitent or nun).
MALA Y MANUSCRIPTS. 2 1
said princess asked her of Batara Guru, but meeting with
a refusal, as she was to be the wife of Ktirtapati, dropped
her with her two waiting-women into a forest, where she
led the life of a penitent, and changed her name and that
of her companions. She is afterwards married to the
prince, here passim called Raden Inn, who succeeds his
father with the title of Pdrabu Anom ing Knrlpan, the old
king retiring to the woods to do penance.
No. 26 (folio of 239 pages). This Panji tale goes by
the name /*j~~ \) ?*> <&l^=>. The commencement is
almost the same as that of No. 23, relating the birth of
/;/;/ Kartapati, and that of the Raden Galuh Pnspaning
rat, surnamed Chanddrakirana. Going in quest of his
love, who is carried off by Batara Kala into a forest, the
hero takes the name Mesa* Tainan Panji Jay eng Kasuma.
Afterwards in the course of the narrative he is called Sira
Panji Wila Kasuma (p. 73 of MS.), but often merely Sira
Panji. After a great many adventures he becomes king
of all Java.f Werndly in his " Maleische Boekzaal "
mentions a Hikayat Mesa Tainan Wila Kasuma, and van
Hoevell, in his annotations on the Sair Bidasari, has given
quotations from a Hikayat Panji Wila Kasuma, pp. 301,
326, 334, 339, 362, 363, and 374.
Nos. 27 and 28. See under No. 23.
No. 29 (folio, 645 pages ; ends abruptly). This Panji
tale has the lettering Hikayat Naga Bersru (on the fly-
leaf within). I dare not decide whether this is right, as I
did not succeed in finding the reason for this title. Leyden,
in his "Dissertation on the Indo-Chinese Nations" (As.
Res. x.) speaks of a Hikayat Naga Bisaru& or story of a
princess of Daha, who was changed into a serpent, and
banished to a lake. It is a pity he gave no explanation
of the name. At all events this manuscript belongs to the
Panji tales. It opens with the king of Kuripan asking
* Jav. Matsa (Mahisa, buffalo) is frequent in proper names of Javanese per-
sonages, and is sometimes rendered by the equivalent Javanese kZbo.
t Amuter jagad jawa (Jav.).
This bisaru and bersru ot the lettering, I should like to explain by^ w*> O
in the sense of to cry invoking the gods.
22 THE ROYAL ASIATIC SOCIETY'S
for his son, the Raden Inn Kartapati, the hand of the
princess of DaJia, Chandarakirana, The hero is here
passim called Sira Panji and Sari Panji.
No. 30 (small folio of 74 pages ; the wrong lettering
on the back of the cover, SalasilaJi nabi Muhammad, is
owing to the first words, which make Nuru-ddln a descen-
dant of the prophet's) contains the $y &js*2*> y^>^. It is
a genealogical account of the kings of Cheribon (properly,
Chi-rcbon\ commencing with a confused tale about Sheykh
Nuru-dd{n y surnamed the Suhunan Gnnnng Jati, one of
the apostles of the Islam in Java. It is probably translated
from a peculiar dialect of the Javanese, its language being
anything but Malay, and mixed up with Javanese and
occasionally with Sundanese words too. Besides the said
Suhnnau ("Reverend"}, other celebrated apostles, as the
Suhunan's Kali Jaga t Ampel Danta, and Bonang are per-
sonated here as people endowed with miraculous gifts, and
the conquest of Majapahit, Bantcn (Bantam), and Pajajdraii
(called here too by its ancient name Pakuivaii) is briefly
narrated. The Panembahan Sura SoJian, called also
Moivldna Plasanu-ddln, introduces, according to this
chronicle, the Islam in Pajajaran, the Lanipong country,
Indarapura, Bangka-nlu (Bencoolen), and Balo. His elder
brother, the Panembahan Pakung Wati rules the country
from Krawang to Cheribon, he himself that from Bantam
to Krawang. This manuscript makes use occasionally of
the linguals .} and L.* From it some valuable materials
might be gleaned for a work on Javanese history, the last
pages containing an account of the kings of Cheribon
down to Sultan Anom.
No. 3 i (folio of 41 1 pages) contains the jUj> sli- &lo.
The hero is the son of Shah Partsad\ Inddra Ldqsana,
king of TMraf t \ situated in the neighbourhood of Mount
Qdf. This king, although powerful, was forced to pay
tribute to the monkey-king Baliya Inddra^ whose residence
* e.g. ^S^^y^ and^aJ (bhatani). f
According to the Malay history of Kama the same as />a/i, and brother of
Si/^r/wa. Malay compositions borrow from each other proper names ; so, for
MALA Y MANUSCRIPTS. 23
was Kurdari (<&^j^). The king's eldest son, called in
the commencement Qnbdd Lcla Inddra* and afterwards
Shah Qubad JoJian ' Arifin, could not put up with his
father's disgrace, and resolved to deliver his parent from
the allegiance to the monkey-king. He is in several ways
assisted by genii, who prove to be his relatives, and wages
war against the powerful enemy. This manuscript ends
abruptly, the last words being
Another copy, in which some of the proper names are
different, goes by the name of ^*JiJ! ^^ <&l=* (No. 6, large
folio of 414 pages). The father of the hero is called here
Shah Pftrmat Inddra Laqsana, and the residence of the
monkey-king Kdrdar 0^)- An entirely different version
is I. (85 pages, and ending abruptly) of No. 58 ; it goes
by the name of IjJuJL* I^Jol ^f^ s^ g|^ dJ.K^. The
hero is in this recension the son of Bakariiia\ CJiandara,
king of Baranta Inddra. A specimen of the readings of
No. 3 i and No. 6 deserves being inserted :
No. 6 :
instance, we find Inddra Kila (mountain, where Arjuna lived as penitent),
Alinlaraga (name of a cave, where Arjuna did penance, Sanscr. and Kavi
I'itaraga. passionless), and other proper names from the Kavi poem Vivaha
(in Malay, Z^ A-J <q^^ ^i^^ ^ rom a Titan conquered by Arjuna)
occurring in other compositions. Such proper names ought to be received
into the Dictionaries.
* Werndly in his " Maleische Boekzaal " mentions a tale about a person of this
very name, and Bahru-ddln (list of Malay works to be had at Surabaya) has a
AJ uol<J g)j <&
t See under No. 12.
24 THE ROYAL ASIATIC SOCIETY'S
***'
No. 32^ (? folio) contains :
I. (11 pages). An account of various ceremonials,
customs, and laws e.g., of the chief ministers a king should
have, the flags they wear, &c.
II. (5 pages). A short story about Indarapura being
* Dulaurier in his list speaks of two folio volumes, both containing \* c JJ 1 >
but I have only found one, on the back of which the number was obliterated.
That number is consequently all but certain.
MALA Y MANUSCRIPTS. 2 5
attacked by todak- fishes,* and the stratagem by which they
were defeated.
III. (5 pages). The first arrival of the Portuguese, and
their stratagem to get possession of Malaka.f A trans-
lation of it by Sir Stamford Raffles is to be found in the
"Asiatic Researches," xii. p. 115.
IV. Coloured figures representing the flags used by the
sovereign and his chief ministers (belonging to I.).
V. (63 pages). A tale the commencing words of which
look more like a chapter than like a separate tract.
They are :
&c.
The last words are :
It relates the adventures of Dewa Bisnu, son of the
king spoken of in the commencement ; from which it is
probable that the title should be ^~j ^ &l^ .
No. 3 3 (folio) contains :
I. (i i pages). A collection of laws, commencing with
the finding of goods, and what is to be done with them.
II. (44 pages). Laws, some of which are maritime.
III. (8 pages). Fragments of a law book, beginning
with the fencing of cultivated fields.
IV. (6 pages). =^ ^\^ ^/^V ^ <-^ About the
ominous qualities of the days of the months, having mystical
names, mostly those of animals. The same is found in II.
of No. 74.
* The same is told of Singapuni (see " Malayan Annals," p. 83) and of Barus
according to the Sair Raja T^tkt^mg ( ix2J gl) v*-^).
t The same narrative is found in one of the last chapters of No. I.
26 THE ROYAL ASIATIC SOCIETY'S
V. (3 pages). &*> <^~ c/^V ^ <-^ On the
seven ominous times. The same is XVI. and XXXVII. of
No. 34, and IV. of No. 74.
VI. (6 pages). ^J <^L^ ^^ ^ J*=^ On the
five ominous times. Compare the " Bataksch Woorden-
boek," p. 419. The same in No. 34 (X. and XXXV.)
and No. 74 (V.).
No. 34 (folio ; the number obliterated, and on the back
of the cover, nndang nndang) contains :
I. (i page). A fragment from a law book.
II. (i page). ^ ^J^JL. ^_Tlj ^f\s^ Ja c^b On
the serpent turning itself round in the sky, the position of
which is to be known, especially when going to war. 1
III. (3 pages). Charms and antidotes.
IV. (15 pages). Malay laws., commencing with the
fencing of cultivated fields. The maritime part has been
published by Dulaurier in the sixth volume of Pardessus's
" Collection de Lois Maritimes."
V. (7 pages). Treaty between the Admiral Speelman
and Hasanu-ddln, king of Gowa, and other Mangkasar
chiefs (compare No. 10).
VI. (i page). Chronicle of Mangkasar, commencing with
&c. (continued in VIII.).
VII. (i page). Contract of 'Aliyn-ddin of Gowa with
the Malay merchants.
VIII. (3 pages). Continuation of VI. (continued in XII.
and XVIII).
IX. (9 pages). A chapter on the law of inheritance.
X. (2 pages). See VI. of No. 33.
XI. (6 pages). Customs and laws commencing with
the duties of the Bandhara, Tumanggung, and other
functionaries of the Malays.
XII. (2 pages). Continuation of VI.
XIII. ( i page). A fragment about the discontinuance
of praying according to the words of the prophet.
1 [Compare " Bataksch Wooidcnbock," p. 327.]
MALA y MANUSCRIPTS. 2 ^
XIV. (i page, 54th page). Formulas used as charms.
XV. (i page). On ominous days ( ( j~s > ).
XVI. (3 pages). The same as V. of No. 33.
XVII. (i page). c^jJl JU; * ^l^- J.3 J^3
XVIII. (4 pages). Continuation of VI. (continued
in XIX.)
.XIX. (3 pages). A fragment of a work on supersti-
tions and continuation of VI. (continued in XXVI.) on
charms, commencing with the means of seducing a
woman, &c.
XX. (i i pages). Receipts against diseases, commenc-
ing with a precept about the regular course of a woman's
sperm (^^y> ^ W>)
XXL (p. 78). Table of ominous events, which have
to be expected on each day of the month.
XXII. (p. 79). Receipts, commencing with a prescrip-
tion against stomach-ache.
XXIII. (p. 81). The letters of the alphabet with
their mystical meaning under each of them.
XXIV. Regulations for the chief of the Malays settled
at Mangkasar, his power, &c.
XXV. Prescription to conquer a woman's obduracy.
XXVI. (p. 82). Fragment of a chronicle (VI.) and
continued in XXVIII.
XXVII. Continuation of XXIV.
XXVIII. Continuation of VI.
XXIX. Combination of letters attributed to prophets,
angels, and holy men.
XXX. (p. 98). A precept of the wise Loqmdn about
the future of a just-born child.
XXXI. (p. 99). On the ominous signification of
earthquakes, lightning and eclipses, according to the time
of their appearance. A fragment of a similar work is to
be found in de Hollander's Reader, p. ^|A.
XXXII. (p. 103). About the choice of the ground to
erect a house upon, to make a field of, &c.
* See Herklots' " Customs, &c." p. 395.
28 THE ROYAL ASIATIC SOCIETY'S
XXXIII. (p. 1 06). Means to know how a man and
woman live together.
XXXIV. (p. 1 1 o). Means to know whether stolen
goods may be recovered.
XXXV. (4 pages). See VI. of No. 33.
XXXVI. Astrological tables of the planets according
to the days of the week.
XXXVII. See V. of No. 33. On p. I2O ; an illustra-
tive table.
XXXVIII. (p. 121). A figure illustrative of the ser-
pent's position (see II.).
No. 35. See under No. 18.
No. 36. See under No. 7.
No. 37. See under No. 9.
No. 38 (small folio of 87 pages), No. 59 (small /j.to of
138 pages), and No. 71 (small 4to of 196 pages) con-
tain the <u^J ^ dLK &Ko * In the last-named
number the introduction is wanting. Some fables from
this book have been published by J. J. de Hollander in his
" Malay Reader," p. I 8 sqq. I possess a manuscript of it
(4to. of 205 pages). A specimen of the various readings
of these four manuscripts may not be out of place. The
reader may compare with it the fable published on p. I 8
of the above-cited work.
No. 38 :
Jy ~& t^j| cjo^ <
* On the west coast of Sumatra it goes by the name of
si-tandmli} after the name of the bull who became the lion's friend
MALAY MANUSCRIPTS. 29
No. 5 9 :
jJ^^" t_3l cJo^
oJ*^ tMJy ^ <^
1 ^1 ^ ^Usi 1 J1
cMJU,
No. 7 1 :
My Manuscript :
Hence it appears that the manuscript from which de
Hollander published some fables must belong to another
recension than these four manuscripts. All these versions
are from the Persian.*
* The Tamil version has been translated by Alxhi-llah the Moonshee and
published at Malaka [in 1835, 85 lithographed folio payes]. It is divided in
the same way as the Panchatantra, and bears the title of ^A^J^J ^/-J . f A
30 THE ROYAL ASIATIC SOCIETY'S
No. 39. See under No. 18.
No. 40 (folio of 320 pages) contains the
j*^w^. The hero is the son of a king of ^ in West Java.
This king had two wives, the younger being Amas Ajcng,
who bore him a son called <^_~J ^U^. She slanders the
elder, making the king believe she had tried to poison
him. The elder queen is defended by her son, who in
consequence falls into disgrace, and is incarcerated. The
queen herself is conducted into a forest to be killed, but
the executioner, pitying her condition, leaves her in a
grotto, where ^bj ^Lq supplies her wants. She is there
delivered of a son, who receives the name of lr>! ,j~-y ^^
jz~*j&u*. The story ends in a strange and abrupt way,
as if not finished. I do not think it probable that this
composition is the same as that mentioned by BaJiru-
ddm under the title +y~& ^ ^ f~* &l5se, which is deci-
dedly a Panji tale.
No. 41 contains a Malay translation of a Javanese
Wnkon* It is a miserable composition, not readable
without the Javanese original.
No. 42 f and No. 64. Two copies of the c5ft LJLJl gti.
This work has been published with a Dutch translation by
P. P. Roorda van Eysinga (" Der Kroon der Koningen."
Batavia. 1827). A great many quotations in Werndly's
" Maleische Spraakkunst " are from a better manuscript
than that used by Roorda van Eysinga.
No. 43. See under No. 19.
No. 44 (4to of 303 pages) contains the J&tf ^U &l&,
a Panji tale ; the title is derived from a banner (tunggnl),
the baneful influence of which occasioned a great mortality
in the land,J being pulled up (c/iabut) by the hero. It
opens with the god Naya Kasuma (see under No. 23)
r.ew edition, by the writer of this account, appeared at Leiden in 1866; it
has a valuable introduction, and philological and critical notes. A Dutch
translation, by Klinkert, came out at Zalt-Bommel in 1871. See also de
Hollander, 1. 1. p. 368; L. 1729, 1757; Bat. 184, 5.]
Raffles' "History of Java," i. p. 745 sqq. "Tijdschrift voor Indisclic
T?al- Land- en Volkenkunde" (Batavia, vol. vi. and vii.).
4- See also Nos. 17 and 47 II.
J Compare Cohen Stuart, 1. 1. p. 153.
MALA Y MANUSCRIPTS. 3 1
descending into the world, and taking the name of Mesa
Pdrta Jaya Kdlana Baujdran. He becomes king of
J\IajapaJiit, with the title Pdrabn Wira Kdrta, after having
married the only daughter of the old king, who retired to
do penance. His sons become kings of Knripan, Daha t
Gagdlang, and Singasdri. In the course of the narrative
Kdrtapati and CJianddrakirdna are again the most conspi-
cuous characters. In his perambulations the said prince
calls himself Ki-rmnang Panji WanJian* (?), and the
princess of Daha, when leading the life of a penitent,
assumes the name of ^jy>^ <^U*J jj| (compare under
No. 25). The language of this tale is crowded with
Javanese words and expressions. As humble pronoun of
the first person, pun titiyang\ (the man) is here used as in
the Balinese.
No. 45. See under No. 23.
No. 46 (large 4to of 306 pages) contains the &l=^
^Jjw* jjJ. The hero's father is Karma ludara, king of
Kangsa Inddra. This king has heard of a certain white
elephant, and orders Parba Indara to catch it. Parba
Inddra, failing in executing the orders of his master, is
discarded from the court, and leaves with his family. He
arrives at a hamlet, where a Sheykh Jddld was living in
religious solitude, and settles there. He afterwards begot
there a daughter called Siti\ Mangdrna Lcla Chahya,
with whom the new king, Pakdrma^ Raja, falls in love
when coming accidentally to her father's hermitage. Siti
Mangdrna is after due time delivered of a son, who is the
hero of this tale. This prince leaves the residence, and
rambles about to increase his knowledge of the world. In
the course of his rambles he meets with the white elephant,
who was a princess of the name of Lela Rdtna Kumdla, and
had been transformed by a demon, of the name of Dewa
* The manuscript : -.Y^} A Hikayat Mesa Kiramang is mentioned
in the "Journal Asiatique," 1833, by Jaquet.
f Compare the use of ulun as pronoun of first person, being the same as iihm
(Lampong) and tihma (Malagasy spelling olomi) \vhich signify man, and
ngivang (pronoun first person) and wtvang (man) in Kavi.
The Arabic g % See under No. 12.
32 THE ROYAL ASIATIC SOCIETY'S
Rdqsa Malik, out of spite, as he wanted her for his wife,
but met with a refusal at her father's hands. Another
copy of this tale is in the library of the India Office
(No. 2,871, folio volume), 1 where the introduction is entirely
different. According to the last words of that manuscript
this tale goes also by the name of I^Jol ^+* g\j Sjl=*
el? rA
No. 47 (4to) contains :
I. The^w^^jJ [;Jol <j^~ &K=*. The hero is a son
of a king of Kuripan by Sakarba* a daughter of Indara, a
king of Rating, who had made himself universal sovereign
of the world (c*_>l^ ^^\ \j&^) ; having subjected the kings
of Gujarat, Mogol, Abyssinia, MacJiulipatam, Bengal, &c.,
he sends a fleet to conquer Java, going himself thither
with his sons through the air. Mesa Inddra Dciva
Kdsuma opposes the conqueror. Amongst the places the
Indian king besieged is Pajajdran, the king of which had
o
a son called Amas Tanduran, and two daughters, called
Raden galuh Kuim'tda Rdsmi, and Raden galuh Dcwi
Rina (?). The opening of this tale is anything but clear.
It is besides crowded with Javanese expressions, as for
instance, mdngambaJi jumantdra\ (to tread the air).
II. (64 pages). Fragments of the c ^ r LXJI glS (see
No. 42).
III. (18 pages). o Jji J&u^ &&\ ^^j&tyu*. Another
copy in No 62. This short tale about Muhammad's
miracle of making the moon pass by halves through his
sleeves, has been published by Robinson at the end of
his " Principles to Elucidate the Malay Orthography." t
There are a great many manuscripts of this legend.^
IV. (5 pages). ..XJ ^y <&K^. On the duties of a
married woman, about which the heroine of this tale
J [Now 210; B.M. i; Bat. 157, 158.]
* Corruption of the Sanskrit Siiprabka. [See " Bataksch Leesboek," vol.
iv., p. 115.]
T The Sanskrit dyumantara.
J P. 222 sqq. of the Dutch translation by E. Netscher [p. 181 of the original
work].
( >ne in the possession of Mr. II. C. Millies at Utrecht, and another in mine
(small 8vo, of 28 pages).
MALA Y MANUSCRIPTS. 33
consults the prophet* A copy is in the possession of
Mr. H. C. Millies, where the proper name is spelt UJ^3.
No. 48 (small 4to of 210 pages) contains the 3 QJZ*
j^=>^JI ay&je* j*z&& ^J\ <-_aJaJ.f It is translated from
the Arabic of Ibrahim Laqanl, by the Sheykh Shihabu-ddin,
surnamed the Pilgrim, and son of 'Abdu-llah Muhammad,
sumamed the Malay (^jWJl).
No. 49 (4to of 56 pages).^ A poem, the title of which
is uncertain. It contains the celebration of a king of
Bintan, and the splendour of his palace, garden, &c. The
first verses are :
The last verses are :
<^.. 0^
* It goes also by the name of &*Ju,v c^ 1 "^ ^i^^ according to de
Hollander, 1. 1. p. 331, No. 14. [On the Makassar recension, see
1. 1. No. 33, III.]
t Another commentary on the same work is called, - yZ Jo i^Jl
^ The lettering on the back of the cover (Karangan Banian] is wrong.
Dulaurier infers from it that it is about the foundation of Bantam.
The Arabic ^-s^* .
SECOND SERIES. VOL. II. D
34 THE ROYAL ASIATIC SOCIETY'S
No. 50 (small 4to of 96 pages) contains the
^^IjJl i.e., the adventures of Tamlum-dddri, an inhabitant
of Madlnah, and originally a Christian. It is taken from
the t^l^aaJI pfc. He was carried off when bathing
during the night, which the prophet had prohibited, by a
spirit (jin) to the country of the genii, that were yet infidels,
and stayed there seven years and four months. He meets
in the course of his rambles with the Antichrist (Jlij),
appearing in the form of a large bitch big with barking
puppies, and becoming large when hearing bad reports
about the Muslims, and small when they are favourable ;
with female cannibals on a certain island, with the angels
Jabardil and Mikail, and the prophet Hilir (^*oc*), who
gives an explanation of the wonderful things Tamlm sees
and cannot account for. He meets a bird, too, which
gives him a delicious beverage out of its bill, and is no
other than the bird of Ishdk, and leads the erring faithful
upon the right way. He sees a man filling out of a pond
a tub with the bottom off, being a usurer. On his return
to this sublunary orb, he finds his wife re-married, and
squabbles with her husband. ' Umar ()*-) could not
settle the quarrel, as Tdmlm, not having shaved and
pared his nails during his absence, looked quite another
man, and was not recognized. 'All (<J^-) then recollects
a communication from the prophet about a sign by which
Tdmim could be identified, being a whitish spot as large
as a ddrham behind the knee. 1
No. 5 i . See under No. 1 9.
No. 52 (4to of 140 pages) contains the e_j>b g^&Ke*.
In the opening a king of Gunung bdrapi rantuw panjang
tdbing bdrukir is introduced, called ^y> &U* jt?&^=>. He
had forty wives, one of whom only, called Inddra Sort,
became pregnant. Sending away the other thirty-nine
on account of their sterility, he was cursed by them to have
1 [L. 1719 and p. 178. A lithographed text edition appeared at Singapore,
a. II. 1297. On the Makassar recension, see Mntthes, 1. 1. No. 10. Con-
cerning 7'amini ibn AAsu-dd&ri, see Slane, " Ibn Khallikan's Bibliographical
Dictionary," vol. ii. p. 21. ]
MALA Y MANUSCRIPTS. 3 5
a hog as a son. After a pregnancy of seven years the queen
was delivered of a boar of a terrifying appearance, with tusks
as yellow as a ripe plantain fruit* The king ordered his
minister to throw his son into the woods, where the young
hog conquered the king of the hogs, being assisted by a
princess who was doing penance on the field of their
contest. Having been victorious, he was bathed by that
princess, and treated in her residence as her son. After
taking leave of her, he is carried away by a ji/t, &c.
This composition is replete with pantuns, and the text is
not much corrupted. The language is genuine Malay as
far as I have read it. After a great many adventures,
the hero returns in a human shape to his father's residence,
and is then called Inddra Btirma Kala. 1
No. 5 3 (small 4to) contains :
I. (98 pages)f the ^<~v*~4 d)K=*. It has been.
published at Singapore (lithographed). 2 There are a great
many manuscripts of this tale. The one in my possession
has been vitiated by a Batavian transcriber, who has, for
instance, changed ^iU~ into *.^^^ !
II. (26 pages) ^J j*&, ? It commences with exhor-
tations to children, and is a miserable jingling of rhymes
about a great many fishes, introduced in it as would-be
poets. The composition may be serviceable in correcting
the existing Malay Dictionaries in the wrong pronunciation.
of fish-names. I possess a manuscript of it.
No. 54 (small 4to of 293 pages) contains the &K=*
^Uvs} A*=J . In the commencement there is a kind of
summary, relating that the hero was harmed by a genius
called ,q*&~ )^=> , that on his rambles he came upon
Mount Langkdri Ratna, where he saw two princes of the
genii, that he encountered the princesses Sakanda Kumdla
1 [B.M. 1 8.]
+ The lettering Angkasa Dewa is a mistake owing to the tale commencing
with these words. % Si-miskin ("the poor one.")
- [In 1857 and 1879; Bat - J 75 17&]
3 [Printed at Batavia in 1865.]
n 2
36 THE ROYAL ASIATIC SOCIETY'S
Indara and Bumdya Inddra, fought the king Makiitcv
Inddra on account of the first-named princess, and was
thrown by order of that prince into the lake Inddra Sdtu-
nang, where he was swallowed up by a serpent, in whose
belly he met the princess Bdranta Maya ; and a great
many other adventures of the same kind are told. The
hero was the son of Sahfar Tsaf Inddra* king of Bu-
rangga Dciua. In one of the chapters ,3^^ ^ is said
to be king of a State situated in the cavern of Mount
Dew a Rangga Inddra.
No. 55. See under No. 9.
No. 56 (4to of 412 pages ; on the cover Badiiilzaman
Anak Hamzah). In the commencement are contained the
adventures of ^^ 1 --^> sa ^ * ^ e ^ e son f Hamzah;\
then follow those of his father, and of 'Uniar Maya, with
whose death it closes. Perhaps it is but a part of the
No. 57 (4to of 332 pages) contains the l^Jol
Manginddra CJmwdcJia, king of Inddra Pdr-
changga, had two sons called Raja Shdhjohan Mangindara
Rnpa and Raja TJiahir\ JoJian Shall. The king having
dreamt of a wonderful musical instrument which sounded
one hundred and ninety times when but once struck, and
longing to have it, the two young princes go in quest of
it. They are adopted by a ghost, of the name of ^Ju* gl;, '
who tells them where to find the wonderful instrument.
He changes their names, calling the eldest prince Inddra
Mahddcwa Sdqti, and the youngest Bisnu Dcwa Kainddra-
an, surnamed Inddra Ldqsana. The brothers are separated
afterwards, each of them achieving a great many stirring
feats by the assistance of the jin, their adopted father.
Extracts from this composition are to be found in
t ()!' the &j4* <&=> de Hollander in his "Reader" (p. 82 sqq.) has
published extracts; and another extract is to be found in Roorda van
Jiy-si lira's " Beknopte Maleische Spraakkunst" (Breda, 1839), p. 102 sqq.
1 [There is a MS. at Leiden, 1. 1. p. 178.]
MALA Y MANUSCRIPTS. 3 7
Marsden's "Malay Reader" at the end of his Grammar,
according to a manuscript but slightly differing from this
one, of which the lettering on the back of its cover, Indra
Jayangan, is a mistake for Inddm Kiydngan, as the extracts
published by Marsden have it.
No. 5 8 (small 4to) contains :
I. See under No. 31.
II. (34 pages). ^5oU >a^ ^>^ &L&^ , which is a
more elaborate version of this tale than that published by
de Hollander (Breda, 1845), and corresponds more with
that of the manuscript from which quotations are found
in Roorda van Eysinga's " Maleisch-Nederduitsch Woor-
denboek" (s. vv. /laram, haji, cldyuui, churi, khiyanat,
kJicjiiali, darah, dapat, dakap, dandam, diri, ridla, rambut y
zadaJi, saldsey, sdlam, sandscJiaya, swat, sayid, sisi, shetan,
tsahib, pandey, and saJidya). In this manuscript the
heroine's brother is called JJbU, <__^ . There was, and
perhaps still is, also a MS. copy of this tale in the posses-
sion of Mr. Frederick Muller, at Amsterdam. On the
west coast of Sumatra the heroine goes by the name of
Johor Malbgan, which name occurs in Bahruuddln's list too.
Of the Sumatra version * I possess an incomplete copy.
No. 59. See under No. 38.
No. 60 (small 4to of 106 pages) contains the &Ks*
^y* sU*.f Another copy is No. 66 (small 4to of 223
pages). The hero, assuming in the course of his rambles
the name of Indara Jaya, this very popular tale goes also
by the name of <^U l^Jol djKsu . Part of it has been
published by de Hollander in the first edition [and
in the fifth from p. \pv to p. \6^] of his " Handleiding
bij de beoefening der Maleische Taal- en Letterkunde,'*
* About a Javanese version, see Raffles' " History of Java," i. p. 394 sqq.
[L. 1691, 1727, 1777 ; Bat. 121-3. ^ ee d e Hollander, 1. 1. p. 351. An
edition by the same appeared at Breda in 1845.]
f The Persian pronunciation, sM-i mardan has occasioned the name
^sl J w* :^Ar cited by van Hoevell in the annotations to the "Sair Bidasari."
^^ \*> i
On the west of Sumatra <%JV- &vw is pronounced sa-i alani ; hence confu-
sion in the title of the dwarf deer between shd-i 'a/am di riinba and shaykh
*(] lain di riinba,.
3 8 THE ROYAL ASIATIC SOCIETY'S
p. nv-lTl- It is also called after the hero's father,
& S^ f& * ^^^ * fc ^ s men tioned by Leyden
(" Asiatic Researches," x.) under the title Hikayet Biker-
madi(tya). A translated extract about the creation of
the world is to be found in No. 60 of the " Indo-Chinese
Gleaner." A copy, too, is found in a volume, containing
the Hikayat Palanduk Jdnaka, belonging to the library of
the India Office (No. 2673 ?). I possess a manuscript
copy of it of 38 pages folio. 1
No. 6 1 (small /j.to. of 150 pages) contains the <*
^Jl ^j~** (sJiamsUr-lbarri). The hero, called "the sun
of the earth," f is the son of Ddrma Dikdra, king of
Paruwa Chnkdra Nagara in Hindustan. His name he
owes to his being predestined to be a powerful king, ruling
over the earth and sea, whence he was surnamed " the
moon of the sea." J When twelve years old he was
carried off by an infidel jin, in consequence of which he
had a great many adventures before he returned home.
On his return he succeeds his father with the title of
Sultan Qamru-lbahrin.
No. 62 (quarto) contains :
I. See under No. 16.
II. See under No. 49, III.
III. (About 60 pages) JjL~ <-T~}.J~> - It contains one
thousand questions put to Muhammad by a learned Jew
of the Khaybar tribe. Having been answered by the
prophet satisfactorily, a great many Jews of the said tribe
embrace the Islam. It is translated from the Persian.
A manuscript of this composition (small 4to of 156
pages) I saw at Barus in the possession of the Tuwanku
of Sigambo-gambo. A copy is also in the possession of
Prof. H. C. Millies at Utrecht. It is a very interesting
work, and reproduces the popular belief of the Malays
* Wikramaditya. No. 60 has ,5 s
1 [I-O. 373; B.M. 14; L. 1733; Bat. 149, 150, 151. There is a Makassar
recension, 1. 1. No. n. See de Hollander, 1. 1. p. 347 f.]
t The manuscript explains the Arabic name by artina matahari di darat.
v*3* explained by bulan yang dit&pi Idut.
MALAY MANUSCRIPTS. 39
about a great many questions of the Muhammadan faith.
The orthodox priests condemn it as well as the
IV. (8 pages) ^ ^j J 5&.L* ^y <OJ)
Moses' ascent on Mount Sinai.
V. (5 pages) )\ <ul?U &l=* is a tract about the
duties of a married woman, expounded by the prophet to
his daughter. A copy of it is in the possession of Prof.
H. C. Millies at Utrecht. 2
VI. (6 pages) jZj^j} <*M1 Jj~^ 3t^=^- The prophet is
shaved by Gabriel, and his hair gathered by the celestial
nymphs for the purpose of making amulets of them.
Published at Batavia (1853, m I2mo, Lange and Co.). 8
No. 63 (4to of 349 pages ; on the cover, Raja 'adil\
It contains the ^^i=s? &lc*.* The wrong lettering is
owing to the commencement, where a just king (Raja
'ddil) whose name is not mentioned, is forced to flee from
his dominions, and is afterwards made captive by an
unjust king. This MS. however, contains another version
than that from which de Hollander has given extracts in
his "Reader" (p. 131 sqq.) and transliterated in his "Hand-
leiding tot de Kennis der Maleische Taal " (Breda, 1845).
The name of the person who found the child of which
the queen was delivered during her flight with her con-
sort, and which she was forced to leave, is here ^Jw^.
The last tale in this manuscript is the story of Solomon
and the Queen of Saba.f The text is pretty good,
but occasionally corrupted. The introduction especially
differs widely from that of other versions I know. I
possess a copy (folio of 50 pages), wherein the number
of tales told by Bakhtiydr amounts to nine. Its version
1 [Bat. 82-86 ; Leiden, 1. 1. p. 178. Lithographed at Singapore in 1879.]
[L. 1744 ; Bat. 94.]
3 [L. 1720; Makassar version, Matthes, 1. 1. No. 32.]
* The Persian original was translated by Lescalier (" Bakhiiyar, ou le
Favori de la Fortune." Paris, 1805).
t Of this story there is an elaborate novel on the west coast of Sumatra,
where it is called ,j~jUj i^jZi &\&* . I possess three manuscripts of it,
all written in the Menangkabovv dialect. [An edition of it, with translation
and notes, by Mr. D. Gerth van Wijk, appeared in the xli. vol. of the
" Verhandelingen van het Batav. Genootschap.'']
40 THE ROYAL ASIATIC SOCIETY'S
differs from that of de Hollander's text. This tale goes
also by the name of -1^)1 ajKs^ , t^s- &A\j <sjl5vs* (from
the name of the hero's father), and according to de
Hollander, also ^A^ 3i\s, which last name, found in the
Leiden MSS., is by far not so popular, and does not
convey an idea about its contents. 1
No. 64. See under No. 42.
No. 65 (small 4to of 152 pages) contains the
^jjyM j*&, a poem the plot of which is taken from the
Javanese. It belongs to the Panji tales* relating the
adventures of Panji and Angarcni, daughter of \hcpati7i,
with whom he fell in love after having been betrothed
to Sckar Taji, the princess of Kadiri. His father ordered
Angdrcni to be killed when Panji was absent, having
gone in quest of game.t This composition proves to be
the story which has suggested the plot of the ^ ^*
g&>j*4$ (see under No. 7). This manuscript breaks off
abruptly, and is to such an extent replete with Javanese
words, that a Malay would not understand it.
No. 66. See under No. 60.
1 [L. 1718, 1723, 1750; Bat. 132-4, 179. Printed at Batavia in 1880.
Makassar translation, ap. Matthes, 1. 1., No. 89.]
* Other tales belonging to this cyclus, and not existing in this collection,
are i. t>)~*& PrriW> f~~* ^i^^ (jayeng kdmrna is Javenese, and means
"victorious on the battle field," and is often changed in Malay into Jaya
Kasiuna), from which quotations are to be found in van Hoevell's annotations
(p. 301, 326, 334, 362, 363, and 374). n. ^ \y$ }+, >~*~- ~3 &Kr>
(after an assumed name of Chandarakirana, when dressing as a male, and
roving about to subject the States she came upon). in.
(No. 2,602, India Office). 2 iv. \j jj ^ ajr> (?)
v.
~-jJ> j+.~* dj^(.see under No. 40). vn. (see II. under No.
7). vili.^l^jij 45^ ^i^^ (a translation of which into Mangkasar is to
be found in Mattlies's " Makassaarsche Chresthomathie"). There are more tales
belonging to this cyclus, as may be inferred from some manuscripts in the
British Museum.
2 [Now No. 69. See the account of it, 1. 1. p. 389.]
t The plot does not differ materially from that of the tale of which Mr.
Taco Roorda has given an elaborate analvsis (see Lotgcvallen van Raden
Pandji in the " Bijdragen tot de Taal- Land- en Volkenkunde van N.I."
Vol. vii. Nieuwe Volgreeks). Compare also Raffles' " History of Java," ii. 88.
MALA Y MANUSCRIPTS. . .\ i
No. 67. ^U ^b 41^- Published by Dulaurier in
his " Chroniques Malayes." A list of countries dependent
on Majapahit, found in this manuscript, is published by
the same in the "Journal Asiatique'' for 1846.
No. 68. See under No. 18.
No. 69 (small 4to of 128 pages). According to the
lettering on the back of the cover, the title would be
ty~J ;jl J~ J^ J&J1 c_ib. It is an ethic work,
laying down rules for ministers and great functionaries as
to their conduct when officiating. It is illustrated by
tales. In the commencement the manuscript says that
the tale came from Sultan Aliyu-ddm Shah, son of
Mansur S/ta/i, king of Pdtani. On page 10 there is a
story about the sagacity of the dwarf-deer (palanduk*)
settling a contest between an alligator and a young man
about the propriety of the alligator eating the young man,
who had delivered it when about to die on the dry land.
The last tale is about a Sultan Al-alam Shah.
No. 70 (small 4to) contains :
I. (186 pages). The seventh book of the ^lx~J
^yJ^lsJI , an ethic work illustrated by a great many tales.
According to the last words of this manuscript, the
whole work is divided into seven books. On p. 163 the
narrative about Siti 'Abasah (see No. 76) is found.
II. (8 pages). A small collection of tales,t belonging
most probably to a larger composition (to I. ?). The
first tale is about Moses and Qdrun, who bribed a preg-
nant woman to say that Moses had committed adultery
with her. The second is about a certain $k trying to
* The ^^s *Ju3 cut* , two copies of which are in the Library
of the India Office (Nos. 3,049 and 2,603 [now Nos. 67, 96, and 373. A
text edition was brought out by N. C. Klinkert at Leiden in 1885]), has a great
similarity with the European tale about Reynard the fox. (See " Kort Verslag
der Maleische Handschriften van het E. I. House te Loriden.") The/dUtaufoe
acts in the Indian Archipelago the part that the fox acts with us.
t Such small collections of tales, the title of which is either arbitrary, or not
to be fixed from the contents, are often found in the possession of the poor,
who cannot afford to buy manuscripts of the extent of the Bustdmt-ssaldthtn,
Taju-ssaldthin, and the like. A collection of the same kind is the
V^nj **$ in Xo. 2,603 (Library of the India Office).
42 THE ROYAL ASIATIC SOCIETY'S
outstrip Moses by the force of his penance. The third
is about Moses' death. The fourth is about a woman
called dJ^JjJI fcuJj getting ten-fold back what she had
given to the poor. The fifth is about a man of the
name of ^~~ ^1^ seeing the mercy of God to the just.
No. 71. See under No. 38.
No. 72 (small 4to of 44 pages ; on the back of the
cover, Kitab rasnl). It contains the < Ji- gbW**
another version of the " Story of King Skull "* g)j
the plot being the same as that of the
_jlv*J f There are a great many copies of this tale.
Prof. H. C. Millies at Utrecht, possesses a copy (small
8vo) bearing the title sUv^b J^- ^Ks> (the Persian/ is
here strange). Of the story of King Skull there is a
copy in the India Office, being the third tale in the
J^ij ^.J djKs> (No. 2,6o3). 1 I myself possess two copies
of it (one evidently mutilated by a Batavian transcriber).
No. 73. See No. 4.
No. 74 (small 4to) contains ;
I. Laws of Malaka, Johor, and Salangor.
II. (n pages). See IV. of No. 33.
III. (i page). Receipts, commencing with that
against a kind of leprosy.
IV. (10 pages). See V. of No. 33.
V. (8 pages). See VI. of No. 33.
VI. Fragments of a religious work. On the last
pages are found coloured tables representing the five
ominous times (belonging to V.)
No. 75 (small 4to ; on the cover, iindang undang) con-
tains :
I. (6 pages). A fragment from a law book. The
first chapter is about people having plantations and
neglecting to fence them.
II. A fragment from some work on Muhammadan
*' Translated, " Asiatic Journal," 1823.
t Edited by Fraissinet under the title of " Geschiedenis van Vorst Bispoe
Kadja" (Jircda, 1849).
1 [Now 96; L. 1781 ; Bat. 106-109.]
MALAY MANUSCRIPTS. 45
law, commencing with the rules about selling and buying
( -j), and ending with the law of inheritance (^/\y>).
III. (2 pages). A fragment from an Arabic work on
law with Malay interlinear translation.
IV. (15 pages). An Arabic-Malay Dictionary.
Under each Arabic word the corresponding Malay is
written. The last seven pages are not filled up with the
Malay. I possess a complete copy, and a fragment of
another work of the same kind.
No. 76 (small 4to) contains :
I. See under No. 18.
II. (9 pages). <u*L.. ~z~* &K^. It is properly but a
tale taken from the ^^UJI c)^^ (* No - 7) but often
found separately. Two copies are in my possession
(small 4to of 20 pages, and small 8vo of 24 pages).
III. (23 pages). A fragment from a work on religious
observances, commencing with the sacrifices (^1^3).
IV. (small 8vo of 13 pages). Fragments of a work
containing Malay laws, and commencing with goods found
on the road.
No. 77 (small 4to) contains :
I. (4 pages). &M ^^ <^Ju* j*J~.
II. (6 1 pages). Maritime laws.
III. (20 pages). Orders issued by Sultan Ahmad
Tdjuddln Hallm Shah of Kddah (sJo), some of which
refer to the suppression of piracy (Muh. year 1133).
No. 78 (small 4to) contains:
I. (62 pages) <^Uj vi ( j^^s>\ y ^*^. A poem, cele-
brating the conquest of Java by the English forces under
Lord Minto. The first words are :
It closes with the description of a market, and teems with
Batavian Malay words.
* The Dutch Maarschalk (Marshal).
t Meester Cornells is the name ot a district of Batavia, where there are
barracks.
44 THE ROYAL ASIATIC SOCIETY'S
II. (2 pages) &]ji ^AJ^ j*&. A short love-letter* in
verses, of which the following lines may serve as a
specimen :
j
No. 79 (small 4to) contains :
I. (46 pages) the laudatory terms Malay letters com-
mence with, varying according to the rank of the person
addressed. The specimens given here are nearly all in
Arabic (continued in III.)
II. (7 pages). Fragments of a Muhammadan law work
containing the fines to be paid for wounds inflicted. The
\Ui^ is here explained by JJ> ^^o <tfj3 <j~^ * f^\
\lj ^j~> <__ -J ^>lj. Each of the Arabic law terms is
explained by a Malay phrase written under it, and con-
taining the amount of the fine ; an example will suffice :
(wound touching the pericranium) is explained by:
* Jo
III. (49 pages) continuation of I. It closes with the
model of a letter to the Dutch Governor-General and the
Dutch India counsellors (Raden van Indie). I. and III.
are consequently fragments from the kit ah tarasitl, a book
in which precepts are laid down how to write letters.!
IV. (7 pages) c^laJ! *\SL~, d)L; (a figurative title, "the
physic of hearts "). The author calls himself Nuni-ddui
Ibn 'All Ibn Hasanjl Ibn Muhammad Hauudi\ This is
* In No. 2,609 (India Office) there is a love-letter, the title of which is yet
to be ascertained. Its commencing verses are :
jo j|
ij
It contains 18 pages.
t See Newbold, ii. 338, and Matthes, " Makassaarsch Woordcnboek,"
j. v. tnrasolo.]
. (See about this author, Note I in the Appendix. )
MALA Y MANUSCRIPTS. 45
a treatise about the sense to be attached to the word
&jli. He composed it, he says, in order to combat
those that entertain wrong opinions about the nature of
God.
V. (8 pages). A tract, the title of which I could not
ascertain without reading it through. It begins with
stating the best time for building a house, and contracting
a marriage, and closes with a recommendation of forbear-
ance towards a slave, even when guilty. It is addressed
to 'AH (^^^\ eac h article ending with ^- b,
VI. (2 pages). Questions and answers about the sense
of ~wJLo (testimony), perhaps belonging to IV.
VII. (9 pages). <0y jA &LO ^f^ ^ &\~>} A tract
on the qualities of God.
VIII. (34 pages), jl&^l SJ^a^ .* This treatise is
divided into two introductory chapters (px&), four books
(c_>lj), and one concluding chapter
The first introductory chapter :
The second introductory chapter :
The first book :
The second book :
The third book :
The fourth book
IX. (15 pages). Arabic fragment from a commentary
on the Qur'an, with Malay translation.
No. 80. See under No. 18.
B. FARQUHAR COLLECTION. f
No. i (small 4to of 5 I pages ; within, Chcrita Siiltan
Iskander}. It contains a pretty good copy of the
* It is translated by jljl^S 1 -j&v* ^^^ ^ note ^ ^ ie transcriber calls
the author ^l jj 1 ^y -i^Cj (the same as the author of IV. ?)
f The manuscripts of this collection were not numbered. I have put numbers
on them in nccordance with the list Dulaurier gave of them, with the exception
of two volumes he did not examine.
4 6 THE ROYAL ASIATIC SOCIETY'S
]j commencing with what is reserved for the
Sovereign. The seventeenth chapter is about people
going to hunt. 1
No. 2 (small 4to of 202 pages ; imperfect at the end).
It contains the ^y^l ) ^AxCJ g\j &K& . The last
pages give the history of the defeat by Alexander of a
king who was a worshipper of the sun. A small extract
from this tale is to be found in Roorda van Eysinga's
" Malay Reader" at the end of his " Beknopte Maleische
Spraakkunst" (Breda, 1839), pp. 120-123 ; and innumer-
able quotations from it are to be found in Werndly's
" Maleische Spraakkunst," and in Roorda van Eysinga's
" Mai. Nederduitsch Woordenboek ;" some also in van
Hoevell's " Aant. op de Sair Bidasari." 2
No. 3 (small 4to of 1 7 5 pages) y>j} ^j <^^ J*^
.^.c^ ^^ , relating the war of the Dutch Company with
the Chinese, and the well-known murder of the Chinamen
of Batavia under Valkenier. It is translated from the
Javanese.
No. 4 (small 4to of 80 pages)
This interesting work was composed at the request of the
Sefior Gornador ^^jj'at Malaka in the Muhammadan
year 1 193. It is an account of Malay observances during
the pregnancy of the wives of chiefs, the birth of their
children, &c. After the introduction it continues thus :
UK On p. 71 there is an
elaborate description of the bier of a king.
No. 5. See No. 18 of the Raffles' Collection.
No. 6 (small 8vo) contains :
I. (17 pages). An erotic poem, the title of which I
could not ascertain. The first verses are :
1 [B.M. 20.]
2 [L.
2 [L. 1696, and at p. 178; Bat. 112, 113; a Makassar translation, ap.
Matthes, 1. 1. No. 87.]
3 De Bruijn ?
MALAY MANUSCRIPTS. 47
And the last :
II. (14 pages). A love-letter in verse. The last verses
are literally the same as those on the two last pages of II.
of No. 9, commencing with
The last verses are :
III. (i i pages). The same as I. of No. 9.
IV. (27 pages). According to the end the title is
ta. The beginning verses are literally the same
as those of II. of No. 9.
V. (23 pages). A poem without title commencing:
The last words are :
v^;L. t^U^ ^y ^^
^s > >l^t>
No. 6^ (small 4to) contains :
I. (14 pages). The same as I. of No. 6.
II. (19 pages). A love-letter in verses. The last
verses are :
48 THE ROYAL ASIATIC SOCIETY'S
No. 7 (small Svo of 5 5 pages). According to the end
the title should be ^^ g)j <5^ ^^ )*> It is a
tragic love-story, as the hero dies.*
No. 7* (small Svo of 55 pages). Another copy of
No. 7.
No. 8 (small 4to of 175 pages, imperfect at the end ;
within : " Presented by Colonel W. M. G. Colebrooke,
6th July, 1832"), contains the &JJv&J! ajJjJ .f The Malay
author calls himself Muhammad sayn& son of Jalalu-ddm y
an Achinese of the Shafi-'i sect. A quotation from this
composition is to be found in van Hoe'vell's annotations
on the " Sair Bidasari," p. 378, where he cites p. 983
of the manuscript. The author of this work says, that he
took the subject from the ^^fcl^Jl II of Abu 'Abdillah
Muhammad ben Yusuf Assanusl Alhasani.
No. 9 (small 4to) contains :
I. (17 pages) c^i j** , A poem where flowers are
introduced singing pantuns, in this way :
* On page 45 we find :
txsv*^ (frequent spelling of the Ar. gyc*) mcrat, is in poetry used for
"to die," of princely persons whose disease is compared to an ascension to
heaven.
f Also mentioned in Bahru-ddln's list.
lie is the author also of a Malay work called ^-J ,q3 jl^j) t Q.;^
+] yc*\ & w-j^J JSA- d-jjJ) (a copy of which I saw at Batavia, small 410,
of about 30 pages). [See Van den Berg's " Verslag," p. 7, No. 36.]
Nila-utdma, name of a celestial nymph ("Tobasche Spraakkunst," 30,
VII. /;).
MALA Y MANUSCRIPTS. 49
II. (7 pages). A collection of pantuns, commencing
with :
s*J ^ US;
(See No. 6, II.)
No. 10 (small 4to of 53 pages).
c, . This work deserves being published ; its language
is pure, and the text, as far as I have examined it, not
corrupted.
NOTE i.
About the Author of IV. of No. 79 (Raffles Collection).
The author, who calls himself also ^^-Jl^Jl , from Rdnlr,
the place where he lived (U&w**), composed, besides the
JJ ^UwJ (No. 17), also the following works:
I. (No. 39, large 4to, India Office at Batavia)
II. (No. 3, small 4to, India Office at Batavia)
3 c/ 1 c
III. (No. 24, Library of the India Office at Batavia)
vJ^-^l? C^^ ^>*^* ^ ^LJ^I j\j~\ , commenced under
Iskandar II. 'Ala uddin, and finished under the queen
Tdju-Palam Tsafiyatu-ddln?
IV. (No. 14, Library of the India Office at Batavia)
1 [Van den Berg, 1. 1. p. 8, No. 40.] 3 [/*., p. 3, No. 10.]
SECOND SF.RIES. VOL. II. E
50 THE ROYAL ASIATIC SOCIETY'S
V. c^b ~1
composed by order of 'Sultan Muqul
Mardyat S/id/i* against the tenets of the Pantheistical
sect, the followers of which were put to death by the said
king of Achih, their books being burnt before the mosque
Bcyturrahman. I saw a copy of this work at Barus
(small 4to of 40 pages), from which I took this notice.
VI. &> aJL^ ^j] <u^lo J1=JI ^$f> <> &^J
^ ^yL.1 i ^J pjb i$f- o^^-i a refutation of
Shamsu-ddiris heretical tenets (cf. VII.).
VII.
J C*J>
Ai c) 4 ^ c^^ 1 ' composed under queen Tdju-l-
'dlam Tsafiyatu-ddln SJiah,\ daughter of Sultan Iskandar
Muda Johan bardowlat, son of Sultan 'Aid u-ddin 'All
Ri-dyat Shah, son of Sultan Farmdn Shah, son of Sultan
Mutlafar\ S/id/i, son of Sultan 'Indyat Shah. It is divided
into two books, the first giving an account of the religions
from Adam till Muhammad, and the second summing up
the heterodox tenets of several Muhammadan sects. The
purpose of the author was to combat the opinions of
Shamsuddln of Pasey and his followers. A copy in
small 4to (of 72 pages) is in my possession.
VIII. c^UJI JaOJ &LJI C U A fragment of this
work is found in a manuscript belonging to the Batavian
Society (No. 5 5 ?).
0> ->
IX. ^Jo^l jjJ ^iJs*=Jl &a=*. A copy of this work
exists in the Library of the Batavian Society (No. ?).l
t atfc,
j| Jsu^J as the Arabic introduction has. Shamatara is an Arabic
corruption of Samuddra, the ancient name of Pasey, which occasioned the
whole island to be called by the Portuguese, who sailed with Arabic pilots,
Sumatra, a name with which natives, not used to mix with Europeans, are not
acquainted.
H X. ^JbUv*JI Ll^-o. See Roorda van Eysinga's "Indie," III. I,
pp. 413-435 [and Van deu Berg, 1. 1. p. I, No. 3.]
MALAY MANUSCRIPTS. 51
Most of these works are directed against the popular
writings of HamzaJi of Barns* and the above-named
Shamsu-ddin of Pasey. The works of Hamzah are, as
far as I know :
I. g*)\sb\ )\y~, 1. I saw a copy of this at Barus
(small 8vo of 24 pages). I read only the preface, which
says, that it is an abridgement of a greater work of the
same name and by the same author ; and that there are
three works of this name, the two already mentioned, the
large and the abridged one, and one treating on <$&> ,
^5^1^ and ^j&ju*. This is all I could read, as the
owner would not lend it me even for a day.
II. <^M ;>c-~ j*&> an allegorical poem/f wherein the
soul of man is spoken of as that of a bird (kalow tdrbagn
siburung pingcy, 'aldmat badan di makan ulat, if the
pingey flies away, it is a sign that the body will be eaten
by the worms).
III. &\y> j*. An allegorical poem, wherein mankind
is spoken of as a vessel tossing about on the waves. A
small fragment is in my possession.
IV. }*> gV~ j** . A copy is in my possession
(small 4to of 14 pages). It is also an allegorical poem,
speaking of mankind as forlorn and indigent.
V. ^la*~J1 .qUtfJl j~$\ < &&, a short exposition of
God's nature, qualities, and works. Werndly knew it
(see his "Boekzaal," p. 354). It is quoted in the second
book of the Tabydn (see above, in the specification of
Ranir's writings, No. VII.) as a book deserving to be
burnt. \
^^jA^uJu! Fantsur being the ancient name of Barus [see above,
vol. i. p. 164]; hence the Barus camphor *lj >*^") * s ca ^ e ^ in- Arabic
t The poems of Hamzah were yet much read in Valentyn's time, but that
he was a native of Barus that author did not know (see " Beschrijving van
Sumatra," p. 21).
J The other books, the author of the Tabydn speaks of in this way are the
1 the &*&. the
the d-JJ M and the
E 2
52 THE ROYAL ASIATIC SOCIETY'S
VI. ^o j^w ^UeJ^. Jj cs^-" <-r*^- It is men-
tioned in the Tabyan, and seems to be an exposition of
the sayings of the prophet.
VII. j&\ j j*& . A fragment is in my possession. It
has the same tendency as No. III.
The works of Shamsu-ddln * of Pasey are :
^ . It is cited in the second book of the
Tdbydn. A badly mutilated copy is in the Leyden
University Library (No. 1,332). The Sultan in whose
reign it was composed is there only called u^L* t>j^y*
II. <^jj*ojJd\ &y*z* ^q^^O CJ*^ ^ saw a ^ P a ^ang a
copy (8vo of 1 6 pages), but the owner would not part
with it. It is a commentary on the anything but trans-
parent poems of HamzaJi of Barns.
III. ^^Jl i\j+. Werndly (" Maleische Boekzaal")
knew it, and says of it, that it is divided into 211 questions
and answers, explaining the principal religious terms. In
the preface to his " Grammar " a small quotation from
this work is given.
NOTE 2.
TJie Manuscripts of tJie India Office not mentioned in my
" Kort Verslag der Maleische Handschriften van
het E. I. House te Louden?
1. ^ gVjl^ ^K=*. (See No. 15 of the Raffles
Collection.)
2. Another copy of the ^U ^!JCL &y> &\A . (See
No. 2 of the Raffles' Collection.)
* He calls himself sometimes <OJIj^- j| . lie seems to have lived
at Achih (Ar. 6^\ ). A namesake of his is cUs^ $ U I ^.jJ l^^a-iv
<OJI \*CL9 ^j|, and is cited as the author of the above-mentioned cCj^l*
11 .J1.
MALA Y MANUSCRIPTS. 53
3. ^JsxS^.4.^ (17 pages in No. 2,906,* 4to). Arabic
with an interlinear Malay translation. It contains the
first precepts of the Islam in questions and answers. The
commencement is : " If people enquire of you : ' What is
the iman f ' the answer is : ' I believe in God/ " &c. The
author is p+>\J\ ^\ yc* ^\ ^ VX^SK* <^-JM f\, sur-
named of Samarqand (^JjJj^vJl). This little book
goes universally by the name of Samarqandl. Copies
with an interlinear Javanese translation t are numerous
in the west of Java. A commentary on it (,5^ Cj^*
^JvxSj^vJJ) is in the Library of the Batavian Society
{No. 29) ; it has an interlinear Javanese translation.
Two copies in Sundanese are in my possession, one of
which is in the Arabic character.
4. (No. 2,672, folio) contains :
I. (133 pages). Another copy of the <j-U) ^W- fA*
It is of the same version as the other manuscript (No.
2,610).
II. (127 pages). Another copy of the J}^ <k\=*
^-Xws^ (or il^3 c;U djK=> , so called after a miraculous
chopping-knife the hero was possessed of). It seems to
belong to the same recension as No. 2,877.+
AMSTERDAM, November 25, 1865.
* The other 51 pages of this volume contain: (i) the several positions of
the body when praying ; (2) the application of the five letters of J**^J I
to the five obligatory prayers ; (3) the formulas of prayers for the dead ; and
{4) on marriage (^^J AZ\
f A copy is in the Library of the Batavian Society (No. 26).
There may be other Malay manuscripts in the Library of the India Office
which I have overlooked, the Persian, Arabic, ram Malay manuscripts being
mingled together. I am in hopes the deficiencies in this notice may be filled
up by other scholars, who will also call attention to the many valuable Malay
manuscripts in the Libraries of London. A New Malay Chrestomathy is
urgently needed at the present time, as those published by Marsden, Meursinge,
and de Hollander, are anything but trustworthy, each of the texts they contain
having been taken from a single manuscript only. It is only by a careful com-
parison of many that a text can be furnished which may be depended upon by
persons desiroXis of obtaining an adequate idea of the grammatical structure of
the Malay language, and reluctant to trust the assertions of those who pretend
that Malay is devoid of grammar.
54 THE ROYAL ASIATIC SOCIETY'S
NOTE 3. 1
The proper name si-lindiing dalima (Raffles Coll.,
No. 7, III.) had better be taken to mean "she who excels
the granate-apple in beauty." Thus we have in Menang-
kabau tales the name of a princess si-lindnng bnlan, "who
surpasses the moon in radiance" (compare the Batak
name nan-chilok mata ni ari, " Miss Sun-stealer," see
" Bataksch Leesboek," IV. p. I 5 7). In the prose edition
of the poem, the heroine of which goes by the name of
j*J J cJcJu- , the proper name ^J^ 9^v also occurs
(p. i 5 of the I.O. MS.) ; the heroine's father is there called
Dewa Laqsana, who had transformed himself into a
granate-apple which, being eaten up by Sari Buniyan,
became the cause of her pregnancy.
NOTE 4.
To Raffles Collection, No. 62, V.
Mr. Engelmann informs me that there is also a
Sundanese recension of this work, a copy of which (small
4to, 36 leaves) is in his possession. In it the prophet's
advice is not only given to his daughter Fdtimah, but
also to another female whom he calls Mnrtasiyah. I
suspect that this proper name is nothing but the feminine
C J
of ^cfbSy* (a cognomen of Fatimah's husband l Alt\ so
that the second female recurring in the Sundanese
recension has got in through misunderstanding. It is
worth noting that <j& has become s, since this consonant
is pronounced / both in Sundanese and Javanese (e.g.,.
vialarat i.q. Arabic yu). But as ^ is pronounced
as a sibilant by the Persians and Indian Muhammedans,
it is not improbable that the Sundanese redaction was
subject to Kling or Bengali influence. The forming of
female proper names by means of the termination ah is a
1 [The following additional Notes (3-10) .ire found in the Dutch translation*
of this " Account." See the note at the beginning of this article.]
MALA Y MANUSCRIPTS. 55
well-known fact.* May not the Malay word rubiyah be
explained as formed from ^) (my lord), as the Malays
generally know so little of Arabic that it could not well
*
occur to them that ^j is properly used only in reference
to God ? On the west coast of Sumatra they call rubiyah
a female who keeps the five prescribed prayers (j^^^
jj&yi), reads the Koran, and, in short, lives in the fear of
the Lord, while strangely enough ! mande (mother)
rubiyah, is a nickname for a procuress, or the keeper of a
house of ill-fame. The name d*-J; (Raffles Coll., No.
70, II.) appears to convey an illusion to rubiyah. The
usual cognomen of Fatimah in Malay is lyb^l ; hence
they say of an unchaste woman : *J
NOTE 5.
Ad No. 79, II., and No. 77, III.
On aJ^S for &j see the "Annotation to Panjatandaran"
(Leiden, 1866), p. xxii.
NOTE 6.
Ad Raffles Coll., No. 2.
Mr. Cohen Stuart informs me that from an interpreta-
tion given him by a Balinese, he has also been led to
correct bisuwa into bkishawa a reading which I had
already suggested in my pamphlet, " Taco Roorda's beoe-
fening van't Javaansch bekeken" (p. 22, note 2).
NOTE 7.
Ad Raffles Coll, No. 7, II.
Concerning inu compare vol. iv. of the "Bataksch
Leesboek," p. 213.
* The Javanese from Bulkiyah (for Bulqis] is probably due to a Menang-
kabau form Bulkik (cf. "Tobasche Spraakkunst," p. 50).
56 MALA Y MANUSCRIPTS.
NOTE 8.
Ad Raffles Coll., No. 14.
Rawi sarangga, as one of the names of Panji, must
mean "ray of the sun," and have been chosen as a
pendant to ckandrakirana> " ray of the moon," the cog-
nomen of his betrothed. Compare in Batak the use of
bulan or mata ni bulan for a daughter, and of mata ni ari
for a son. Sarangga must be a corruption of sdrangga,
" light," so that rawisarangga literally means " sunlight."
NOTE 9.
Ad Raffles Coll., No. 30.
Chi-rebon is Sundanese, and originally the name of a
river. The Javanese, not understanding its meaning,
have changed the word into Cherbon, Chrebon.
NOTE 10.
Ad Raffles Coll, No. 65, note.
Jayeng kasuma has been rendered according to native
ideas (jayeng kusuma being, i.q., jayeng sekar). On this
subject I owe to Mr. Cohen Stuart the following interest-
ing note : " I would rather think of a real field or
meadow, than of a battle-field, and take kusuma in the
sense of a woman (see 'Bataksch Leesboek,' vol. iv.
p. 127, note 2). Such terms of honour, taken from love
conquests, are not unknown in Java ; I have known at
Solo a certain Jayeng Resmi * (' conqueror in love '),
Jayeng Rana (the last word in the sense of pudendum
, &c."
* Resmi, however, must mean "beauty," and not "love ;" see "Tobasche
Spraakkunst," p. 52, note 2.
t In Malay rana ( ,\|^) means " a princess," and is a corruption of ratna
(M.,P. 50.
JOURNEY TO GUNONG BENKO. 57
IX.*
MEMORANDUM OF A JOURNEY TO THE
SUMMIT OF GUNONG BENKO,!
OR THE SUGAR LOAF MOUNTAIN IN THE INTERIOR OF
BENCOOLEN. 2
["Malayan Miscellanies," vol. ii. (Bencoolen), 1822, pp. i-n.]
THIS mountain, which stands detached from the regular range of
hills, forms, by its peculiar and remarkable shape, an excellent
landmark on this part of the coast. It lies about 18 miles N.E.
of Bencoolen, but its exact position and distance had never been
correctly ascertained. Two attempts had been made by Euro-
peans to ascend the mountain, but without success, and a general
impression prevailed that it was utterly impracticable to gain the
summit. Remarkable mountains of this description are generally
believed by the natives to be the residence of spirits, arid their
summits are considered as Kramats or places of peculiar sanctity. 3
A Kramat of this nature was said to exist on the top of the Sugar
Loaf, and it was reported that the natives sometimes adventured
to visit it from motives of superstition. It was therefore resolved
to make another trial, in the expectation that it might afford the
means of correcting and extending the observations already com-
menced on the coast, with a view to a more accurate survey of
this part of the country.
* [For the foot-notes the editor is indebted to the Hon. D. F. A. Hervey.] '
1 We should spell it Bengkok. This, which means " winding " or " crooked,"
hardly tallies with the English name ; but the Malays doubtless refer to.different
characteristics, possibly indicating the tortuous path which leads to it.
2 The Malay name to which this is the English approximation is Bengkahiilu.
H&lu is source, of a river, and might imply that the Bencoolen river had its
source in the Bengkok mountain. But while the mountain is here called Benko',
it figures in Baron Melvill's map as Boengka (Bungka) [and in Favre's "Dic-
tionary," as well as in the account of the Dutch expedition (" Midden-Sumatra,"
Reisverhaal, 1881) as Bukit Bungkuk, or Dwarf Mountain. Veth, in his " Geo-
graphical Dictionary," calls it Bukit Bongso.]
3 For accounts of similar legends regarding Mount Ophir near Malacca and
Gunong Dato in Rembau, see "Journal S.B.R. A.S.," No. 13, p. 257 ; and " N.
and Q.,"No. 2, p. 41, and Note ; [on haunted mountains generally, "Journal Ind.
Arch." i. 319 ; G. A. Wilken, " Het Animisme," p. 139 ; J. B. Neumann, " Het
Pana- en Bila-stroomgebied in Sumatra" (1886), p. 294, and "Midden-Sumatra,'*
Reisverhaal, vol. i. pp. 150 and 405.] Besides these Kramats, there are those
formed at the tombs of holy or great men, where Chinese, Hindus, c., all
come and pay their vows, whatever may have been the religion of deceased,
so long as he has established his reputation.
5 8 MEMORAND UM OF A JO URNE Y TO
A party of gentlemen accordingly proceeded from Bencoolen on
the loth of June, 1821, for the purpose of effecting this object.
They crossed the Bencoolen river a little above Tanjung Agung,
and proceeding through the Lumba Selapan district, halted the
first night at Lubu Pooar, a small Rejang village on the banks
of a stream which falls into the Sungey Lamow. 1 Thus far the
journey was accomplished on horseback, but it was found im-
practicable to carry the horses any farther, and the party pro-
ceeded on foot to Punjong, a respectable village situated on the
banks of the Simpang-ayer, and the residence of the Pasirah of
the tribe of Marigi, the chief of the four into which the Rejangs
are divided. The others are called Bermani, Saloopu, and Joru
Kallang. On the third day they reached Rejak Bessi, the last
village in the direction of the mountain, where they rested for the
night. It is situated on the Ayer Kiti, a stream which falls into
the Simpang-ayer below Punjong. The journey from Lubu Pooar
to this might with ease have been accomplished in one day
instead of two had the weather permitted.
The mountain was now to be attempted, and in order to ensure
success, it was arranged to pitch a tent in the forest in case the
ascent could not be accomplished in one day. From Rejak
Bessi they proceeded over hilly ground gradually rising for about
five miles, when they found their progress impeded by the in-
creasing steepness of the ascent, and then halted under an over-
hanging rock, where the tent was pitched, as it was impossible to
carry it any further, even if space could have been found to erect
it on. The course from Rejak Bessi was through deep forests
which precluded them from seeing the mountain. The last view
they had of it was at Rejak Bessi, which it appeared to overhang,
and whence they were able to form some idea of the difficulties
they were likely to encounter from the steepness of the ascent
and the precipitousness of the declivities. Soon after quitting
Rejak Bessi they crossed a small river on a temporary bamboo
bridge thrown across a deep chasm between two rocks, which
confined the stream within a narrow channel after being precipi-
tated over a fall of considerable height. A fine view of this fall
was commanded from the bridge, which was itself suspended
about one hundred feet above the stream, and the whole formed,
with the surrounding forests, a beautiful and romantic scene.
About ten o'clock they commenced the ascent of the cone along,
the rocky bed of a mountain torrent, until they arrived in front of
a perpendicular face of bare rock stretching completely across the
ravine which had hitherto afforded a passage, and seeming to bar
all further progress. This difficulty was surmounted by placing
two of the longest bamboos against the rock underneath where
the bare root of a tree projected from above ; by the aid of these
1 More correctly Li mail or Lcmau.
THE SUMMIT OF GUNONG BENKO. 59
held fast at the bottom, and afterwards secured by a rattan at the
top, they succeeded in clambering up to the tree which overhung
the precipice. The next acclivity terminated at the head of
another ravine, where their progress was again checked by a
jutting rock rendered moist by the trickling of a small spring of
water from among its crevices. Here the guides declared that
further ascent was impracticable, and that from thence the party
might return as soon as they pleased. The fact is, they were
extremely averse to their proceeding, fearing the vengeance of the
evil spirits if they conducted strangers to the summit ; they were,
therefore, advising to return at every difficulty, and the ascent was
ultimately accomplished without their aid, or rather in spite of
them. The appearances around were calculated to confirm this
assertion, but before determining to return they examined the
extent of the precipice, and crossing the ravine, perceived that the
opposite side, though almost perpendicular, had a thin coating of
soil and moss, with numerous roots of trees half laid bare, by
laying hold of which with the hands and placing the toes in
the niches, they at length reached the ridge which formed the
right-hand shoulder of the hill. Along this a path was found,
sometimes along the base, sometimes over the face, of a suc-
cession of bare masses of rock, which it was necessary to
clamber over by the aid of such twigs and roots as occa-
sionally fastened themselves in their fissures. The last of
these precipices was perhaps the most dizzy and dangerous, as it
was necessary to make a step or two on a narrow ledge on the face
of a cliff of such height that the eye could not discern the bottom,
and thence catch at a dry stump barely within reach, by swinging
from which it was possible with a considerable effort to clear the
rock. The denseness of the moss and the stunted appearance of
the trees now indicated their approach towards the top, and at
length about two o'clock they found themselves on the summit.
This was a bare spot of not more than four or five yards in.
breadth with a precipice on each side partly concealed by brush-
wood. Of those who set out together from the foot of the hill a
few only reached this point, by far the majority giving up in
despair at different parts of the ascent, but the labour of those
who persevered was amply recompensed by the view which)
opened from the summit. The line of the coast from Laye 1 on*
the north to a considerable distance beyond Buffalo Point on the
south was distinctly marked ; the vessels in the basin of Rat
Island were distinguishable with the aid of a glass, and the white
ramparts of Fort Marlbro' were easily discerned. To the south,
they looked down on the hills of Bukit Kandees,- or the Lion's
Hump, and Bukit Kabut (the hill of mist), which formed a straight
1 This appears as Lais in the Dutch maps.
2 Kandis is the name of a tree (Carcmia mergucnsis) bearing an edible fruit..
<5o MEMORAND UM OF A JO URNE Y TO
line with the Sugar Loaf. Inland the view was obscured by a
cloud, which was evidently directing its course towards the hill,
and it was necessary, therefore, to take the desired observations
and bearings with all possible dispatch. This was done with a
small compass, none of the larger instruments having been got up.
The character of the vegetation was decidedly Alpine, the rocks
and trunks of the trees being covered with dense moss, 1 and many
of the shrubs belonging to genera of higher latitudes, such as
vaccinium, rhododendron, &c. There is also found here a shrub
which the natives consider a substitute for tea, 2 remarkable by its
thick glossy leaves ; it will form a new genus in the family of the
Myrtaceee. Having finished their observations, they made haste
to descend, as the cloud was now rapidly approaching the hill and
threatened a deluge of rain. They found the descent fully as diffi-
cult as the ascent had been, but it was occasionally facilitated by
fastening a long rattan to a tree above, and then sliding along it
down the steepest places. It was necessary, however, to be
cautious not to slide with too much velocity in order to be able to
keep a footing when the rattan slipped from the hand. When
they had got about half-way down, the cloud, which had now
enveloped the hill, burst in a flood of rain, and rendered the
footing still more insecure. The steepest parts, however, were
then past, and the trees for a short while afforded some protec-
tion, but by the time they reached the lower ravines, the waters
began to swell, and the latter part of the descent was in the very
bed of the torrent. They arrived at the tent about an hour before
sunset, and found the spot completely flooded ; the rain had in no
degree abated, and it was impossible to find shelter for the whole
party of natives, &c., which was very numerous ; it was, therefore,
determined to make a push forward to Rejak Bessi, rather than
pass the night in so uncomfortable a situation. A sharp walk
brought them to the village soon after dark, and a good night's
rest repaired the fatigues of the day. The next day was spent at
the same place, both for the purpose of resting the people, and of
bringing up the tent which had been left in the forest. On the
1 6th they travelled to Punjong, and the following day they com-
menced their return by another route, striking across the country
in the direction of Bukit Kandees to the Bencoolen river. Sam-
pans had been previously ordered to be in readiness at Tanjong
1 On Gunong Beliimut in Johor, at from 2,000 feet onwards, the trees and
saplings are thickly clothed with dripping moss, making stems only wrist-thick
appear thigh-thick ; hence the name "the bc-mossed mountain," from himuf,
moss.
2 There is a similar shrub on the summit of Mount Ophir at a height of about
4,200 feet, a spray of which stirred in boiling water gives it a very pleasant
aromatic fragrance. This quality was discovered at the suggestion of Governor
.Sir F. Weld, who discerned in the plant a likeness to others he had seen
possessing this quality on a mountain in one of the Australian colonies.
THE SUMMIT OF GUNONG BENKO. 61
Sanei, and they arrived there about eleven o'clock, having in the
latter part of the journey forded the main stream of the Bencoolen
river no less than eleven times. About twelve they embarked on
the sampans, and placed the baggage and some of the followers on
bamboo rafts ; the first part of the course was a constant succes-
sion of rapids, in shooting down which some management was
necessary to avoid being upset upon the trunks of trees and other
obstacles that lay in the way. Twice, by being driven against
these, the boat was filled with water and with difficulty saved from
being swamped. Below the junction of the Rindowati, 1 the depth
of the river increased and the current became more regular ; and
at length they landed near Bencoolen about nine at night, having
thus accomplished, aided by the rapidity of the stream, in one day
what would have occupied several in ascending.
Gunong Benko' is not estimated to exceed 3,000 feet in height,*
but its shape, and its standing boldly out from the general range
of hills, render it the most remarkable visible from Bencoolen. It
is almost entirely composed of masses of basalt or trap, which is
the most prevalent rock along this part of Sumatra. The whole
of the country traversed on this occasion is exceedingly broken
and irregular and but thinly inhabited. In the neighbourhood of
the hill it is a complete forest and very wild, presenting an infinite
number of romantic and beautiful views. The soil near the rivers
is remarkably rich, and that of the forest tracts is little inferior,
particularly in the bamboo groves, which indeed are generally
found to prevail on the finest lands. The greater part of the rice
is cultivated in ladangs, 2 but there are a few sawahs. At Tello
Anou is a small nutmeg plantation where the trees have never
been manured, yet seem as thriving as any about town. The
forests abound with noble timber trees ; few animals were seen ;
of monkeys the Kra (S. fasticularis), and Chingkau (S. cristatd),
were the most common, and the loud cry of the Siamang (S. syn~
dactyla) was frequently heard, though they did not come in sight.
It is very singular to observe the young of the Chingkau and
Simpai (S. mclalophos) embracing their mothers, that of the former
being fawn-coloured, while the adult is nearly black, and the latter
having the young black while the mother is fawn-coloured, ap-
pearing exactly as if they had exchanged young ones.
At about half the height of the mountain the temperature of a
small shallow spring was tried where it oozed from a cleft in a
1 Perhaps for rindu hdti, desire of the heart i.e. an enchanting stream.
* [In the new atlas of the Dutch East Indies, by Stemfoort and ten Siethoff,
the height of the " Suikerbrood " or Sugar Loaf Mountain, is given as 1,029
metres; in Veth's "Geographical Dictionary" as 3,287 Rhenish feet. In the
latter it is also mentioned that on the Ilth of August 1857, Lieut. F. G. Steck
made a successful ascent.]
2 Ladang, plantation on high ground, dry cultivation ; sazuaA, padi or rice-
field in swampy ground, wet cultivation.
62 MEMORAND UM OF A JO URNE Y TO
rock and found to be 68 Fahrenheit. The temperature might,
however, have been lowered by evaporation ; therefore it can
scarcely be assumed as a true mean temperature, or employed in
calculating the height. It may, however, be remarked that the
mean temperatures given by Mr. Leslie for the level of the sea in
the different latitudes will certainly not apply to the low latitudes
in the eastern islands : 83, which is given as the mean tempera-
ture in latitude 3, is far too high for Bencoolen, where the range
of the thermometer throughout the year is usually from 74 to
5, rarely falling below 70 or rising above 87 or 88 . 1
The people who inhabit the interior are Rejangs, and speak a
different language from the Malays ; they extend northward as far
as Laye. From the Sillebar 2 river southward, the Serawi tribe
prevails, and the space between that river and the Bencoolen is
occupied by the tribe of Dua-blas. Similar customs, with slight
shades of difference in each, prevail among all these tribes. At
every village where the party stayed for the night, the gadises, or
virgins, paid a visit of ceremony in the evening, making a present
of betel or siri, and receiving some trifling articles in return.
This custom is general, and it is necessary to be provided with a
sufficiency of fans, looking-glasses, or such like articles in con-
sequence, as the number of the young ladies is often very con-
siderable. Sometimes an entertainment is given in honour of the
visitors, and then all the beauty of the surrounding villages is also
called in.
These entertainments, which take place also on occasions of
marriages, &c., are not unamusing, and to a European have the
additional interest of novelty and originality. They are given in
the Balei, or public hall, a large building generally in the middle
of the village, appropriated to such purposes and to the accom-
modation of strangers, &c. When European visitors are present,
the ceremonial is generally as follows : The gentlemen being
seated near the upper end of the room, the gadises, dressed out in
their best attire, make their appearance about nine o'clock, and
seat themselves on the floor, previously spread with mats, in a
semicircle, with their attendant matrons behind them; each brings
her siri box of various material and elegance according to the rank
or wealth of the parties. The chief of the village or one of the
elders then makes an harangue in the name of the ladies, wel-
coming the strangers to their village, and concluding with the
presentation of the betel. An appropriate answer is then to be
made, and, after taking out the siri leaves, a small present is put
into each box, proportioned in some degree to the rank of the
parties ; this, however, may be put off at pleasure till the conclu-
sion. The amusements of the evening then commence, which
1 This is almost identical with the Singapore temperature.
a That is Salebar or Selebar, from Ubar, broad.
THE SUMMIT OF GUNONG BENKO. 63
consist, on the part of the young people, of dancing and singing ;
and of the old, in smoking opium in a circle apart to themselves.
The musical instruments are commonly kalingtangs, which are a
species of harmonicon formed of a series of small gongs arranged
on a frame. 1 A space is cleared on one side for the dance, which
is performed by five or six of the young gadises ; the step is slow
and sailing ; the salindang, 2 or scarf, is adjusted in a particular
manner over the shoulders so that the ends may be taken in the
hand, and the motions of the arms and management of the flowing
scarf are not the least graceful part of the performance.
The singing of pantuns in alternate contest is an amusement
which seems to be peculiar to the Sumatrans, and of which they
are very fond. It may either be formally commenced by two
parties, who seat themselves opposite to each other after having
danced together, or it may be begun by one of the ladies from the
place where she happens to sit. She begins a series of pantuns
in a kind of recitative or irregular song ; a bujang, or young man,
answers her in the same manner, and the contest is kept up in-
definitely, or until one of the parties is unable to give the proper
answer. The girls and young men relieve each other occasionally
as one or other happens to get tired.
The Malay pantuns, strictly so called, are quatrains, of which the
first two lines contain a figure or image, and the latter give its
point or moral. Sometimes the figure or comparison is accurately
suited to the subject, and then the application may be omitted in
recitation, the more to try the ingenuity of the respondent ; some-
times the whole is couched under one or more figures ; while in
many the beginning seems only intended as a rhyme, or, at least,
has not obvious connection with the subject. 3 Among the Rejang
.and Serawi people a greater latitude is allowed to the seramba* or
pantun, the figure is pursued to greater length, and a kind of
measured prose is often employed in place of confining themselves
to the trammels of verse. The pantun is frequently framed into
a kind of riddle, whose meaning it requires some ingenuity to dis-
cover, and a blundering answer to which excites much mirth.
These pantuns frequently contain words derived from the lan-
1 On the same principle as the Javanese gambang, in which pieces of wood
take the place of the glass in the harmonicon, ranged over a hollow case or
trough.
2 In the Straits selendang.
3 In the majority of Peninsula pantuns no meaning is to be attached to the
first two lines as far as any connection with the remaining two is concerned.
ISee H. N. van der Tuuk's "Bataksch Leesboek," vol. iv. p. 107, ff. ; and J.
Pijnappel in "Bijdragen" for 1883, pp. 161-75.]
* [The scope of the serambahs is somewhat wider than that of the pantuns.
They are called berduwi, when young people sitting in rows sing them in the
daytime ; begandai, when sung standing ; and nyambri, when they are sung in
the evening, sitting. Specimens are given in "Midden- Sumatra," iii. 2 (iS8i),
PP- IS-35-]
64 MEMORAND UM OF A JO URNE Y TO
guage of Sunda, which has been partially introduced into the
poetry of all the tribes to the southward of Kataun, while to the
northward the Menangkabau dialect prevails. The origin of this
distinction is referred to the period of the wars between Imbang
Jaya, a Javanese prince, and Tuanko orang Muda of Menang-
kabau, the traces of the Sunda dialect marking the limit of the
possessions of the former.
In these contests the pantuns are supposed to be extempo-
raneous effusions, and perhaps sometimes are so in reality, but in
general their memories are so stored with established verses, that
they are not often put to the task of invention. Of their force
and meaning it is extremely difficult to convey a just idea by any
translation : whoever has attempted to transfuse the spirit of an
Oriental composition into a European language must have felt
the difficulty of doing so satisfactorily, where the whole structure
of the language is so different, and the whole current of ideas seems
to flow in another channel. This is particularly the case with the
pantun, whose chief merit consists in conciseness and point, and
in conveying a deeper meaning than is contained in the literal
words and expressions. The figures and allusions are often quaint,
but occasionally evince a considerable degree of poetic feeling and
force of imagination.
It is not only on these set occasions that pantuns are employed,
they enter largely into their more common intercourse, and are
essential accomplishments to all who aspire to a character of
gallantry, or who hope to woo and win their lady's love. Skill
and readiness in this kind of poetry is with them a passport to
female favour, much in the same way that a readiness at compli-
ment and flattery in conversation and the art of saying soft
nothings serves the European candidate for the smiles of the fair ;
much of this kind of flirtation goes on independently of the open
and public display of skill, and is often accompanied with the
interchange of flowers and other mute symbols, which have all a
mystic meaning intelligible to those who have been initiated into
this secret mode of communication. Making due allowance for
difference of customs, of wealth, and of progress in civilization,
there seems to be much in the conduct of these entertainments
and in the general deportment of the Sumatrans towards women
to indicate that they possess somewhat of that character of
romantic gallantry which marked our own earlier ancestors, and
there might be found as much delicacy of feeling and perhaps
more of the poetry of the passion in their courtships, than in the
over-refinement of modern English society. It must also be
remembered that no people can be more jealous of female honour
than the Sumatrans, and that all this is conducted with a strictness
of decorum far greater than is observed in the free intercourse
permitted by European custom.
THE SUMMIT OF GUNONG BENKO. 65
A few examples of the different kinds of pantuns may not be
unamusing, though it would be as difficult to convey an idea of
the effect with which they are applied at the moment and on par-
ticular occasions, as to record the sallies and evanescent sparkles
of wit that sometimes enliven our own tables, and which like the
champagne that inspires them, would seem flat and dull if repeated
next morning. Of the Malay pantun of four lines, several exam-
ples have been already given by Mr. Marsden, the strictness of
their form and limits perhaps render them better suited to transla-
tion, but they are considered by the people of the interior as too
stiff and prosaic and as deficient in that boldness of allegory and
recondite allusion which they consider the perfection of their own
longer ones. The following are specimens of the Malay pantun,
applicable to different occasions, such as the opening of a court-
ship, complaints of inconstancy, coyness, &c., expressions of com-
pliment, of affection, of doubt, of ridicule or displeasure, and
others which the reader may much better imagine to himself than
they can be explained by words. In some the connection of the
figure and the sentiment will readily be perceived, in others it is
obscure, particularly where the allusions are idiomatic or have
reference to popular fables or belief, and in others there is none
at all.
Memuti 1 umbak di rantau Kataun
Patang dan pagi tida berkala
Memuti bunga de dalam kabun
Sa tangkei saja iang menggila.
" The waves are white on the shore of Kataun, night and day
they do not cease to roll; many are the white flowers of the
garden, but one alone hath made me distracted with love."
Guruh berbuni 2 sayup sayup
Orang di bumi samoa bembang;
Jika ada angin bertiup
Ada kah bunga mau kambang.
" The thunder rolls loud and deep, and the inhabitants of the
earth are dismayed ; if the zephyrs should now breathe upon it,
will the flower expand its blossoms ?"
Ayer dalam bertambah dalam,
Ujan di ulu bulum lagi tedoh ;
Hati dendam bertambah dendam,
Dendam daulu bulum lagi sumboh.s
1 In the following notes the common (Straits) Malay forms are referred to :
mgmutih, from fritik, white ; petang.
3 Bunyi; Sayup means primarily as far as the eye can see, or the ear can
hear.
3 Seihboh) cured.
SECOND SERIES. VOL II. F
66 MEMORANDUM OF A JOURNE Y TO
" The deep waters have increased in depth, and the rain hath
not ceased on the hills ; the longing desire of my heart hath in-
creased, and its former hopes have not yet been accomplished."
Parang bumban l di sabrang.
Pohon di hela tiada karuan ; 2
Bulan pernama niatalah bindrang,
Sayangnia lagi di saput awan.
" The reed is cut down on the other bank, it is now at the
mercy of the stream, draw it towards you ; the moon is at the full
and shining, a cloud as yet intercepts her light (literally affec-
tion)."
Ulak berulak batu mandi.
Kian berulak tenang jua ; 3
Hindak bertunah tunah ati,
Dewa membawa bembang jua.
" The stream becomes still behind the sunken rocks, and the
waters are smooth and calm amid the eddies ; I try to quiet the
uneasiness of my heart, but there is a fairy that still disturbs its
peace."
Permata jatu di rumput, 4
Jatu di rumput bergelang gelang ;
Kasih umpama ambun di ujong rumput,
Datang matahari nischaya 5 ilang.
" The jewel fallen on the ground, though fallen among the
grass, is glittering still, but thy love is like the dew on the flower/
quickly disappearing when the sun comes forth."
Telah lama tiada ka rimbo,
Bumban berbua garangan kini ;
Telah lama tiada bersuo,
Dendam berubah 7 garangan kini.
" It is long since we have been to the forest, perhaps the bum-
ban (a species of flowering reed) is now gone to fruit ; it is long
since we have met, perhaps thy affections are now estranged."
1 Bcmban ; this plant, except for its more branching character, somewhat
resembles the bamboo. The stem is hollow, but contains a pith ; it is split
up to make sieves and mats, and the leaves are plaited into baskets. There
are two kinds, mangifera thai pa and maranta disticha.
2 The tree is being drawn (by the stream) uncertainly i.e., hither and thither,
benderang.
3 The eddies whirl over the bathing rock; often though they whirl (the
surface) is calm.
4 Rumput, the grass ; but many plants are called rumput.
i 6 Lit., surely. Lit., on the point of the grass.
7 Lit., changed.
THE SUMMIT OF GUNONG BENKO. 67
Jeka sungguh bulan pernama,
Mengapa tiada di pagar bintang ;
Jeka sungguh tuan bijaksana, 2
Mengapa tiada dapat di tintang. 3
" If indeed the moon is at the full, why does she not appear in
the midst of her stars ? If indeed thou art true and faithful, why is
it denied me to behold thee ? "
Unggas bukan, chintayu bukan,
Kira-nia daun selara tubbu ;
Aches bukan, Malayu bukan,
Pandei nia amat bermain semu.
" 'Twas not a bird, neither was it the Chintayu,* 'twas only a
withered leaf of the sugar cane ; she is not of Achinese, neither
of Malayan race, yet is she deeply skilled in the arts of deceit."
Bagimana menangkap landak,
Di hasop pinto-nia dengan api ;
Bagimana mula berkahindak,
Deri mata turun ka hati.
" How is the porcupine to be caught ? smoke his hole with fire.
How is desire first kindled ? from the eyes it descends to the
heart."
A few specimens of the longer and more irregular Seramba of
the people of the interior will be sufficient, and the Serawi dialect
is selected as differing least from the Milay. The following may
be supposed the opening of the contest.
Pandak panjang rantau di Musi,
Maso meniamo rantau Tenang,
Rantau A man pandak sakali ;
Hendak anggan wong ku puji 4
Mimpin bulan sanak bintang
Anak penakan mata hari.
" Long and short are the reaches of the Musi (river), think you
they are the same with the reaches of the Tenang, the shortest of
all the reaches of the Aman ; willing or unwilling I will address
my opponent, I will take the moon by the hand, though she is of
the family of the stars and a daughter of the sun."
It may be answered as follows :
Burong terbang mengulindang
Sangkan terbang pagi pagi,
1 Lit, why is she not fenced in by the stars?
5 Usually in the sense of wise, prudent. 3 Ditentang, lit., meet face to face.
* The chintayu is a fabulous bird, said to delight particularly in rain.
4 Praise or compliment.
F 2
68 A JOURNEY TO GUNONG BENKO.
Hindakkan bunga jeruju ;
Amun wong sintano bulan,
Rinchang sintano matahari
Timbang betating berteraju. 1
" The bird flies swift and straight, it flies early in the morning
in search of the Jeruju flower ; if a person resembles the moon r
and is also compared to the sun, take them up and try them in
scales.
Titiran pikat nibang hari, Ingunan si Jiwo Jiwo, Jadi kampong
burong tiong, jadi koum punei siulan, Bringin di mana garangan
masak, merangei meruntuh daun, sanalah dio maridawan, Amun
sakali kali lagi, Taulah aku di idar' o, Hindak niabong ayam tang-
kap, Hindak berjudi kandong pitis, Hindak siri rai peliman,
Hindak bunga, karang ko tuboh, kundang wong di rindu jangan,
amun asso rindu kan dio, tangisi kian dalam hati.
"The turtle dove kept by Si Jiwo Jiwo calls day by day, the
minas are collected together and the tribe of pigeons ; where the
warringin tree is with ripe fruit, bare and stript of leaves, there they
are all chattering ; Since once more it has come to my turn, if you
wish to fight cocks, take up your bird, if you wish to game, bring
money in your purse, if you wish to eat siri, draw the siri box
towards you, if you wish for flowers, string thyself (i.e., thou art
thyself a flower), if you desire a lover, do not pine for him, if you
do feel a longing towards him, conceal your feelings within your
breast."
As an example of the puzzling questions or figures with which
they sometimes try each other's ingenuity, the following may be
taken :
Ada kayu indan sabatang, Tumbuh di padang raaha leber,
Beringin bukan Beringin, Kruya bukan Kruya, Bodahan ganio 2
ampat dahan, bedaun ganio ampat daun, sadahan chondong ka
langit, niat ka mana bulan bintang, sa dahan chondong ka laut,
niat ka mana raja ikan, sa dahan chondong ka gunong, niat ka
mana gaja indan, sa dahan chondong ka bumi, niat ka mana anak
Adam, Amun teritti 3 sili warang, wong ku angkan dio guru, Amun
de teritti sili-warang, wong ku angkan anak murid.
" There is a great tree, growing on an extensive plain ; it is not
a beringin, neither is it a kruya ; of branches it has only four, of
leaves, too, it has only four ; one branch points to heaven, what
will become of the moon and stars ? one branch points to the sea,
1 Lit., weigh by taking in the hand and scaling.
2 Bgrdahan hanya.
* Form olarti. l\eti i.s found in Straits Malay ; thus tereti'\'->=>tcrarti.
AN ACCOUNT OF THE ISLAND OF BALL 69
what will become of the king of the fishes ? one branch points to
the mountains, what will become of the great elephant? and one
branch points to the ground, what will become of the children of
Adam ? If you understand my riddle, I will take you for my in-
structor ; if you do not understand my riddle, I will take you for
my disciple."
In these examples several words occur which are foreign to the
Malay language; some of these, as wong (orang), indan, sili, &c.,
belong to the Suncla dialect ; and others, as amun (if), peliman,
asso, angkan, &c., are Serawi.
To conclude this paper, the following are the results of a series
of trigonometrical observations made by the late Captain H.
Auber, for determining the distances and height of some of the
more remarkable hills in the neighbourhood ot Bencoolen.
Distance of the Sugar Loaf from Mount Felix, 17-84 miles.
Perpendicular height of the Sugar Loaf, 2601 feet.
Distance of the Laye or Sungey Lamau Hills, 28-37 miles.
Perpendicular height of their highest points, 7,797 feet.
X.
AN ACCOUNT OF THE ISLAND OF BALI.
By R. FRIEDERICH.
["Journal of the Royal Asiatic Society," N.S., vol. viii. pp. 157-218;
vol. ix. pp. 59-120 ; vol. x. pp. 49-97.]
Dr. Friederich's valuable paper was originally published in vols, xxii. and
xxiii. of the " Verhandelingen van het Bataviaasch Genootschap' (1849-50),
under the title " Voorloopig Verslag van het Eiland Bali." Part of it (pp.
1-39 of the present volume) was translated for Dr. Logan, and published in
the "Journal of the Indian Archipelago," vol. iii. pp. 119-137, 235-250,
whence it is here reproduced in a thoroughly revised form. The remaining
.and by far larger portion appears now for the first time in English ; the trans-
lation having been made by Mr. A. II. May, at the suggestion and expense of
Major-General Sir George le Grand Jacob, C.B., K. G.S.I., who is greatly
interested in Balinese literature, and has long been desirous of obtaining a copy
of the Ravi Brahmanda-Purana, according to Dr. Friederich the only Purana
known to the Balinese.
INTRODUCTION.
I MUST request the indulgence of friendly readers for the following
paper on Bali. Not having prepared myself for this labour on Bali
70 AN ACCOUNT OF
itself, I had not the means of collecting and properly arranging all
my materials. I could only use for this purpose a small portion
of the valuable manuscripts of the priests which were placed at
my disposal. I could not avail myself of the information of the
natives as to many points, and I was deprived of a great part
of my manuscripts. These circumstances will, perhaps, in some
degree excuse the many deficiencies, best known to the writer, of
this preliminary account. I have divided this work into three
sections ist, language and literature; 2nd, religion, worship and
cremation ; 3rd, castes and royal races. With this is given a short
description of the Balinese calendar.
In the "Tijdschrift voor Neerland's Indie," IX. vol. iii. p. 340, an
explanation from the Sanskrit is given of the name Bali in the
paper Usana Bali : subsequently the title of a work, Bali Sangraha,
became known to the writer. This work, which however appears
no longer to exist, was presented by a pandita to one of the
princes of Bali. The name is explained thus Baliw^esha^
sangraha = kumpulan. Following the Indian manner of composi-
tion, where the word, which must be taken to be in the oblique
case, is placed before that in the nominative, it is to be explained
thus The gathering of the excellent (the heroes). With this the
Sanskrit sangraha entirely agrees. Bali is then not to be con-
sidered as " offering," but as the nominative of the theme balin r
a strong person, powerful, a hero. The name Bali signifies, thus,
a hero, and the name of the country given in Usana-Bali,
Bali angka, " the lap (birthplace) of heroes," is a very beautiful
denomination of the holy land, and one which expresses the bold
spirit of the nation.
Crawfurd and Raffles first drew attention to the great importance
of Bali in a religious and scientific respect. After their time little
progress was made towards a knowledge of the island, and thus
the Balinese (from their wanting that courtesy which the Javanese
exhibit, which however only shows their submissive character)
have been considered as a rude uncivilized people, from whose
knowledge not much was to be expected. It cannot, indeed,
be said, that the whole population of Bali, in arts (wherein they
clearly are behind) or in science, stand above the Javanese, but the
priests bring before our eye the stage at which the Javanese stood
before the introduction of Mohammedanism. They are, also, the
only remaining preserves of the old literature and religion. To
them must every one repair who desires the elucidation of the
Kavi. They are the expounders of all laws and institutions ; and
of the knowledge of antiquity they have scarcely lost or forgotten
anything from their faithful adherence to traditions.
Should circumstances permit, I hope, after some time, to follow
up this preliminary account by an extended work on this remark-
able island.
THE ISLAND OF BALI.
LANGUAGE AND LITERATURE.
The language of Bali, 1 like that of Java, is divided into a High
and a Low, the first being spoken by the lower to the higher
orders, and the last by the higher to the lower. The High
Language is nearly pure Javanese, but it does not entirely agree
with the present high Javanese. It possesses many words which
now belong to the low tongue of Java, while other high Javanese
words cannot be used in it without giving offence. It is thus
easy for a Javanese to understand the high language of Bali, but
he is not able to speak it with purity. The Low Tongue, on the
other hand, has very little in common with the Javanese, and it
agrees more with the Malayan and Sundanese, so that it is easily
learned by men from Western Java. This language is that of the
original inhabitants of Bali before the arrival of the Javanese.
It has naturally undergone some changes, but, in general, we find
in it a rude Polynesian 2 dialect, which, in the recognized relation-
ship of all these languages, agrees most with the least polished
dialects, the Sundanese and original Mahy ; while it is far behind,
and greatly differs from, the polished language of Java, which, in
the course of more than a thousand years, has been brought to its
refinement. On Bali, four hundred years ago, there were yet
savages or half savages without a finely elaborated language.
The same we may suppose to have been the case with the Malays
before the reception of Muhammedanism, and with the Sundanese
before the kingdom of Pajajaran came into existence. From this
alone, that is, from the original relationship between all the
languages from Sumatra to Bali and further to the east, which has
been only distinctly preserved where the people have remained
in a lower stage of civilization, we may explain the agreement
between the low Balinese tongue and the Sundanese and Malay ;
an immigration of Sundanese or Malay into Bali is not at all to
be thought of. The Javanese conquerors found this language the
prevailing one on Bali, and could not expel it, and, for this reason
in particular, that the population of Bali was very numerous, and
was brought under subjection more by the greater civilization of
the Javanese than by the force of arms. The Javanese conquerors
preserved as a high language the Javanese which they brought
with them ; for their intercourse with the people of the land they
had to learn the original Polynesian tongue, which alone was
spoken by the former, and which, to this day, has a wider
prevalence on Bali than the low language on Java. . It is still
1 [R. van Eck, " Beknopte Handleiding bij de beoefening van de Bali-
neesche taal," p. 1-8; H. N. v. d. Tuuk in " Tijdschrift v. de taalkunde," voL
xxv. p. 245 ; Dr. Brandes, " Vergelijkende Klankleer," p. 108-11.]
2 [Here and in the sequel we should prefer the term "Malayan."]
72 AN ACCOUNT OF
exceedingly difficult for a common man to express himself intel-
ligibly in the high language; and to speak to each rank of a
higher or lower degree with full conformity to the laws of polite-
ness, is an accomplishment which many even of the young princes
have not attained. The agreement between the Balinese and the
Sundanese does not confine itself to words alone. Both have also
only 1 8 letters, while the Javanese possesses 20; these 18 were as
much as the Polynesian organs originally required; the second
</and /are properly foreign to these languages, and the distinct
pronunciation which the Javanese give to them is not easily dis-
criminated by the ear. Notwithstanding, these characters, as well
as the capital letters, exist in the writing of the Balinese, but are
only used to express the corresponding Sanskrit characters / and
d or dh (cerebral), in the same manner as the aksara murda or
gde, the capital letters of Cornets de Groot. 1 Further, the Sunda-
nese and Balinese agree in preserving the pure pronunciation of
the vowel a in all cases where the Javanese corrupt it to o (a).
The a is also in these languages, as in the Sanskrit, of far greater
range and predominance than the other vowels. The only de-
generation is to fepet e, and this may also be considered less as
a short e than a short ejaculated ^, which is commonly used with
a nasal sound following it (;;/ or n and ng).
The language of Java must originally have possessed a closer
relationship to the Balinese. This we conclude principally from
the appearance of Malay, and also (according to Humboldt, vol. i.
p. 198) of Tagala words, in the Kavi. At the period when the
Kavi formed itself, the Javanese language could not yet have been
so refined as it might have been if it had been formed in the
course of ages in civilized Hindu States. 2 The Malay words of the
Kavi, which do not exhibit themselves in the present Javanese,
are ou'ginal Polynesian, and reveal to us the union which once
existed between the languages of Sumatra, Western and Eastern
Java, Bali, and probably all the Eastern islands, and which,
chiefly, in the Eastern or proper Java alone, has been obscured
by a higher civilization. The influence of the polished Javanese
has also, it is true, made itself felt in the Sunda territories, but the
high language of those parts is far less developed than that of
Java ; it probably first began with the establishment of the king-
dom of Pajajaran ; as on Bali with the arrival of the Javanese.
On Bali the division into castes operated most, which rendered
necessary a subordination in the manner of speaking also. By
1 [And of the subsequent authors of Javanese grammars, T. Roorda, J. J.
de Hollander, Jansz, Halkema. The ten letters in question are called capital
because they are substituted, except when final, for their equivalents in writing
names of objects to which honour is due, such as deities, princes, &c.]
2 [The results of the labours of V. der Tuuk, Kern, Brandes, and other
savants tend to modify these conclusions ; see the following notes.]
THE ISLAND OF BALI. 73
the Javanese, however, the language must have been rendered so
complicated, since it was developed by them during more than a
thousand years. A further knowledge of the languages east of
Bali will probably still more confirm this position : the languages
vf all these islands are dialects differing from each other, which
have departed the less from the original parent the less and the later
tfie people have received Hindu civilization. Besides the spoken
languages, we have on Bali the written language ; this is in poems,
with the exception of the more recent, the Kavi, and in the sacred
writings of the priests, the Sanskrit
Humboldt (vol. i. pp. 188-203) has written best on the origin
of the Kavi language. 1 Some modifications, however, in the con-
clusions of Humboldt must be introduced by the fact that pure
Sanskrit writings are still found with the priests on Bali.
Kavi is explained by Humboldt to be "poetical language"
(Kavi "a. poet," kdvya "a poem"). With this explanation that
of the Balinese agrees ; they say that Kavin or Kakavin signifies
** to make comparisons," "to speak in comparisons." This is the
mode in which poetry is formed ; comparisons are the ornaments
and marks of poetry. The explanation of the Javanese by Khavi
(strong) scarcely needs to be mentioned. Khavi is an Arabic
word ; first known in Java in the Muhammedan era, and in Bali
not at all. How could the Arabs have given the name to a
language which they neither produced nor cultivated, but, on the
contrary, have nearly destroyed, because it was the prop of
Hinduism and of all the institutions on Java which the Arabs
sought to overthrow and cast into oblivion ? It is due to the
Arabs and their followers that the Kavi is no longer understood
on Java, and that Kavi works have nearly disappeared there,
while an abundance of them has been carefully preserved on Bali.
The verb kavin or kakavin has caused the works to which that
name is applied by the Balinese to be regarded as marriage
poems, because it reminds us of the Malay kavin (to marry).
Both words, the Balinese and the Malayan, appear to be referable
to the same Sanskrit word. From kavya by the suffixing of the
Polynesian an, kavyan is formed ; this, by the contraction of ya
to e commonly (however improperly) used in Java, gives kaven ;
and from this, by a careless pronunciation with the common
permutation of e and *, are formed kavin and kakavin. This is
then at once the Balinese word for "poetry" and the Malay for
" to marry," because the marriage songs (Jiymenaea] form a prin-
cipal part of the festivity, and that which most strikes the ear.-
Respecting the origin of the Kavi language, it would seem that
1 [See tlie later contributions, ap. Brandes, 1. 1. p. 73 ff.]
2 [It is obvious that the Malay and Javanese word kdwtn, marriage, to
marry, which is a Persian loan word, has nothing but the sound in common
74 AN A CCO UNT OF
some new ideas must be kept in view. The priests did not hold
the Kavi but the Sanskrit as the sacred language ; this language is
still found on Bali in the Vedas, the Brahmandapurana and other
mystic writings or tuturs [that is Sansk. tantra. ED.]. We cannot
therefore agree with Crawfurd, who considered that the Kavi was
the language of the priests (Crawf. "Arch." vol. ii. pp. 17, 18).
The Hindus, and particularly the Hindu-Brahmans who came
to Java, brought with them the Sanskrit in their sacred writings,
and, perhaps, also a Prakrit dialect. That they knew and could
speak a Prakrit dialect may be concluded from the comparatively
late period of their arrival from India, which we place at the
highest 500 years after Christ; at that time, however, the Sanskrit
had been at least 800 years a dead language in India. On the
other hand, against the idea, that they spoke Prakrit, pleads
strongly the fact that we do not find a single Prakrit word in the
Polynesian languages, that none of the assimilations, contractions
and elisions which characterize the Prakrit appear in the Indian
words of the Kavi ; but it is this very fact which points the way to
an explanation of the origin of the Kavi.
In the Sanskrit words on Java and Bali we find corruptions,
which have not originated in an Indian mode. To this class
belong the contraction of va to o, ya to e, the indistinct pronun-
ciation, and the permutation thence arising, of u and 0, of / and e ;
further the permutation of ra and re (kcrret, formerly recognized
by me as ri-vocalis], which however, as well as the preceding cor-
ruptions, never appear in good Balinese manuscripts. To this
class belong also the corruption of the prefix pra into/tfr and
per ; the omission of the initial a in Sanskrit words, for example
jutgraha for anugraha, which are interchanged with the non-signi-
ficant initial letter a of Javanese verbs. The pronunciation of
Anusvara as Jig, e.g. in ong, should not be ascribed to a corrup-
tion ; as this pronunciation appears to stand nearest to the
unsettled sound of the Indian letter. The change of the Indian
v to b in Byasa, Balmiki, Baruna, is to be considered less as a
corruption than as an accommodation of the Sanskrit idiom for
the preservation of the vocalic pronunciation. I, therefore, believe
that the few changes in Sanskrit words have had their origin in
Java, and that not a single Prakrit word has been introduced into
the language of that island.
Thus the Hindu immigrants into Java, though they certainly
spoke the Prakrit, as we must presume if we consider the time of
their arrival, appear to have abandoned that language at once and
adopted the dialect of the country. The reason for this must be
sought in the circumstance of the Hindus arriving but in small
numbers and finding a large population of natives; further, in
with the Old Javanese and Balinese word kawin t which is a poem framed
after a Sanskrit original.]
THE ISLAND OF B A J.I. 75.
their being partly Buddhists, the adherents of which creed always
adopted the manners and language of the nation to be converted,,
in the different countries into which they came. By the Buddhists
the devotees of Brahma were likewise compelled to yield with
regard to language, in order not to irritate the people whom they
wished to subject to their own worship and institutions, and to
give thereby full play to the Buddhists. Thus Buddhists and
Brah mans lived together in Java on peaceful terms, and the
worship of each became not indeed blended with, but augmented
and modified by, the dogmas of the other. We have noticed this
already on an earlier occasion when viewing the ruins of Pram-
banan and Boro Budo; in the course of this account more distinct
proofs will be given of this hypothesis in different places. The
Kavi works are written partly by Sivaites, partly by Buddhists ;
both use the same dialect, and the works of both are held in high
regard by the people, though the Siva Brahmans of Bali appear to
entertain a predilection for the genuine Sivaitish works.
Those friendly relations appear to be one of the chief causes of
the existence of the Kavi language. The introduction of a foreign
language was not practicable on account of the Buddhists, and
because the original population of Java was too large ; still the
necessity was felt of augmenting the dialect of the country in
order to express, in the tracts written for the people, ideas relating
to worship and science, for which no terms were then existing. In
this way the people became accustomed to a number of Sanskrit
words employed by their instructors in religion, and by gradually
introducing more and more foreign words, a distinct language
was formed, destined exclusively for writings and teaching. This
language could not of course adopt the inflexion of the Sanskrit,
for, in order to understand it, the people would have had to be-
inade acquainted with the entire Sanskrit grammar, which would
have been too troublesome for a nation like the Javanese to
acquire, and moreover the imparting of it was not for the
interest of the priests, whose secret writings, containing un-
adulterated Sanskrit forms, remained unintelligible for the rest of
the people.
The fact that the Buddhists formed the Kavi without the intro-
duction of words from the Prakrit, seems also to prove that their
secret writings were in Sanskrit. In Ceylon and the further Indian
Peninsula the books of the Buddhists were composed in Pali (a
dialect of the Prakrit) ; but, in China and Tibet, in Sanskrit ; the
promulgation took place earlier in the northern parts than in those
towards the south, and, for this reason, the books were still written
in the ancient sacred language of all India. If, therefore, the
Buddhists brought their books to Java composed in the Sanskrit
language, their introduction must have been in a comparatively
ancient time. It has been observed already that this newly
76 AN ACCOUNT OF
formed dialect 1 was chiefly intended for the converted people,
while the priests preserved in the Sanskrit the religious books
used by them alone (the Vedas), and whatever they wished to
keep secret from the people (Brahmandapurana and the Tuturs).
The Kavi contained all those works by which the religious
ideas and the cherished mythology of the priests were communi-
cated to the people. It thus became a sacred language to the
people, and the holiness attached itself to all the words, prin-
cipally however to the Sanskrit, which were rendered conspicuous 3
by capital letters (the aksara g'de or murdd). For the priests of
Bali this language is that of pleasure ; they always use it for their
poetical compositions ; almost every one of them composes a poem
of greater or less extent, which is communicated to their colleagues
and scholars. But the Kavi is not sacred to them ; they greatly
distinguish between Kavi and Sloka. Sloka is the usual Epic
measure of India, in which, in Bali, the Mantras (secret writings)
and also the Vedas are written. The name Sanskrita, as signifi-
cative of the language, is unknown in Bali. It is not even of a
very old date in India, having come into use to contradistinguish
it from the Prakrita, the vulgar language, ^loka (the measure
used in the epic poems of India) is used at present in Bali as the
denomination of the works composed in that measure, the language
of which is Sanskrit. Those are sacred and must be kept hidden
from the people (rahasyd). The Kavi has various epochs ; in the
opinion of the Balinese there are three principal ones, viz. :
1. The epoch of Ayer Langgia ; in the compositions of his
age, according to the Siva Brahmans, the Kavi appears in the
most beautiful and oldest form. He reigned in Kediri, and was
one of the ancestors of Jayabaya. In his time the worship of Siva
seems to have been predominant.
2. The epoch of Jayabaya ; of his time is the Barata Yudda,
less esteemed than, for instance, the Vivaha, and indeed of a more
recent style, also many works of Buddhist authors. His period
cannot be ascertained from the Balinese records ; according to
them he reigned in Barata Varsa (India), but this is the India
transferred by the Barata Yudda into Java. His period would
appear to comprise the reigns of several rulers, since so many
works are ascribed to him.
3. The epoch of Majapahit, where we meet with still greater
admixtures of the vulgar language, and less acquaintance with the
riches of the Sanskrit. This period is succeeded by a fourth one,
formed by the continued poetical compositions of the priests and
of some princes in Bali. These, at least the priests, have pre-
served the knowledge of the Kavi, and even augmented it by new
1 [On the nature of Kavi and the position it holds with regard to Javanese,
.see the note to V. d. Tuuk's article " On Malagasy," and the references.]
2 [See, concerning the term "capital," the note above at p. 72.]
THE ISLAND OF BALL 77
Sanskrit expressions, which they take from the secret writings.
From this we are inclined to trace their immigration into Bali, and
the large stock of knowledge they are still in possession of, to
another part of Java, perhaps Kediri, and not to Majapahit. The
tale of Ibiva Brahmans having come to Majapahit from India
shortly before the destruction of that empire is altogether unknown
in Bali. How is it, moreover, possible that those Brahmans
should have acquired so speedily the knowledge of the Kavi and
of the native language ? The priests of Bali have been in Maja-
pahit, how long is uncertain; but they descended from Kediri,
and from thence probably it was that they brought their greater
knowledge. These accounts can be brought into accordance
with the account in question of the arrival of &va Brahmans at
Majapahit, if we here, likewise, bear in mind the transfer of
Baratavarsa into Java, Kediri with its king Jayabaya lay in
Baratavarsa ; Majapahit seems not to have been comprised in it.
The literature of Bali from its nature is divisible into
1. Sanskrit works, with Balinese paraphrase; they include the
Vedas, the Brahmandapurana, and the greatest part of the
Tuturs.
2. Kavi works : (a) the epics sacred to the people, viz., the
RAmayatfa^ Uttarakdnda and the Parvas : (b) the lighter Kavi
poetry, as the Vivaha, Barata Yudda, &c.
3. Javanese-Balinese compositions, written partly in the native
measures (Kidung), such as Malat ; partly in prose, as the his-
torical narratives Kenhangrok, Rangga Lave, Usana, Pamendanga.
Some of the works in prose, especially the law-books, cannot be
classed in the third category ; they exhibit the ancient language
strongly intermixed with Sanskrit, yet they cannot be called Kavi
works, from the absence of measure, and this alone is the charac-
teristic of the Kavi language. From this also the poetical language
is determined.
To the accents which are used in the writings of Bali (vid.
"Tijdschrift" IX. vol. iii. pp. 254-56) must here be added a sign for
the long M differing from the ordinary Suku, and everywhere used in
good manuscripts, where the Sanskrit exhibits the long u. This
long M is called Suku Mud, and according to this, "Tijd.," ib. %
p. 255, 1. 3, is to be corrected; the kerret (ri-vocalis) is called Guung
makerret (Guutig is chakra, makerret, joined to kerret). The long
/, with a small point in the common figure, is called ulu mija. The
Balinese have very indistinct notions respecting long and short
vowels ; however, they, at least the learned priests, use the long i, the
long u, and the tedung as signs for the long a, precisely following
the tradition where they must stand according to the Sanskrit.
The priests are also in possession of a work on the euphonic
laws, called Sroyanchana.
In earlier accounts it has been noticed that in Bali no inscrip-
78 AN ACCOUNT OF
tions on stone or metal are met with, nor any older characters
than the present current writing. This is naturally explained from
the letters only having been introduced since the fall of Majapahit
or a very little before. Although we meet with no modes of
writing of a more ancient date, yet in the new writing all the
richness is preserved which ever was possessed by the Sanskrit
writing in Java. It is only in the Balinese manuscripts that we
find reproduced, with the greatest purity, the numerous signs of
the Sanskrit, which were superfluous and unpronounceable in
Polynesian idioms. Nearly all doubt which may be entertained
of the proper powers of the Sanskrit letters, as they have been
received in Java and Bali, will be removed by the examination of
the writing of such manuscripts as the Vritta Sanjaya, and prin-
cipally of the numerous Sanskrit words occurring there ; any pos-
sible faults will be corrected and excused by those who are con-
versant with the subject, if they consider the many transcriptions
of such manuscripts which are made in Bali, and how easily some
corruptions and inaccuracies might find their way into them
among a small nation, shut out from the source of their civiliza-
tion, and for 400 years dependent on themselves.
SACRED WRITINGS NEWLY DISCOVERED.
The first rank in the Balinese literature, as in that of the Hindus,
is occupied by the Vedas. According to the communications of
the priests, they are not complete in Bali, but only fragments,
although, to judge from appearance, tolerably large ones, of all the
four Indian Vedas viz., i, of the Rig- Veda ; 2, Yajur-Veda (com-
monly inaccurately spelt Yayur Veda) ; 3, Sama- Veda ; 4, Artava-
Veda (a corruption caused merely by the transposition of the r,
easily explained by the mode of writing the Indian-Balinese r
above the line ; the Indian name is Atharva- Veda}. The author
of the Vedas is Bagavan Byasa ( Vyasa in India).
The Vedas contain the formulas of prayer as well for the private
worship of the Panditas, performed in their houses, as lor the
feasts, great offerings and cremations of the people, when the
Panditas mumble them inwardly. They are a mystery to all
except the Brahmans, and the Panditas instruct the younger
Brahmans in them in secret. The metre appears to be the epic
loka, as further illustrated in the Article on Metre, and the lan-
guage a pure Sanskrit. From their being written wholly in Slokas,
we may suppose either that the Vedas were brought into that
metre in ancient times, and in that form introduced into Java and
Bali, or that the knowledge of forming Slokas existed in Java. If
the names of the Vedas were not well known, I should rather
incline to suppose that they never had been in possession of the
genuine Vedas, since in India the metre of the Vedas is guarded
THE ISLAND OF BALI. 79
by ample commentaries, and must be regarded as an integral and
sacred part of those ancient scriptures. The whole of the Brah-
mandapurana has been communicated to me on the condition of
my not making any uninitiated person acquainted with it. In the
same way, I may hope to obtain, also, further information about
the rest of the mystic writings, and about the Vedas themselves.
The Vedas have also been in Java, since the priests of Bali are
of Javanese derivation and had their abode in Kediri and Maja-
pahit. Any direct arrival of Brahmans from India is not known
in Bali, and even the immigrants into Majapahit, shortly after the
destruction of that empire, appear not to have adopted the Vedas
of India but of Java, and it is even doubtful whether they arrived
directly from India, or only from some other part of Java, since
the Panditas know nothing of such an arrival from India, while
they nevertheless trace their genealogy through Kediri to India.
From the tenor of the Brahmandapurana in Bali we may draw
conclusions as to the character of the Vedas. The genuine Indian
pieces in the Vedas, which appear to be written in lokas, are,
probably, accompanied by a Balinese or Kavi comment, which,
after the lapse of some time, became necessary even for the priests,
in order not to lose the true sense of the original texts.
It is an object of the greatest importance to get possession of
the remains of the Vedas in Bali. The religion can only by their
means become thoroughly intelligible ; they further give the
standard for the determination of the state of Hinduism when it
spread to the islands, and, if compared with the antiquities
of India, especially through a more intimate knowledge of the
history of the Vedas in that island, will be of service in
ascertaining the age from which the Indian influence, and the
civilization of Polynesia consequent on it, may be dated. Sdrya-
sevana (worship of the sun) signifies not only the religion of the
priests, but also the book containing those parts of the Vedas
which are used for that worship. I saw the outside of the manu-
script; it contained about eighty lontar-leaves. In respect
of contents the Brahmandapurana come nearest to the Vedas ;
it is also called shortly Brahmancla. We find in India eighteen
Puranas, among which is the Brahmandapurana. These eighteen
are the sacred writings of all the different Indian sects. Six
are especially holy to the votaries of Vishnu, six others to those
of iva, and six keep the mean. The more special sects have
embraced chiefly one Purana, as representing the abstract of
their worship, as the worshippers of Krishna the Bhagavata-
purana. In this way it is easily explained how, in Bali, the
Brahmandapurana only should be in use, and how the Panditas
should not have preserved even the slightest recollection of
the other seventeen Puranas, so little indeed that the names
mentioned by. me were altogether unknown to them. . We
8o AN ACCOUNT OF
find in Bali but one Sivaitic sect, and the adherents of it have
acknowledged the Brahmandapurana, perhaps already in India, as
the only book of instruction. The Puranas are, as we know, the
sacred books of the sectaries, and the priests in India did not
trouble themselves much with the sects and their controversies,
but, adhering to the more purified worship of the Veda, held the
religion of the other people in contempt. Hence it is that the
Puranas in India are, chiefly, in the hands of the people. In Bali,
on the contrary, they are guarded by the priests like the whole of
the holy scriptures, and even hid from the people. In Bali,
everything relating to religion is in the hands of the priests, and
on the great ignorance of the people in all that is necessary
according to the sacred literature for their temporal and celestial
happiness, is founded the unlimited power of the priests, who are
the organs of the Deity for the blindly believing people.
The contents of the Brahmandapurana are : the creation, the
ancestors of the world under the various Manus, the description of
the world according to Indian notions, the history of the ancestors
of old dynasties, besides mythology and mythic chronology; it is
composed by Bagavan Bydsa (the holy Vyasa). He is also known
in India as the author of the Vedas^ of all the Puranas and of the
Mahabharata ; his name signifies [expansion, amplification, in
contradistinction to samdsa, i.e.] composition, and Prof. Lassen is
of opinion that it is a personification of the recension of those holy
writings. (In what period did this take place ?) It is worthy of
remark, however, that in Bali he (as the compiler of the said
works), as well as Valntiki, the author of the Ram ay ana, are
known, since from this we may complete the traditions from
India.
The Brahmandapurana is written in Slokas like the Indian
Puranas ; and it is to be lamented that we cannot get possession
of the Indian Brahmandapurana ; a comparison of both of them
would furnish us with a large amount of revelations on the pro-
gress of the literature, as well as on the relation of the Balinese to
the original Indian worship. The Slokas seldom follow each
other unbroken ; generally, we meet with only a fourth or the half
of a Sloka, followed by an extended paraphrase in the Balinese
language. Under the head of Religion we shall give a few
examples.
EPIC POETRY.
Rdmayana. This is the oldest Indian epos, composed by
Valmiki, who is also in Bali acknowledged as the author of it.
Here, however, it exists as a Javanese elaboration by M'pu Raja
Kusuma, also called Jogisvara, or prince of the penitents, father of
M'pu (Hempu} Tanakung and of another poet M'pu Dharniaja,
composer of the Svaradahana. The language is pure Kavi, with
THE ISLAND OF BALL 81
a peculiarly large number of Sanskrit words. The Indian Ra-
mayana contains seven Kandas, large divisions, again divided into
Sargas, chapters ; in Bali we find no Kandas, but the whole nar-
rative of the first six Kdndas is placed together and divided into
twenty-five Sargas. 1 The yth, the Uttara Kdnda, is no part of the
narrative, but forms a separate work in Bali, the author of which,
however, is accounted to be the same Valmiki. The separation of
this Kanda from the rest of the Ramayana is a proof that it was in-
troduced from India as a different piece, not forming part of the
large work, in favour of which position the contents also speak, the
Uttara Kat.ida giving an account of the history of the family of Rama
after his death. From this we conclude that in India, at the period
when the Ramayana was communicated to the Javanese, the
Uttara Kanda was not yet annexed to this work. We, likewise,
do not find, in the Java-Balinese Ramayana^ the long stories of
the Bala Kanda, the history of Rama as a child, where Vasista,
the priest of the house, tells him tales of the time of old. Those
narratives, partly very beautiful, such as that of the Sagarides and
the descent of the river-goddess Ganga on the earth (vide A. W.
von Schlegel's " Indische Bibliothek"), are episodes not forming
part of the Ramayana ; they have, however, so many charms,
especially for a people like the Javanese and Balinese, who take
every story for truth, that the absence of those tales in the Java-
Balinese Ramayana is surprising. We ascribe their absence to
the same reasons as the separation of the Uttara Kaiida from the
Ramayana ; at the time when the Ramayana found its way into
Java, it was not so voluminous as at present in India, and com-
prised exclusively the history of RAma. As to the Mahdbharata,
it has long since been discovered by European scholars from the
contents, and the form of different parts, that in this work, as it at
present exists, we have before us a conglomerate of Indian myths,
which have been interpolated, partly in recent times. The same
seems to be the case with the Ramayana, though the interpola-
tions are not met with so repeatedly, and are not spread through
the whole work. For a careful critical comparison of the Indian
Ramayana with that of Bali I am at present in want of an edition
of the Indian one. In Java, up to this time, there is only known
a Javanese elaboration of the Kavi composition, the Romo ; this
is far behind the Balinese Kavi work both in language and style,
and is looked upon by the Balinese as a corruption. The Romo
probably was not composed until the Muhammedan era, and pro-
bably when, on the cooling of the religious zeal, the beautiful
ancient literature was still remembered, while the knowledge of
the Kavi was forgotten.
I have borrowed a good manuscript of the Ramayana from the
highest and most learned priest in Badong, the Padanda Made
1 [Kern, in "Bijdragen" for 1883, i. p. I.]
SECOND SERIES. VOL. II. G
82 AN ACCOUNT OF
ALENG KACHENG in Teman Intaran. It contains the Ramayana
complete on 210 lontar-leaves, and is written very fairly, with
great care in the use of uncommon signs, and with attention to
the euphonic laws. Of this manuscript the last leaf with the sig-
nature is wanting, so that it cannot be ascertained how old it is.
For my use the little that was deficient has been transcribed from
the text of another manuscript. This latter was written in the
year (of Sakci] 1693, corresponding to the year of Christ 1771;
and in Bali at Bandharapura (the Sanskrit name of Badong).
Badong signifies as well the small kingdom of that name, as the
residences of the princes of Badong, situated at no great distance
from each other. We may translate Bandharapura, " the town of
union," or "the united palaces of the princes, " pura meaning a
town and a royal palace. The Balinese word badong has also the
same meaning. It is written with alpasastra (small letters), which
makes us think of capital (Kavi and Sanskrit) letters. The usual
Balinese letters may indeed be said to be small ones (alpa\ it
compared with old writings still existing in Java. However, we
find no other letters in Bali than the common recent current
writing, and even the learned priests have lost every recollection
of more ancient letters. Inscriptions on stone (as noticed already)
are not found, and the letters of the Sanskrit shown by me to them
were perfectly unknown to them. We can thus make nothing
more of alpasastra than that the writer humbly acknowledges that
he makes use of the imperfect letters, since the want of greater
knowledge does not permit him to write better and more correctly.
The last words contain an invocation of the Deity, and we find
them with slight variations at the end of several manuscripts ; they
are pure Sanskrit, and correspond to the invocations at the begin-
ning of Sanskrit works : Siddir astu, tatastu, ong Sarasvati namah,
cng fmung Gaiiapataye namah, ong sri Gurubyo namah, " Be this
the accomplishment, be it thus (?) : Ong adoration to Sarasvati,
Ong adoration to Ganapati, Ong to the gurus adoration ! " The
word fmung is not very clear nor Sanskrit. The invocation of
tat-astu (let this be) appears also superfluous ; if we explain it by
tatha astu (may it be), the sense becomes no better. Sarasvati
is the goddess of letters, the consort of BrahmCi. In every
Balinese year she has a feast, where the whole of the manuscripts
are brought forth and consecrated in the temple. Ganapati or
Ganesa, the son of Siva and Parvati, is the god of arts and
cunning, the Indian Mercury. His cunning is invoked in India
as well as on Bali, in order to overcome the obstacles which are
likely to be met with in the composition of an important work.
The gurus are on earth the parents and spiritual teachers ; here,
however, are meant the celestial gurus, the Pitaras, or " spirits of
the departed members of the family," who receive a daily worship.
The Ramayana is divided into twenty-five sargas or chapters. It
begins with the incarnation of the god Vishnu in the family of the
THE ISLAND OF BALL 83
Icing Dasarata of Ayodhya (the present Oude) ; he becomes the
son of Dasarata by his wife Kosalya (Sanskrit Kausalya) ; his half
brothers are Barata by Kekayi and Laksmana by Sumitra. His
teacher is the Muni Vasista, who instructs him above all in the
Danurrcda, " the art of arms/' At an early age, the pious king
Visvamitra, the rajarsi, royal rishi (vide the Rajarsis in Bali, his
successors), when he was recognized as an incarnation of Vishnu,
invoked his aid to deliver his hermitage from the Rakshasas who
had made war against it. This he accomplished, and bent the
bow of Parasu Rama. From this the tale turns to his nuptials
with the fair Sita, and to the intrigues of his stepmother Kekayi,
who forms the design to raise her son to the throne. After that
he voluntarily retires into a hermitage, and subsequently into the
forest of Dandaka, accompanied by Sita and Laksmana. Laks-
mana mutilates the Raksasi Surpanaka who wooed for his love,
and by this excites the hatred of RCivana, the prince of Langka
(Ceylon), and brother to Surpanaka, against Rama and his com-
panions. JRiirana ravishes the beautiful Sita, and Rama seeks for
her in vain. He makes an alliance with the monkey-king Sugriva,
and his son the swift Hanuman. Hanuman discovers the hidden
spot where Sita was concealed, and then begins the war of Rama
and his monkey- warriors against the Raksasas of Langkapura. A
large part of the work is filled with instructive conversations
between the monkey-princes and Rama, and their relations, espe-
cially between Vibisana, the brother of Rdvatia, and the latter.
Finally RCivana is slain by Rama, who with his supernatural
weapon chakra cuts off his ten heads. Sita is purified by Agni
(the god of the fire), and disappears in mother earth. Rama
becomes king of Ayodhya, and retires in old age to the forest
hermitage, where he dies.
The Ramayana and the Paruas are to the Balinese a sort of
pattern for princes. The adat of the princes, and of the second
and third castes, is contained in those works, holy to them, whilst
the Vedas and other secret writings furnish the rules for the Brah-
mans. The princes and the chiefs of Bali are to regulate their
lives in accordance with the Epic writings, and as long as they do
so peace and quietness shall prevail and increase in the country.
In the present time, however, many princes are charged with
indifference to the sacred precepts, and with being, thereby, the
cause of the diminution of the fortune and prosperity of Bali. A
virtuous prince, before undertaking the smallest matter, examines
first the conduct of the old Kshatriyas and demigods, as it is
described in the ancient holy literature. The conduct of those
ancient heroes is ever in the recollection of the princes of to-day,
in order to regulate their actions according to the holy patterns,
wherever they may find themselves.
A king is to have the accomplishments of the eight gods of the
points of the compass viz., Indra, Yama, Surya, Chandra,
U 2
84 AN ACCOUNT OF
Anila, Kuvera, Baruna } Agm (according to Ramayana, lontar-leaf
181).
UttaraMnda. This, as we have seen, is the last (seventh) divi-
sion of the Indian Ramayana. The author is likewise Balmiki
( Vahnlki). Up to this time I have not had access to it ; it is,
however, the history of the brothers of Rama, and contains also
stories altogether unconnected with the family of Rama. A more
recent Kavi work is the Arjuna-rijaya, which borrows its subject
from the Uttarakanda ; of which hereafter. Kanda (compare the
" Kanda" of Raffles, vol. i. p. 373 et seg.), division in India, is used
in Bali like Parva for all sacred writings ; those Kavi works,
however, whose names are Kanda and Parva, are chiefly destined
for the princes and nobles of the second and third caste in Bali,
whilst the works written in Slokas are confined as holy to the
priests and Brahmans. The Ramayana and the Parr as (of the
Mahabharata) have not been long known to the whole people ;
they were a secret of the priests and chiefs, and contain rules for
the latter in their government and for every action during their
temporal life. In every undertaking and in every event, persons
of rank are bound to conduct themselves in accordance with the
precepts contained in those works. Contempt or indifference in
following those sacred writings would bring disaster on princes and
people alike, and the entire happiness of the country is indis-
solubly dependent on the imitation of those holy works.
Parvas (of the Mahabharata'}. The second great Indian epos
is the Mahdbharata, composed by the Muni Vyasa (Bal. Byasa}.
The name of Mahdbharata is not known in Bali, but its eighteen
divisions or Parr as are known. The names of those eighteen are
correct. Six exist entire and two are incomplete. From the
name of Mahdbharata being unknown, it would appear that tJiis
ivork at the time it was brought from India to Java did not bear
this name, nor perhaps any general ?iame at all, but that its divi-
sions were already regarded as sacred writings. In that case, the
name Mahabharata is only applicable to a small part of the whole
work, since the war of the Bharatas, that is, of the Pandavas and
Kurus, occupies not more than 20,000 Slokas, whereas the whole
work contains above 100,000. The rest consists of interpolated
narratives of various descriptions, which, as occasion admits, are
inserted loosely or annexed. How much the Balinese Parvas did
contain of the Indian ones, it is impossible for me to decide,
without being in possession of the Indian Mahdbharata ; the
pieces contained in them stand, however, in high esteem, and are
faithfully copied. They have :
1 Adiparva
2 Virataparva
3 Bismaparva
4 Musalaparva
5 Pnutanikaparva
6 Svarga-Rawanapan'a
and pans of
7 Udyoga Parva and
8 AsramaweuafMva.
THE ISLAND OF BALI. 85
The names of the remaining ten they give as follows :
9 Saba Parva.
10 Aranyaka Parva
1 1 Drona Farva
12 Kama Parva
13 Salya Parva
14 Gada Parva
15 Svatama Parva
1 6 Soptika Parva
17 Stripalapa Parva
1 8 Asvamedayajnya Parva.
Along with them they mention also the Santika-parva, although
they expressly said there existed no more than eighteen Parvas: this
can, therefore, be nothing but another name for one of the above eigh-
teen Parvas} Vyasa, the author, whom we have already mentioned
in speaking of the Brahman dapurana, is the son of Parasara, the
grandson of Sakri, who is the son of Vasista, the domestic priest
in Ayodhya, teacher of Rama, and supposed progenitor of one of
the most distinguished castes of the Brahmans. This family was
nearly extirpated through Sakri, the son of Vasista, being devoured
by one of the Raksasas. Vasista was ready to immolate himself
by the flames, but was prevented on hearing from out of the wornb
of the mother, the cries of his grandson, who afterwards was called
Parasara. He then resolved to spare his life for the education of
the child. Upon this he performed his domestic worship, and
while muttering the Veda a fire broke out, into which all the
Raksasas were drawn down by an irresistible force and destroyed.
This furnished the subject of a painting in the private temple of
the raja Kassiman of Gunong Rata, where we see the holy Vasista
performing his worship in the manner still observed to-day by the
Panditas, and hosts of Raksasas, by the power of his words,
falling into the self-existent fire.
The Balinese maintain that the family of Vasista lived in
Baratavarsa (the eldest holy name of the Brahmanical India,
which, however, comprised only a part of the valley of the Ganges
between Ganga and Jamuna). Vyasa, the writer, is also called
Hempu or Mpu Yogisvara. This is a name of frequent occurrence,
and signifies even the highest divinity Siva. It is, however,
explainable by the fact that a saint or Padanda, who retires from
the world, becomes identified with the Deity, and is himself called
Siva. In a certain sense, the Deity is himself the author of all the
holy scriptures, since he enters into the composer and speaks and
acts by him.
* These are, especially, the works whose deficiency the Brahmans, who spoke
with Crawfurd, regretted. They requested me to communicate them to them,
which I did as far as my pieces extended, with the promise to provide, also,
the large remaining part. The Indian books themselves are of no use to them,
since they do not know the writing. I was thus obliged to dictate them word
by word.
1 [See, on this specification, Weber in his " Indische Studien," vol. ii. pp.
136-9. Also van cier Tuuk, " Notes on the Kawi Language and Literature,"
(1881), p. 7; and Kern, "Over de Oudjavaansche Vertaling van't Mihab-
harnta," (Amsterdam, 1877), pp. 2-4].
86 AN A CCO UNT OF
The Bismaparva contains 102 lontar-leaves. The Adiparva is-
nearly of the same size. The Prastanika-parua, which I saw, con-
tained only sixteen lontar-leaves, but was not complete. The
names are all Indian ones with the exception of Svatamaparva,
which seems to be a corruption of Atvatth&maparva, thus called
after a hero of the Mahabhdrata, a son of Drona. Stripalapa-
farva is called in the Sanskrit only Strlparva ; palapa seems to be
formed in the Polynesian manner from alapa (harangue). 1 The
language of the Pandas is, like that of the Ramayana, pure Kavi^
and more difficult to be understood than the other important Kavi
works. In addition we have a Kapipan'a, containing the history
of Sugriva, Hanuman and their ancestors in the monkey-dynasty.
There exist also the Chantaka or Khdaka-Parva ; this is a sort
of dictionary, where all the synonyms are classed together after
the manner of the Javanese dasanama ; it was compiled by Kavi-
dasi, the follower of Byasa ; it commences with the numerous de-
nominations of the gods, and is for that reason of great importance
for the mythology. It is, however, written in prose, and, like the
Kavipawa, strongly separated from the eighteen holy Parvas. An
Agasti (or Anggasti] Parva came also to my knowledge, in which
the holy Agasti (the star Canopus and leader of Rama in his cam-
paign against the south of India) gives instruction to his son Dre-
dasya ; this work is not to be confounded with the Pawas of the
Mahabhdrata.
To the ancient Indian literature pertain further the books of the
laws, especially that of Manu. The Balinese law-books are, like-
wise, drawn from them, although they are written neither in Slokas
nor in Kavi, and we shall, therefore, speak of them after the Kavi
literature. The original law-book of Manu, Mdnavadharma-sdstra,
is not known in Bali either by that name or by that of Menava
Sastra (as the name is said to be on Bali by Raffles, vol. i. p. 991).
Prabu Manu, however, is mentioned as the founder of the law,,
and the Indian origin of the Balinese law and law-books is thus
certain. The Purvadigama or Siva Sdsana, especially, is said to-
have Manu for its author. ( Vide infra.")
COMMON KAVI LITERATURE.
i. Barata Yudda. With respect to its contents, the Barata
Yudda stands nearest to the Parvas. For a considerable time it
has been regarded as the only version of the Indian Mahabharata
in our islands. But we have now found on Bali the original pieces
of that epos. The Barata Yudda is formed after four of the
Parvas viz., after the Bisma, Drona, Kama, and Salya-Parva ;
the author is Hempu (or M'pu) Sdah, who lived in the time of Sri
Paduka Batdra Jayabaya, Prince of Kediri, and wrote his works.
1 [Or, rather, to be a corruption of praliipa.]
THE ISLAND OF BALL 87
by the order of the latter ; the design of the Prince was to obtain
by the composition of the work a kadigjayan, a subjugation of the
world. In this also an Indian idea is conspicuous ; by the per-
formance of great offerings, by sumptuous works of architecture,
and by works of literature, the prince thus engaged becomes not
only famous, but he also acquires extraordinary power, by which
he is enabled to subject the universe to his will. Such was also
the aim of the great offering of the prince of Lombok (in Sep-
tember, 1846), who, not being recognized by all as the legitimate
chief, sought, by offerings and abundant alms, to prove his royal
right and to strengthen himself for warlike enterprises. The time
at which the manuscript of which I made use was composed is
the year of Saka 1724 (corresponding to the year of Christ 1802).
To judge from its outward appearance, I should have taken it to be
much older; in forty-six years the lontar-leaves have already become
much injured, and it seems to prove what is said, also, of Indian
manuscripts, that they cannot survive a hundred years. This,
probably, is also one of the causes that in Java, in so short a time,
almost the whole of the ancient literature was lost, and that, when
the desire for the old literature was revived, hardly any of the old
manuscripts could be discovered. In Bali, also, we must not look
for very old manuscripts ; however, those which are guarded and
transcribed in the families of the priests may almost be considered
as original, since in these families the knowledge of language and
religion is preserved with the minutest care. Some faults are, of
course, also possible here.
The place where the manuscript was written is Svechchanagara,
also called nagara Sukavati, situated in the kingdom of Gyanyar.
I have noticed above that Badong has also a Sanskrit name (Ban-
danapurd) ; this is the case with all distinguished places in Bali ;
this place has even two nearly accordant Sanskrit names. Suka-
vati is the city abounding in pleasure ; Svechcha-nagara, the city
of well-being. We perceive here, again, how far the Indian
element, and thereby the Indian language, has penetrated into
Bali. However, all those places have also Polynesian names for
the populace the Sanskrit ones are frequently known to the
princes and priests only.
The name Barata Yudda was formerly translated "penance,
combat," and commonly written BrdtCi Yudda ; brata (Ind. vrata)
is penance, and the heroes of the combat acquiring perfection by
penance, the explanation had appearance in its favour. But we
find in the manuscripts of the priests of Bali constantly Barata
Yudda, with the capital b (according to De Groot), corresponding
to the Sanskrit bh t and followed by the fdung (or tarung], repre-
senting the long a ; the name cannot, therefore, be brought into
accordance with brata, which originated in vrata. Barata, as we
find it written, signifies, however, " a descendant of Bharata " (the
88 AN ACCOUNT OF
old Indian ruler of the universe), and we have thus in our work
" the combat of the descendants of Bharata." Now the Kurus and
the Pandavas are descendants of that ruler, and nothing can be
more appropriate than such a title for the work. This explanation
has already been offered by Raffles, but the reasons which render
it irrefutable we first learned from the good Balinese manuscripts.
The conclusion of this work agrees much with that of the Ramci-
yana, and is Sanskrit : Ong sri devyebyo namah, ong fmung Gana-
pataye namah, ong siddir astu, tat-astu hastu, ong dirgayur astu.
" Ong adoration to the happy gods ! Ong adoration to Ganapati !
Ong may the accomplishment be, may that be ! Ong, may there be
long life ! " Devyebyo must be devebyo. What gods, however, are
meant is not clear. Sarasvatl and Ganesa cannot be intended, since
the latter is invoked separately ; tat-astu is here made more forcible
by the addition of another astu ; the word fmung here likewise pre-
cedes Ganapati. Dirgayus, " long life," is a thing for which the
Indians and Balinese, and especially the composers of literary works,
always supplicate the Deity. It is not necessary to draw the
attention of those who are acquainted with Sanskrit to the inflected
Sanskrit forms, and to the proper observance of the difficult
euphonic laws of that language, occurring here and at the conclu-
sion of the Ramayana. In an enumeration of the Kavi works of
a less sacred character, the Barata Yudda is placed at the head,
because the contents are closely connected with the holy Parvas.
It stands, however, in less esteem, and is more recent than some
other Kavi works e.g., the Vivaha. The language is also not a
very pure Kavi, but more intermixed with the ccmmon bhasa.
2. Vivaha. This is known from the Javanese paraphrase of
Gcricke, published in the twentieth volume of the " Transactions of
the Batavian Society." The contents and arrangement of the
narrative in the Kavi Vivaha is the same as in the translated para-
phrase. The language is a very pure and beautiful Kavi ; it is
likewise not composed in the common Javanese song-form, but in
the metres derived from India (to be afterwards described). The
author is M'pu Kanva, not Kanno, as we find in the Javanese
Vivaha, which word has been formed by the usual Javanese corrup-
tion of va into o. Kanva is the name of an Indian Muni or Saint.
Our Kanva, however, we may be sure, was a Javanese, perhaps of
an Indian descent. He, too, lived in Kediri under Ayer Langgia,
the ancestor of Jayabaya.
Hempu S'dah and Hempu Kanva seem to have been adherents
of the Sivaitic sect. We find here few or no traces of Buddhism
in the Barata Yudda and Vivaha.
3. Smara dahana, the burning of Smara (the god of love) : a
well-known Indian myth. The god Siva is interrupted in his
penance by Smara (or Kama), that is to say, he loses the fruits
of his penance by falling in love. Enraged by this, he burns the
THE ISLAND OF BALI. 89
god of love in flames which issue from his body. The god of love
is therefore also called Anangga (the bodiless), because his body
was burnt by sira. This work, too, is of the time of Ayer Langgia,
Prince of Kediri. The author is called M'pu Darmaja, son of
Raja Kusuma, the composer of the Rdmdyana Kavi.
4. Sumdna Santaka comprises part of the Indian Raghuvanta.
Raghu, the ancestor of Rama, begets the Adia ; she is permitted
to chose her consort after the Indian royal custom (Svayamvara,
also so called on Bali). Her husband Devindu died, and she then
gave birth to Dasarata, the father of Rama. This work also is
composed in Kediri or Daha under Ayer Langgia ; the writer is
M'pu Monaguna (the name signifies " whose prominent attribute
is silence, mauna"). The writers of the three latter works bear
Sanskrit names, and belong to the Sivaitic sect ; the names of the
Buddhist writers are in the language of the country, and, in this
circumstance, likewise, the characteristic of that religion is con-
spicuous, which made its way chiefly by yielding to and adopting
the manners of the numerous and widely different countries into
which it was propagated ; whilst Brahmanism, rigidly adhering to
the ancient traditions, and holding in contempt all that is foreign,
is nowhere found beyond India except in Java and Bali, and per-
haps in parts of Sumatra and Celebes.
All the three above-mentioned works are in a peculiarly good
style, and highly esteemed, and this chiefly because they are of
Sivaitic authorship.
5. Bomakdvya ; the song of Boma (or Bhauma)*, " the son of
the earth ; " he is begotten by Vishnu from Pritivi (the earth),
and has, as son of the earth, a demon form and disposition.
He is a Ddnava (that is, like the Grecian Giants and Titans).
He waged war against Indra, the god of (the lower) heaven, and
triumphed over him. (Indra is also overcome by Ravana, the
giant-king of Ceylon, and his power appears everywhere as
secondary, against which the evil spirits are proof.) One of the
higher gods (Vishnu or Siva) must subject his adversaries in order
to restore peace and order on earth. Here it is sang Krisna, the
well-known (eighth) incarnation of Vishnu, who kills Boma, and
delivers Indra from his distress. Boma is killed by being lifted
up from the earth, which constantly re-invigorates him. The
author is M'pu Bradah Boda, that is, " a Bauddha, a Buddhist;"
he wrote in the time of Jayabaya of Kediri. Under that king
Buddhism seems to have found its way for the first time into
Kediri (the largest empire which existed in Java before Ma-
japahit}.
6. Arjuna Vijaya (" the triumph of Arjuna ") is formed after
the Uttarakdnda in like manner as the Bdrata Yudda after the
* Buma Kalantaka, by Raffles ; the name Anraka Sura has not yet come to
my knowledge in Bali. (Raffles, vol. i. p. 388).
90 AN ACCOUNT OF
four above-mentioned Parvas. It contains the combat of
Arjuna with jRavana and his victory. Rdvana is here bound, but
not yet killed, because his time has not yet arrived. He is to be
destroyed by Rama. Whether we are warranted in supposing,
here, an expedition of the Brahman Hindus against the South of
India and Ceylon, previous to that of Rama (who is considered
to be a personification of the subjugation of those regions), further
research must show. The composer is M'pu Tantular Boda t
likewise a Buddhist in Kediri M&feTjayabaya.
7. Suta Soma. The ratu Detia (Danawa, Demon). Pur-
usada had made captive all the kings of Baratavarsa and
conquered the ratu Darma. He is overcome by Suta Soma
and his relative Prabu Maketu. It contains many episodes,
and also the history of Rama. The subject is said to be taken
from the Ketaka Parva (vide supra), although we should not
have expected it from the nature of that work. The author is
the same who composed the Arjuna-Vijaya viz., the Buddhist
Tantular of Kediri.
We have thus compositions from older works in the epoch of
Jayabaya, or at least of the successors of the King of Ayer
Langgia ; it appears that the older Kavi language then began to
be difficult of comprehension, and that the favourite subjects of
literature were, therefore, translated into a more comprehensible
language. The influence of the Buddhists in this innovation is
not to be mistaken.
8. Harivangsa. This likewise is an Indian poem, commonly
joined to the Mahdbharata (the Indian one is translated by
Langlois in Paris and obtainable in Calcutta) ; this piece, too,
invites us to a comparison between India andfava, the Kavi and
the Sanskrit. The contents, according to the priests, are : the
conduct of Krisna towards Rukmini (his wife), and the war
against the two princes Jarasanda, father-in-law oi : . Kama, ruler of
Magada, and Chedi or Sisupala. This work was written in Maja-
pahit* and is thus of later origin than the preceding ; the author is
M'pu Penulu Boda, likewise a Buddhist. The King of Maja-
pahit at this period was Brayang V basing Suka, father of Bra
Vijaya (Brovijoyo), who, according to Javanese records, was the
last (Hindu) prince of Majapahit.
These are the most important works of the Kavi literature, so-
far as I am yet acquainted with it. 1 With these, however, we are
far from having exhausted Balinese literature. We have besides
* Majapahit is the literal translation of the Sanskrit Vilvatikta (corrupted
Vilatikta, Us. Java), the bitter vilva (aegle marmelos) ; this then at least is
not a fictitious fruit, and the name of Majapahit not unmeaning, as it was
formerly considered (vide Raffles).
1 [See also R. van Eck in tlie Introduction (pp. vi.-viii.) to his edition of
Megantaka % in the Batavian " Verhandelingcn," vol. xxxviii.J
THE ISLAND OF BALL 91
them, first, the law-books written in prose; further, the Tuturs, or
" instructive writings," of which nothing can as yet be ascertained,
since they are for the most part secret writings. Further, the
Babads, or historic-genealogical works, partly written in Kidung
i.e. the newer (Javanese) measure, partly in prose. Moreover, we
have pure Polynesian myths ; above all, those of Panji, which
are likewise written in Kidung. Then there also exist little essays
on the transmigration of the soul, on erotic subjects, &c. ; and
finally there is the Balinese " Kalendar," a work of the utmost
importance.
BABAD, OR HISTORICAL ESSAYS.*
1. Ke?ihanrok. He is a son of Brahma and progenitor of the
rulers Q{ Kediri, Majapahit and Bali. It has not as yet been
ascertained in what epoch he must be sought for. His residence
was in the Kampong M'dok, whose situation is not known in
Bali, but is supposed to be in Baratavarsa. It is written in
prose, and contains forty or more lontar-leaves. I am only in
possession of the first part, which has no more than seventeen
leaves. His mother is called Kenhendok. The god Brahma met
her, much in the same way as the Greek Zeus knew how to win
his numerous loves, whilst she, as a married woman, was amusing
herself in the field.
2. Rangga Lawe. Siva Budda (N. B.), ruler of Tumapel, is
made captive for misgovernment by the King of Daha or Kediri,
and his empire Tumapel is overthrown. The chief minister of
Kediri is Rangga Lawe ; he at a later time disagrees with his
sovereign, and is finally vanquished and put to death. The work
contains a minute description of the Court of Kediri and the
position of the grandees of the empire, and may serve as a pattern
of the constitution of the old empire in Java. It is, especially r
maintained on Bali that the Court of Majapahit was altogether in
the same style, and that all the rules of the Court of Kediri were
carried to Majapahit. For this reason it would be desirable to-
have this work published (text and translation), accompanied by
the necessary notes ; this, however, can only be usefully done in
Bali. The manuscript in my possession contains sixty-seven
lontar-leaves, each of four lines front and back, and is written
very neatly. It was written in Garogor (Glogor) in Badong^ on
the day of Saneschara Kaliwon Landep, in the month Kasa, the
thirteenth day of the increasing moon, in rah 9, tcnggck 6,
corresponding with the year of Christ 1847, Saturday, the 26th
* The name Babad is also met with in Java (vide Raffles, " Literature," vol. L
P- 393) an d it a l so comprises, following him, all the historic works and new
chronicles. Raffles spells it Babat. In Bali I find the word written Babad*.
[See also J. J. Meinsma, " Babad Tanah Djawi," vol. ii. pp. 1-15.]
9 2 AN A CCO UNT OF
of June. It commences with a metre of fifty-two syllables in each
line, the stanza as usual of four lines.
3. Usana Java. " The ancient institutions of Java," a work
containing the subjugation of Bali by the Javanese of Majapahil
and the settling of the Deva Agung in Gclgel, with the distribu-
tion of the lands amongst the grandees of the Court. One
manuscript of it had twenty-nine lontar-leaves, and was derived,
as they told me, from Pasuruang ; it, however, probably, came
from Bali to this place, and seems to be little or not at all known
in Java. It is written in prose. In that work a predilection for
Arya Damar and his family is plainly manifested, whilst it passes
over the Patih Gaja Madda, the founder of Mengui and ancestoi
of the powerful family of Karang-Asem, almost in silence. For
this we may find reason in the circumstance that it was originally
composed by a follower of the dynasty of Arya Damar. Accord-
ing to the postscript it was written in Galogor by Pam'chuttan (in
Badong), on the day Rediti Pahing (Sunday), in the week Dun-
ulan, in the month of Kanam (the 6th), on the thirteenth day of the.
dark half, in the year i (rah), of the tenggek* 6. This would be
the year 5 1 ; if we take the eighteenth century, we should have
1751 of Sa&a, corresponding to the year 1839 of the Christian era.
4. Usana Bali. The contents of it are known from the " Tijd-
.schrift voor Nederlandsch Indie/' Qth year, vol. iii. pp. 245-373.
There I have said that it is a work exclusively intended for the
people, and not esteemed by the priests. It is otherwise with
the Usana Java , which is held in honour by all castes, at least in
Badong.
5. Pamendanga.\ A sort of chronicle of more recent times. It
-contains sundry confused histories of priests and kings, of the
distribution of Bali amongst the original Pungavas of Gclgel, and
genealogies of kings, of Karang-Asem, for instance. Respecting
the division of the vice-regencies among the Pungavas, this work
widely differs from the Usana Java, and its value and style are far
inferior to those of the latter. It is also written in prose. Other
Babads are found in the family of every prince ; if it were possible
to gather the greater part of them from the different States, they
certainly would spread much light on the history of Bali, if care-
fully compared with each other.
TUTURS, OR DOCTRINAL WRITINGS.
These are divided into two classes : the secret writings of the
priests, and such as are also current among the other castes, espe-
* Tenggek is a period of ten years. Rat, a single year of that time. Sup-
posing the era to be known, we find from it the year of SaAa.
t From the Pamendanga a play is derived, performed by a single person in
.topengs (masks) ; (it represents the more ancient history of Bali viz., of the
Deva Agungs.
THE ISLAND OF BALL 93
cially the second and third. The former are extremely numerous,
but since they are kept secret, we can only mention the names of a
limited number of them. They seem to be written, like the Vedas,
in &lokas. The names I obtained are the following :
1. Buvana Sangksepa (the shortening or contraction of the
world or of men).
2. Buvana Kosa (the treasure of the world).
3. Vrihaspati Tatva (the Tatva, truth, the essence of Vnhas-
petti, the star Jupiter, teacher of the gods).
4. Sarasa Muschaya (sarasa is explained by isi, the contents ;
it is, however, probably sdrasa, the lotus ; muschaya is not very
clear, but is explained by kumpulan, " accumulation," " gathering ;"
this is one of the works enumerated by Crawfurd. 1
5. Tatva Jnana (knowledge of substance, essentia).
6. Kandampat.
7. Sajotkranti.
8. Tutur Kamoksa (vide infra). Under this denomination exist
numerous works ; it means, " instruction for blessedness, or for
delivery from the transmigration of the soul."
The second class of the Tuturs, current, also, among the other
castes besides the Brahmans, are, for instance :
1. Rajaniti (wisdom of kings) ; it contains rules for the policy
and the government of kings, and it is in many respects analogous
to Machiavelli's "Princeps."
2. JNitipraya or Nitisastra (superabundance, or manual of
wisdom) : it is of a similar character with the former.
3. Kamendaka Niti (rules of wisdom of the sage Kamendakd}.
4. Naranatya (nara "men," natya " the mimic").
5. Ranayajna (the sacrifice of the battle).
6. Titi dasa gunita ; this belongs properly to the first division,
but has been made by Padanda Vahu Ravuh into Kavi under
the name of Nitisara (compendium of wisdom).
LAW-BOOKS.
These are written in prose. They comprise most of the Balinese
books which are mentioned by Crawfurd and Raffles. The
accounts of them, however, differ from each other. Raja Kasiman
names them :
i. Agama. 2. Adigama. 3. Devagama (somewhat difficult to
understand).
4. Sarasamuchchaya (the same we have just met with among
the secret writings).
5. Dustakalabaya (the fear of the malignant Kala), a law-book,
in which in particular the faults committed by children are pun-
ished.
1 [It should be s&ra-samuchchaya t the aggregate of truth.]
34 AN A CCO UNT OF
6. Svarajambu (the voice of Jambu), that is, "the command,
the law of India," fambu-Dvipa.
7. Devadaiida (in very old language), it conies in use when
Vishnu appears incarnated upon earth. 1
8. Yajfiasadma (yaj/la "sacrifice" sadmaT)
The Pandita in Taman Intaram mentions only
i. Agama* 2. Adigama, the two law-books mentioned by
Raffles as the basis of the law for the common people. Raffles
calls the latter Degama.
3. Piirvadigama or Sivasasana, the above Adigama, or "the
command of Siva, 1 ' of value exclusively for the Brahmans.
4. Devagama, the dgama of the Devas.
5. Svajambu Svarajambu; the meaning is doubtful; perhaps
svara, "voice," "command," and/aw/^, in lieu Qi Jambu Dwipa
(I?idia), thus, " the voice of the law of India."
The principal law-book from India (ap. Raffles' Menava Sastra,
Ind. Manava-dharma-sastrd] is wanting, according to all inquiries
for it which I made among several priests and persons of rank.
They, however, are aware that all their laws have been derived
from Prabu Manu (the ruler of Manu), who, in different ages,
under different names, holds the government of the world. I
found it mentioned only in the Sivasasana, the law-book of the
Brahmans, under the name Dharma-sastra Kutara- Ala navadi ;
adi has here, it would seem, the true Indian sense "and so forth"
so that the translation will be " the law-books, that of Kutara
Manava and the others" Kutara t is also mentioned by Raffles
as "a law-book," and is not explained by Humboldt. Kutara
appears to me to be the same as Uttama viz., the name of the
third in the line QlManus. The conversion of Uttama into Kutara
is quite possible, and supported by a passage of the BraJunanda-
pur&na: Utara Manu, lont. n. Uttara is the comparative, "the
higher" and Uttama the superlative, "the highest" degree. The
1 [A Dutch translation of this short code (35 sections) appeared in the
Batavian "Tijdschrift," vol. xviii. pp. 295-309.]
* Agama is explained by Wilson, Sanskrit Diet., "a Sastra or work on.
science and of divine origin." In the Malayan and common Balinese language
signifies religion ; in the names Agama, Adigama, Dcvagama, it has evidently
more the old Indian meaning, and especially that of law-book. Adigama seems
to have originated in Adhi and Agama, with the omission of the first A of
Agama, a carelessness which is frequently met with among the Sanskrit words
in the Kavi e.g., Svatamaparva for Aivatth&maparua. The a in Polynesian
words is a euphonic prefix, which was then omitted in the Sanskrit words
likewise.
t Kutara is, following Wilson, " the post round which the string passes
that works the churning-stick." This explanation is here in no way applicable.
[The whole question as to the existence, on Bali, of a Manava-sastra and a
Kutara-sastra has been fully treated by Dr. J. C. G. Jonker in his work,
"E'jn Oud-Javaansch \Vetboek" (Leiden, 1885), pp. 11-20].
THE ISLAND OF BALL 95
k before Utara I am inclined to regard as the Polynesian prefix,
added through ignorance. Opposed to this conjecture, it is true,
is the fact that the law of India must have been framed by the
first Manu, Svayambhuva Manu ; but we have various law-books,
and these are even yet not all known. Possibly the original
Balinese law-book has been derived from another Indian one,
although the contents are upon the whole the same as in that of
Svayambhuva.
This Dharmasastra Kutara Manava is either now in Bali and
kept secret, or it is one of the works which existed in Java, but
were lost and were not brought to Bali. It is mentioned along
with the Sarasamuchchaya, which we learn to be one of the
Tuturs ; further, along with the Kamandaka, a Tutur for obtain-
ing advantage or intrepidity. A learned Brahman is expected to
be acquainted with all these works. It was not without the
greatest difficulty that I got the Sivasasana into my hands ; how-
ever, I may hope to obtain in the like way insight into the
remaining law-books and the Tuturs. The Sivasasana was bor-
rowed by me on the same condition as the Brahmandapurana
viz., not to show it to any one of the people. The manuscript of
the Sivatasana in question was written on the day Mahulu Pahing
Anggara (Tuesday), of the week Sungsang, in the year of Saka
(Sakawarsa) 1682 (A.D. 1760), in the month Sravana, on the
eighth day of the white half, in Vilatikta* After this the writer
makes his excuses in the customary manner for the bad and
careless writing, and he has great need to do so, for the manu-
script abounds in faults ; he pleads his inexperience (muda)
and inferiority (hina dtna). He adds further that the work is
a secret writing (rahasya\ and concludes with the well-known
invocation :
Siddir astu, tat' astu astu
Ong Saras watie namah
Ong g'mung Ganapataye namah
Ong sri Gurubyo namah
Ong ong Kamadevaya namah
Respecting these invocations, we refer to what is said under
Rainctyana and Barata Yudda. Here only is added " Ong
adoration to Kamadeva!" He, the god of love, would thus
appear to be peculiarly the favourite deity of the writer. The
god of love is indeed highly honoured and praised in many of
the newer poems, a circumstance the analogy of which we find
* Where this Vilatikta is to be sought for in Bali remains uncertain ; it is
(Vilvatikta) the Sanskrit name for Majapahit. It is possible that the work
was originally written in Majapahit, and that the copyists in succession retained
the name of the city where it originally was composed, although they them-
selves were in Bali.
96 AN A CCO UNT OF
again in India. We give here the prologue, the text and the
translation of the Sivasasana :
"This is the Purvadigama &asana-sastra-saro-dretta* first
composed by the accomplished old teacher, the raja Purohita,
who knows all qualities, who resembles the rays of the sun, who
dwells in the hearts of all mankind ; Misraharana, who, as the
highest precious stone, outshines all the divine teachers of iv&
(of the Sivaite sect), the lowest, the middle, and the highest;
further is he named the first Guru, the great saint. The same
asked for ashes, after he had obtained permission to ask for ashes
of the children and grandchildren of Sa?2g J3as?na?igkura(?) ; the
same commanded him thereupon to compose the Scisanadigama
Sastrasarodreta for all priests, as many as hold the religion of
o/rtf ; for the Panditas of Siva as well as who live in the cities,
the perfect ones, as also those who choose to dwell partly in
cities, partly in the country, also for the host of the learned, who
take care of processes, who settle disputes between all men, at
the Court and in the country, this is their number. Assuredly the
Adigarnasastra sarodreta must contain the laws for the conduct
of them all." 1
There further exists in Bali a law-book, called Svara, issuing
from the Deva Agung, and in force for all princes and persons
of rank. It cannot, as yet, be ascertained whether it is the same
work as the Svarajambu (or Sva/anibu), but it seems to be a
different one, since the addition of Jambu in the latter points to
its Indian origin. 2
Tatwa or Tutur kamoksa (vide above) contains rules for a
religious life, with special directions from the birth of a man up
to his death. It frequently prescribes fasting (Ind. vrata, brata,
votum). In accordance with those writings not only the Padanda?
regulate their lives, but also the princes and those of rank who
aspire to the condition of holiness; they attain, thereby, the
* This word must be divided, it would appear, into two parts ; Purvadigama
sasana, " the command, laiu of the Puruadigama" and Sdstra Saro dreta, "in
which is contained the essence of holy works." The Saro is inserted instead of
Sara, and we thus find the nominative case in place of the theme in a composi-
tion. This seems to be an error founded very likely in the want of acquaint-
ance with the meaning of the Sanskrit terminations and inflexions, but offering
at the same time another proof of the preservation of the inflexions in the
memory of the Panditas. \Saro-dretta = saroddhrita, "gathered from the
essence of the S.-L"'\ The SivaSdsana or Piirvddiganiatasana is the law-book
for all the Brahmans, in the cities as well as in the country, and for those in
whose hands the jurisdiction is deposited as well as for the rest. It is not,,
however, applicable in the decision of the lawsuits of persons belonging to
one of the three lower castes.
1 [The transliterated Kavi text, with Friederich's explanatory notes, is here,
for obvious reasons, omitted.]
2 [See P. L. van Bloemen Waanders in the Batavian " Tijdschrift," vol. viii.
(1859), pp. 201-27, and the Introduction to Jonker's work previously men-
tioned.]
THE ISLAND OF BALI. 9;
dignity of Rest (a saint, without sin), and the priests become
Brahmarsi, the princes Rajarsi ; the latter, of course, as it is
natural, in consequence of their birth, rank below the former.
Every prince must properly aim at this dignity, and the Abisckn,
"the anointing" of the chief prince, is dependent upon it. Uy
becoming Resi and by the Abiscka^ not only the dignity of the
prince is raised, but he is, thereby, as it were, received into the
caste of the Brahmans the like rule is also observed in India.
The predecessor of the last sovereign of Pam'chuttan was Resi^
and had received the Abiseka ; even as the former Deva Agun&s.
At present there is no prince of Bali who has received the
Abiseka. The Raja Kassiman, however, aims at the dignity
of Resi.
MALAT.
The Malat * contains the history of the celebrated hero Panji,
who had his adventures on Bali also. The work is as voluminous
as the Ramayana ; it is, however, not written in the Kavi measure
or language, but in Kidung, which means the newer Java-Balinese
measure. The subjects contained in it are exhibited to the public
in the Gambuh (dramatic performances by men, who speak them-
selves). The same is the case with the Ramayana. The Barata
Yudda and Vivaha are represented in the Vayang Kulit in the
same manner as on Java. Of the Indian drama nothing seems to
have found its way into this island. The names of the most
famous of the Indian dramas are unknown there. The tale of the
Sakuntala is known from one of the Parvas, and the original nar-
ration we find also in India in the Mahabhdrata. But the
magnificent drama Sakuntala of Kalidasa is not known. The
reason of this is, probably, that most of the Indian dramas are of
late times, and, perhaps, at the time the Brahmans came to Java,
were exclusively found at the courts of the princes of Ujjayinl,
Kashmir, Ayodhya, &:c., so that the Brahmans could not be
acquainted with them. Besides, the drama forms no part of the
sacred literature, and the Brahmans might have neglected it for
that reason.
RELIGION.
The religion of Bali, as is well known, is the Hindu, and in fact
the two great Indian creeds, the Brahmanical and the Buddhist,
exist there. The adherents of the latter are few in number, and
live in Karang-Assem in the dessa of Buddha Kling (Crawfurd)
and in Gyanyar, in Batua?i. These Buddhists, whom no European
has ever visited, appear, however, to hold a modified form of
religion. Crawfurd remarks that the people of Boleleng had
1 [R. van Eck, in "Bijdragen," III. vol. ii. pp. 3-5.]
SECOND SERIES, VOL. II. H
98 AN ACCOUNT OF
spoken rather contemptuously of the Buddhists, but I have not
noticed this in the southern part of Bali. It is true they are said
to be allowed to eat all kinds of animals, cows for example, which
the worshippers of Siva are forbidden to eat, and dogs and other
unclean things, but they are not accused of actually eating them.
As for the relations between Sivaism and Buddhism, the Panditas
state that Buddha is Siva's youngest brother, and that the two
sects exist peacefully sjde by side, although the Buddhists do not
worship Siva, and the Sivaites do not adore Buddha. In the form
of worship, however, an intermixture of the two religions is ap-
parent, for, on great feasts, e.g., the Panchdvalikrama, a priest of
Buddha is invited to join the four Panditas of Siva, and performs
his devotions sitting towards the south, while the other four
throughout the service sit towards the remaining cardinal points,
and in the middle. At the cremation of princes, moreover, the
holy water from a Sivaitic Pandita is mingled with that of a
Pandita of Buddha, and is used in this form by the worshippers
of Siva. The intermixture of the two religions is also shown by
the frequent mention of Buddha in the Kavi writings, and by the
Buddhist composers of these writings, these works being also held
in honour by the Sivaites. This, however, applies more to Java,
whence all those writings came, but it is partly applicable to Bali
also. So much is certain, that the Buddhists in Bali (and in
earlier times in Java) were not fanatics, and that they left the
Hindu Pantheon undisturbed, whilst they worshipped Buddha as
the only true God.
SIVAITES.
The great majority of the Balinese hold the Brahmanical belief,
and belong to the sect of Siva. There is no^trace of the other
sects (Vishnuites) in Bali, and the worship of Siva has absorbed,
as it were, that of all other gods of the Hindu Pantheon. The
religion may be divided into the private worship of the priests and
\\~\z. public worship of the people.
THE DOMESTIC WORSHIP OF THE PRIESTS.
The domestic rites of the Panditas remind us of the ancient
Veda-worship of the Indian Brahmans, and in fact owes its origin
to it. In old times the Brahmans in / India did not worship the
gods of the people ; Brahma, Vishnu or Siva, and all the rest of the
gods connected with them, had no existence for those men they
adored the celestial bodies, especially the sun, and yfo? (Agni] and
various stars. The domestic worship of the Brahmans in Bali has
also the sun for its object, and is called suryasevana (worship of
the sun) ; it is performed without temples or idols and with but
few offerings. Upon asking what the sun meant, I was told that
it was Siva, and therefore we may presume that the Brahmans no
THE ISLAND OF BALL 99
longer hold the ancient faith, and have adopted the ordinary
service of Siva. Siva, however, has become so idealized, at any
rate by the Brahmans, that he may very well be identified with
the supreme (solar) deity, and in the popular creed of India Siva
is also the representative of fire, and bears the sun as the third
eye in his forehead. We, therefore, adhere to the hypothesis,
that the Brahumns in Bali have preserved the ancient worship of
the Indian Brahmans, which is based on the Vedas alone, and takes
but little or 110 notice of the gods of the people, and that, although
they conduct and regulate the worship of the popular gods, they
do not themselves take part therein.
I have been permitted to see the domestic devotions of a
Padanda. They are performed between nine and eleven o'clock
in the morning, on a fasting stomach, and are obligatory at least at
full and new moons, in addition to which most Panditas perform
them on every fifth day (Kalivon, according to the Polynesian
week of five days). Especially holy priests, and those of high
rank, such as the Padanda Made A ting Kachcng in Taman
Intaran, observe them daily. On ordinary days, however, the
service is not so long as on Kalivon, and on this day again it is
shorter than at full and new moons. On the latter occasions,
too, the priest is arrayed in his full vestments. The place of
worship is a Bale, in one of the priest's inner courts. The portion
of the Bale* where the ceremony takes place is surrounded on
three sides with a lattice-work of bamboo : that of my Padanda
was only open to the west. The Padanda is clothed in white,
with the upper part of the body naked, after the Balinese-Indian
manner. He sits with his face to the east, and has before him a
board upon which stand several small vessels containing water
and flowers, some grains of rice, a pan with fire, and a bell. He
then mumbles, almost inaudibly, some words or prayers from the
Vedas, dipping the flowers into the water and waving them and a
few grains of rice before him (towards the east) with the forefinger
and thumb of his right hand, whilst at the same time he holds up
the pan containing fire. After having proceeded with his prayers
for some time, during which he makes all kinds of motions with
his fingers and turns his rosary, he appears to be inspired by the
deity ; Siva has, as it were, entered into him ; this manifests itself
in convulsions of the body, which grow more and more severe,
and then gradually cease. The deity having thus entered into
him, he no longer sprinkles the water and flowers towards the
east alone, but also towards his own body, in order to pay homage
to the deity which has passed into it. The bells are not used in
the ordinary daily worship, but only at the full and new moons
and cremations.
* The names of the Bale are: Yasa, Mahantcn, Mahari, Boat; the holy
water is called Sevamba^ i.e., sdiva and amdAas, " water of Siva."
H 2
ioo AN ACCOUNT OF
By this ceremony the Padanda is completely purified ; all his
actions, even the partaking of earthly food, are holy. He
then eats (but only once in the day) ; while he is doing so no
one but his children, who wait upon him, may approach, and
they keep silence. The remains of his food are like Amrita
Ambrosia), and are eagerly solicited and consumed by those
present -including the princes, if they have a Pandita in their
house, or happen to be in his house. The water which the
Pandita has used during his devotions is looked upon, in accord-
ance with the Vedas, as holy ; it is called toya tlrta (water of a
holy place), and is bought by the people for their purifications,
for sprinkling corpses and for offerings. This is one of the
sources of income to the Panditas. In addition to his domestic
worship, he performs the public religious ceremonies (see below),
and conducts the cremations and the offerings for the departed.
In his own house, moreover, he occupies himself with the Vedas,
with the sacred and the common Kavi literature ; he teaches his
children and those (chiefly princes and men of rank) who come to
him as pupils. He is also the people's astronomer and astrologer,
and alone knows how to regulate the calculation of time according
to the different divisions of the year (see the Balinese Calendar).
Finally, he consecrates the weapons. Every new weapon to be
wrought is brought to him before the operation : he places some
mysterious signs upon it, especially the word Ong (om), and until
this has been done, the weapon is of no value or power. When
the weapon is quite finished, the owner makes offerings and the
Pandita reads the Vedas over it to insure its effectiveness.
RELIGION OF THE PEOPLE. PLACES OF WORSHIP.
The chief places of worship are the sad-kahyangan (the six
temples) so-called /car' e^o^f. They are all dedicated, under
various names, to Siva. The principal and oldest temple, the
founding of which is narrated in the Usana Bali, is (i) in Basuki,
at the foot of the Gunung-Agung, the holy mountain in Karang-
Assem ; the name of the deity is sang Purnajaya, and his weapon
tuak (a sword-like creese).
(2). Vatu Kahu, in Tabanan, at the foot of the peak of Tabanan,
called Barattan or Vatu Kahu ; the name of the deity is sang
Jayaningrat, the weapon panah (bow).
(3). Uluvatu, on the point of the table-land (bukii) in Badong,
picturesquely situated above the sea, over which the rock on which
the temple stands projects. This temple is the prahu of Devi
Dann. The deity worshipped here is sang Manik Kumavang (the
brilliant precious-stone) ; his weapon is tumbak (lance). 'I 'he
access is difficult, through rocks and wild places. The temple can
only be approached with the Sovereign.
THE ISLAND OF BALL 101
(4). Yehjentk (Jcruk- water), in Gyanyar, in the interior, in the
Kampong of Narangkana ; the deity is sang Putra Jnya (the
prince of victory) ; the weapon pedang (sword).
(5). Gira/ara, in K/ongkong, on the coast ; the deity is Sanging
Java (the triumphant one) ; the weapon is sambuk (whip).
(6). Pakcndungan, in Tabanan^ on the coast. The name is sang
Manik Kaleba (kalcba = dumila/i, brilliant) ; the weapon dining
(sword-like creese).
These are the principal temples, in which the rulers make offer-
ings for the whole people. In U/nratu, the feast-day is the
twenty-first of the Balincse year ; in Basuki, on the full moon of
the month Kapat or Karttika (in September or October). A few
other places, although of less consequence than those already men-
tioned, are of special importance.
(i). Sakcnnan, on the island Serangan, belonging to Badong.
The deity worshipped is sang hyang Indra ; his weapon is the
J>njra (Sanskrit vajra\ which really means lightning, but accord-
ing to the drawing corresponds to the so-called thunderbolt. His
feast is on the eleventh day after the Balinese new year.
(2). Jcmpul, in Bangli, also with Indra as its deity.
(3). Rambot Savi, i&jeinbrana, near the frontier of Tabanan.
(4). Samantiga ; and (5). Kcntel Guii, both in Gyanyar. It
is not known what deities are worshipped in the last three. These
places are sacred through the supernatural power which issues from
the gods adored there.
We have besides in each dessa one or more Panatarans (natar,
a court ; the offerings to DurgA, KAla and the Bhutas, are placed
upon the ground, in the court). In these Durga, the wife of the
malignant Kala, and the chief of the Bh&tas or Rakshasas (evil
spirits), is worshipped. The worship of these latter occupies the
people almost more than that of the beneficent gods, for the perni-
cious influence of these beings must be guarded against in all sorts
of ways, whereas the beneficent gods are more easily propitiated.
Other temples are called Puri and Pangastanan ; the former are,
principally, for persons of the highest rank, and the latter for the
people ; here Siva is worshipped with his family. Another name
is ParAryangan^ an assemblage of temples for all the gods and
Pitaras (the shades of the dead). The small temple -houses are
called Kahyangaii) place for a deity (flyang). To these belong
also the Sadkahyangan. Finally, there is in every house a number
of small temples, called Sanggar (in Crawfurd, Sangga). Among
these there is Mcru t a temple with several roofs one above the
other, rising up in the form of a pyramid, dedicated to Siva. The
rest of the small temples are mostly devoted to the service of the
Pitaras. The house-temples of the princes are of some import-
ance and costliness, but they are not built in the best taste.
Among these, besides the Merits, which are of wood, we find also
102 AN A CCO UNT OF
the pyramidal erections of stone. Padmasana (the Padmasana
must be dedicated to the sun ; Siva is the sun), the apex of which
is truncated and provided with a sort of stool, upon which incense
is burnt to Siva, in his three forms viz., Sadd-Siva, Para ma-Sim
and Maha-Siva (the incense being of three sorts : M'nyan, Madya-
gawu and Chandana] ; and Chandi, a complete pyramid, not
truncated. Besides these buildings, one finds several Bales t partly
of masonry (Gdong Chanti-l) inlaid with Chinese porcelain and
glass-work and ornamented at the back with pictures, and partly
of wood (Gdong Tarik\ upon which the offerings are placed.
The Mem and Padmasana are chiefly regarded as the seats of the
supreme deity; the Merits are also provided with li/i^as, which
however are, usually, merely of pointed wood and are fixed in great
numbers in the roofs. The extremity of the Merits and also of the
other small temple-houses is generally covered with an inverted
pot or sometimes a glass, a circumstance that reminded me at
once of Buddhism, since this seems to represent the dome (or
bubble) which is the distinguishing feature of all Buddhist temples.
The Sivaites, however, will not admit this, but they give no
explanation of this ornament. The linga is also found in great
numbers on the wails surrounding the temples, and here is of
stone, shaped like the specimen which I have sent from Bolehng
to the Batavian Genootschap. The original signification of the
linga is almost lost ; the word now means simply "the most excel-
lent one." In addition to the above, we have temples on the sea-
coast, dedicated to the god of the sea, Baruna ; and further, small
houses in the sawahs and on the roads, dedicated to Sri t the
consort of Siva ; in the latter the passers-by strew a few grains of
rice, if they have any with them.
THE GODS WORSHIPPED.
In India, according to the popular belief, Brahma, Vishnu and
Siva, or the Trim Art i (Trinity), are the supreme gods. It is well
known that the Brahmans, originally the first caste, pay but little
honour to these gods, and that the Vedas place other deities above
them Vishnu and Siva, indeed, playing a very subordinate part
therein. The popular creed is further subdivided f into two great
sects, the one worshipping Vishnu, and the other Siva, as its prin-
cipal deity. In India, also, Brahma is not made the object of any
special worship ; as creator he is neutralized, as it were, for his
work of creation is accomplished, while the attention of mortals is
absorbed by Vishnu, the preserver, and the dreaded Siva, the
destroyer. It is certain that no Vishnuites ever came to Bali, nor
probably to ancient Java. The only idols in Java which un-
doubtedly represent Vishnu are mostly found in conjunction with
Siva, and, it would appear, are merely added to make the retinue
THE ISLAND OF BALL 103
of the latter god larger and more splendid. It may be safely
asserted that Vishnu has nowhere been the chief object of
worship.
In Bali all the characteristics, names and attributes of Vishnu
are given to Siva ; he combines in himself the power of all gods,
all others being as it were but other forms of himself. Siva is
the highest invisible firmament (akasa}, or dwells alone in the
heart ; Brahma, fire, icliicli through smoke becomes water f or
Vishnu. Hence it is also that a Padanda is called Siva ; if Siva
were not the all-comprising deity, completely idealized, that
designation would not be applicable to these holy men, whose
power, through the study of the Vedas, is greater than that of
the common gods. Siva's heavens are, the Merit, Kailasa,
Gunitng Agitng (Srarga or Indrahka ; Vishnu/oka or Brahmaloka,
and Srvaloka are the three heavens rising one above the other).
In the Indian mythology we find several gods (Vishnu, Durga,
Ganeta, Skanda, &c.) provided with many^rms, to indicate their
power. In Bali, four arms are given to Siva alone, while all the
other gods, unless they assume a demoniacal (Rakshasa) shape,
have but two arms. Siva also has a third eye in his forehead
(signifying in India the sun, but not recognized as doing so here)
(tnata trinetra} His names are : Paramesvara (the supreme
lord) ; Mahesvara (the great lord) ; Mahadeva (the great god) ;
brlkanda (the throat of eloquence?); Sudasina (with pure
throne) ; Givaka (meaning uncertain) ; Sangkara, Garba (the
foetus); Soma (the moon); Vrekanda (?) ; Kritlirasas (clothed
in a tiger-skin) ; Garbaduta (garbha, foetus, and duta, messenger) ;
Ganggadara (he who carries the river Gangga in his hair) ; Hara
(he who grasps) ; Kamari (the enemy of Kama, the god of love) ;
Vrishaketana (he who has the bull in his standard) ; Durjadi,
probably more correctly Durjati ; Triambaka (he who speaks
the three Vedas] ; Kawandi, Sarvajha (the omniscient) ; Viskandi,
Fisnaki ; 13 a ma (Varna, the left-handed); Mredda, Ugra (cruel);
Suit (he who bears the trident) ; Gandsara, Ganadipa (the lord
of hosts); Isa (the lord); fsana (ditto); Kandali, Matsya-
durita (matsya, fish, and durita, sin); Pasiipati (the lord of
creatures) ; Tripurari or Tripurantaka (the enemy and des-
troyer of the demon Tripura ; also Vishnu's name). Vrikshakctu
(he who has a tree in his standard) ; Sambu, Srava, ear, and
JBara, nature ; Dara (the holding one) ; Krisnarsa, Kusadi (he
who has the Kusagrass as his first attribute) ; Saddakaripu (the
enemy of the Saddaka} ; Sima (sinia, whole, or slma, boundary) ;
Pramesti or Faramesti (the highest); Nandakavahana (he who
rides on the Nandi) ; Kamadahana (he who has burnt the Kama} ;
Girlsa (the lord of speech) ; Pravesada, SAli (domestic ?) ;
Jli'atma (the soul of life) ; Isvanukara, Pitambara (covered with
a yellow garment); Berava (Bhairara, the terrible one, also a
1 04 AN AC CO UNT OF
subordinate deity in the demon-shape). Nllakanta, and Nila-
lohita (with a dark blue throat, from drinking the poison that
comes forth from the troubled sea) ; Sani (saui, the name of the
planet Saturn, and sani, worship) ; fsvara (the lord) ; Dristaketu
(with plainly-visible standard) ; Uinapati (consort of the Uma] ;
Chaturbuja (with four arms).
Part of the manuscript of the Chaniaka-Parva was wanting
here, but many more names were mentioned to me from memory.
Siva appears to have in all not less than a thousand names in
Bali, as in India. The following are further names : Bima (he
who is to be feared) ; Rudra (also a special class of eleven gods
of this name) ; Sara (nature) ; Kapalabrit (he who wears a skull-
chain); and finally Jagammta (the lord of the world). f This last
name always means Vishnu in India, but, in Bali, Siva is the
supreme and almost the only god, and thus Vishnu's principal
name is given here to Siva.
SIVA'S ATTRIBUTES.*
These are different in his various forms and temples. He has
the rosary (gnduha genitri, Sansk. akshamdla) ; the fan (iibas-ubas^
Sansk. chamara). These two symbols represent him as a peni-
tent. He has further the trisida (the pointed trident, to be
distinguished from the trident without points, which, e.g., the
JBagaran Trinavindu in the Batavian Society's Collection carries,
and which the Balinese call Tckan. Both tridents stand with
their points upwards ; the inverted trident (of Vishmi) I have not
yet met with. Siva also has in Bali, as in Java, the Padma (the
lotus), which in India belongs to Vishnu ; this, however, is not
regarded as the lotus-flower, but as a weapon. We have already
mentioned other symbols of Siva in the Sadkahyangan, namely,
Tnak, a sword-like creese ; panah, the bow ; tumbak, the lance ;
pdangj a sword ; sambuk, a whip ; and duuiig, a sort of creese.
With the exception of the creeses, all these symbols are also
Indian, and belong to Siva in India; the creese, however, is
purely Polynesian. The bow and the sword proper are not used
in Bali, and which are only known there from the religion and
the writings.
From Siva is to be distinguished Kala ; originally they were
one and the same, but Siva is the bright (white) god of light,
while Kala is the dark (black), terrible and destroying one.
Kala is worshipped, with Durga and the Bhfttas, in the Paua-
tarans and in the houses. The feast of Bayakala, the day before
* Siva became incarnate as Arjuna Vijaya. His wife, Devi Yajnavati,
commits suicide upon hearing a false report of the death of her husband, and
at the prayer of the holy Pnlastya is called to life by sans hyang Sagara (the
ocean) by means of Mntasanjivana (life-giving Ambrosia).
THE ISLAND OF BALI. 105
the Balinese new-year, is dedicated to him, and he must be pro-
pitiated by bloody offerings. The offerings placed daily before
the houses and in the niches of small pillars, or in the walls, are
also intended for him and the Bhtitas.
Siva's consort is Unui. This is one of the many names of this
deity, but not the principal one in India. In Bali it is used more
commonly than Parvatl ; she is also called Giriputrl (daughter
of the mountain; Parvatl has the same meaning). Durga is
distinct from her, as Kala is from &va. (Durga in conflict with
Mahishasura does not appear in Bali.) She is also called Devi
Ganggd and Devi Danu (the goddess of the mountain lake ; this
great mountain lake lies in the midst of the great Balinese range
of mountains) [Us. Bali, p. 274]; in this character she is wor-
shipped on the Gunung Batur, which rises from the middle of a
mountain lake (Damt, Jav. ranu\ and she is regarded as the
cause of eruptions and of the overflowing of the waters which is
indispensable for the cultivation of rice. As goddess of the rice-
fields she is called Sri (in India Vishnu's wife is called Lakshml,
who here is also Siva's wife) and has her temples on the sawahs
and on the roads between them. She is also worshipped along
with Siva. The same applies to Ganesa, who possesses no
temples and but few images in Bali. On account of his mis-
shapen form, he is not beloved.
Brahma, like Vishnu, has no special temples ; on great festivals,
small-temple houses are erected for both deities, when offerings
are made to all the gods, but after the festival these are taken
down again. The following are other names for Brahma : Cha-
turmukha (provided with four faces) ; Prajdpati (the lord of
created beings, the creator) ; Padmayoni (born out of a lotus ; he
is supposed in the Indian Mythology to have come forth out of a
lotus which rises to the surface of the sea out of the navel of
Vishnu as Narayana, resting upon the bottom of the sea).
According to the Balinese conceptions, he has only one head ;
if he is represented with more than one head, with four arms and
other extraordinary limbs, he is to be regarded as Brahmamtirti,
or as a Rakshasa (m&rti, the body, form, figure, does not precisely
express this idea). The same is true of Vishnu and other gods.
Brahma, the creator, has been deprived of all his distinguishing
features ; he has no temples either in India or in Bali, and he is
entirely subordinated to Siva, the supreme deity, and although he
appears in the Brahmandapurana as the creator of the latter,
Siva, when once created, possesses far greater power than
Brahma the creator. Brahma and Vishnu are looked on in Bali
as emanations or forces of Siva, and as related to each other ;
Brahma represents fire, Vishnu water ; the fire through smoke is
changed into water, and so Brahma's force passes into that of
Vishnu. Siva himself has the dkdsa, the highest firmament, as
1 06 AN A CCO UNT OF
his element, and he dwells in the hearts of the purified.* (See
as to Sadti, Parama, Mahd, Siva, Us. Bali, p. 307.) Brahma's
symbol is the dan da (staff) ; a staff is carried by the Brahmans, if
they become Panditas and hence it is that they are called
Padandas " provided with a staff." The danda, however, is also
regarded as a weapon, and includes the idea of punitive justice.
Brahma's wife is Sarasvatl, the goddess of eloquence ; she, too,,
has no special temples, but she has a feast-day in each Balinese
year, in the week of Vatu Gunong, on the day of Saiicsc/iara
Manis (Saturday). On this day all the manuscripts are brought
into the house-temples and consecrated ; the old prince Kassiman
brings his in procession to Gimong rata (his country residence) ; a
Pandita is called upon, and reads the Vedas over the manuscripts,
whereby their holiness is renewed. At the same time offerings of
rice, kwe-kwe, sirih, &c., are made to the goddess, and the floor
of the temple is sprinkled with holy water. Sarasvati's names,
according to the Chantaka-Parva, are : JBagi (bhaga, knowledge) :
BCisa (language) ; Giva, Givasa, Veda (Science) ; Vidayana (yidya,
knowledge, ay ana, road) ; Baradi, Yani, Sastravid (the learned in
writings); Sudevi (the good goddess); Darl (the holder); Sumari,
Ganggadari (she who holds the gangga) ; Prajnadari (she who
holds learning) ; Kastavit, Darjiinandari, Nilasiki, Satradana.
Vishnu is scarcely worshipped at all in Bali; as god of water
less honour is paid to him than to Banuia, although the latter is a
sea-god of inferior rank. The principal temples on the sea-coast
are dedicated to Siva ; we have already spoken of the erection of
a temple for Vishnu on festivals. Vishnu is nevertheless an im-
portant personage to the Balinese ; in his various incarnations^ he
is the hero of most of the Kavi works ; it might be said that Siva
is the high andinvisible, Vishnu the incarnate god, who has acquired
infinite fame by his deeds on earth, and whose conduct serves as
an example for all the actions of princes and people. His names,
although better known from the Kavi writings than from religious
worship, are as follow : Narayana (he who floats upon or in the
waters) ; Sori (Sauri, also a name for the planet Saturn) ; Chak-
rapani (he whose hand is armed with the Chakra) ; Janardana
(he who is plagued by men with prayers) ; Padniandbha (he who
has a lotus-navel ; see Brahma Padmayoni) ; (the holy) Kesi ;
Kesa (the fine-haired) ; Vekunta ( Vaikiintha, the careless one) ;
Vistara (collection ?) ; Srava (srava, the ear ?) Indravaraja (the
younger brother of Indra) ; Govinda (a name for Krisna as a cow-
herd) ; Garuda-dhvaja (he who has the Garuda in his standard) ;
A'csava (Kesa) ; Pundarlkakslia (the lotus-eyed) ; Krisna, PltAm-
bara (with a yellow garment) ; Siva also is so named above) ;
Visvaksena (visvak, everywhere, scna, an army ; whose army reaches
* SadA-Siva (the eternal Siva) is a well-known name for Siva in India, not
for Ijrahmft.
THE ISLAND OF BALI. 107
everywhere). Svabil (self-born) ; Sangkhi^it who has the angkha %
shell-trumpet) ; Danavara (perhaps Danan'iri, the enemy of the
Danavas, the demons) ; Hanoksaja (?), Vnksa (this must be
vrisha, the bull, also a name for Vishnu); Kapi (the ape);
Basudeva ( Vasvdeva, the father of Krisna, literally " the god of
riches," or Vasudeva, Krisna) ; Madava and Madusndana (the
conqueror of the demon Madii). These names are given in the
Chantakaparva in ttokas ; with a few slight alterations we obtain
pure inflected Sanskrit :
Visnu Narayana Sori, Chakrapani Janardanah
Padmanaba Resi Kesah, Vekunta Vistara Srawa
Indravaraja Hupendrah, Gohvinda Garudadvaja
Kesavah Pundarikaksah, Krisnah Pitambarochatah
Visvaksenah Svabu Sangkhi, Danavara Hanoksajah
Vresah Kapi Basudevah, Madava Madusudana.
Besides these, his avataras are well known viz., Matsya, fish ;
Vardha, wild pig ; Kurma, tortoise. To these must be added
two local ones, not known in India viz., Pati Gaja Madda,
founder of the Karang-Asem family, and the cock Silingsing, the
apotheosis of cock-fighting. Parta and Maruta also are incarna-
tions of Vishnu, slain by Ravana. The following are yet other
names for Vishnu : and Vamana and Tripurantaka (the dwarj
and the conqueror of the demon Tripura ; the latter name we have
also found to belong to Siva, and it has reference to the fifth
Avatdra] ; JVarasing/ia, man-lion (in the fourth Avatard) ; Rama
(in the seventh Avatard]; Purusottama (the most excellent of
men, with reference to his numerous incarnations). His symbols
are, in the first place, the chakra sudarsana (the disc, which, being
well slung by him, penetrates everything, and returns to him); and
then the sanghka (the shell-trumpet), the gadd (club), danda (the
staff, also belonging to Siva and Brahma) ; the same applies to
the tuak (a sword-like creese, which in one of the sadkahyangans 1
belongs to Siva as a distinguishing symbol). In images of Vishnu
and Brahma we also find a circular mark on the forehead ; this is
a third eye, but appears to point to the Indian tilaka, the mark of
the sect. No other remembrance, however, of this tilaka (which
name is unknown here) has been preserved, and the Balinese
seem to draw the mark on the foreheads of the gods in accord-
ance with a tradition which is no longer understood, or endeavour
to keep secret the origin of this sect-mark, in order that the form
of religion, at present existing, may be regarded by every one as
the only and true form. According to the statements of the
1 [/.*., the six heavens. Sanskrit linguals are often represented by the
corresponding (unaspirated) dentals in Kavi and Balinese ; while, on the other
hand, Sanskrit dentals frequently pass into linguals in those languages. Thus,
sad= Sanskrit shad; but danda ^ from which padanda, = Sanskrit danda, \
1 08 AN A CCO UNT OF
priests, Vishnu and Brahma are invoked in the Vedas, and do
indeed appear in the Indian Vedas, although as gods of a very
inferior rank ; if they play a higher part in the Vedas of Bali, we
should again be compelled to entertain some doubts as to the
authenticity and originality of these Vedas.
Vishnu's wife is $ri. We have already met with Siva's consort
under the name Sri, as protector of the rice-fields and goddess of
fertility. According to Indian ideas, this is always Vishnu's wife ;
but just as many of the names and attributes of Vishnu are in
Bali given also to Siva, so Sri, originally the consort of Vishnu,
appears in Bali as one of the names of Siva's wife. We have
already said that Vishnu and Branum are but other forms of Siva ;
and so also their wives belong, as special forces (Sri, goddess
of fertility, of abundance ; Sarasvati, goddess of eloquence and
learning), to Siva, the supreme deity. Laksml is unknown as a
name for Vishnu's wife. As Vishnu's consort, she has no special
temples. The mark on her forehead, and on that of Sarasvati, is
called peryasan, derived probably frornyasas, Jav. yasa : fa?ne, and
in that case meaning excellence. The signification of this word,
however, is not clear; and upon asking whether this were the
sectarial mark (Tilaka), I was answered in the negative.
We have thus found that the three supreme gods of the popular
creed of the Hindus are looked upon as expressions of one and
the same force, are worshipped together, and regarded, as it were,
as one being. Siva in the popular belief also is almost the sole
god; the inferior gods, with Indra at their head, are his le^er
forces. The different names of the gods mean for the ignorant
people, it is true, as many different gods, but the priests hold
other views.
We will here say a few words respecting Indra and the inferior
gods, and will then give an enumeration of the gods, as, according
to the Brahmaiidapuraua, they were created.
Indra. This deity, the prince of the Devas, that is, of the sub-
ordinate gods, who require the Amrita to keep them alive, and
who are often brought into danger and vanquished by doers of
penance and giants, has, singularly enough, special temples in
Bali. Sakennan, in the island oiSerangan in J3adong, sx\&Jempul
\nJBangli, we have already become acquainted with as such. The
explanation of this we think is, that our Sivaitic sect has suc-
ceeded in making Vishnu (and Brahma) of little importance, and
in causing him to be regarded as a part, an emanation or force of
Siva, but did not find it necessary to deprive of his worship the
popular Indra, the example of princes, who is glorified in so many
poems. Indra could never be inimical to the consequence of
Siva, and was therefore harmless to the imported Sivaism, and, by
allowing him temples, the prejudices of his devotees were perhaps
spared from a blow. His temples and attributes are even of
THE ISLAND OF BALL 109
considerable importance. He ha^ also the third eye. The fol-
lowing are among his names: Satakratu (worshipped with 100
offerings) ; Trinetra (provided with three eyes, like Siva) ; Sahas-
ramtra (provided with a thousand eyes ; these are the stars, Indra
himself the visible, heavens, while the higher, invisible firmament,
akasa, represents Siva ;) Devaraja (the king of the Dcvas or sub-
ordinate gods) ; SacMpati) the husband viSachi.
His weapon is the bajra (vajra, lightning; here, however,
represented as a kind of weapon in the form of a thunderbolt).
Indra's wife is SafM, only remarkable on Indra's account.
The eight gods of the cardinal points (Lokapald) are named
very frequently in the writings ; in the religious worship they are
less prominent. In the Ramayana, lont, 181, these eight gods
are enumerated as follows : Indra, Yama, Sfirya, Chandra, Anila,
Kuvera, Baruna, Agni. We meet with the same names again in
India, only Nirriti appears there instead of Surya, and fsaril
instead of Chandra. The order in which they stand, beginning
with the east and going round by the south, is however different,
and in India is tolerably fixed : Indra, in the East ; Agni, N.E. ;
Yama, S. ; Surya (or Nirriti'], S.W. ; Varnna, W. ; Vayu (or
Marut, Pavana, and in Bali Anila, all meaning wind}, N.W. ;
Kuvera, N. ; Chandra (or tsani), N.E. " Usana Bali," p. 261, gives
the eight cardinal points thus : (i) Purva, East; (2) Gneha, S.E.;
this is Agneya (" Wilson," " the South-East quarter "), to be derived
from Agni, fire and the god of fire, whose throne is in the south-
east ; (3) Daksina, south ; (4) Neriti (Sansk. Nairriti, belonging
to the south-west quarter, to Nirriti ; Nirriti, according to some,
Surya, according to others, presides over that point of the com-
pass), S.W. ; (5) Paschima, behind or west; (6) Vayabya (Sansk.
Vayavya, belonging to Vayu, the wind, compare Anila, Pavana],
N.W. ; (7) Uttara, north ; (8) Esania (not Resanid], answering
to the Sansk. aisanya or disani, "belonging to ham" N.E. Here,
therefore, we have the names of the cardinal points accurately
preserved by adjectives derived from the names of the guardian
deities ; even the less-known Nirriti and ham are not forgotten.
It cannot surprise us, however, that in Bali all the cardinal points
are attributed to various forms of Siva, although this alteration
seems to be of later date.
Yama and Baruna are the only gods besides Indra of any note
in the religious worship ; they also are, to some extent, identical
with Siva. With Indra in his heaven we find the Varapsaras (the
most excellent Apsaras], and the Vidyadaras and VidyAdarh (male
and female spirits), as well as the Rests ; the last are the human
beings who have become gods, after having attained, through a
holy life, to Indra's heaven ; his heaven is called Svarga, or
Indraloka, and is the ideal of a royal dwelling ; the descriptions
of it agree with the Indian accounts. Even this heaven and its
no AN ACCOUNT OF
inhabitants are not safe from the attacks of foes (see the Vivaha,
the combat between Indra and Havana [the latter is called
Indrajit, the vanquisher of Indra] ; and other myths) ; its in-
habitants are also liable to become human again, at least they
require the Amrita in order to preserve their divine power.
Indra's heaven lies beneath that of Vishnu (and Brahma), and the
latter beneath that of Siva. It is not until it reaches Siva's
heaven that the soul attains the repose and release of trans-
migration (Moksa), But little is heard in Bali of Vishnu's heaven ;
it is known rather from tradition f than from religious doctrine.
Every man endeavours to reach Siva's heaven (which is to be
sought on the Kailasa, the Meru, or the Gunung Agung, in Bali,
at its highest point) ; but only a Padanda appears to have a right
to immediate entry therein. The rest of the people have to be
satisfied with Svarga, Indra's heaven, where they hope to live for
ever, entirely after the Balinese manner of living, but without care
and with greater splendour. The attainment of the Svarga is in
some cases immediate ; a Bela or Satia who follows her husband
into the fire, passes into that heaven at once ; a prince, who sacri-
fices himself and his adherents in defence of his country, goes with
all his followers to Indra's heaven, where, probably, they fall again
to fighting. Cremations also, if they be accomplished according
to rule which is difficult, as the priests can very easily discover a
fault, if they have not been consulted as to all ceremonies, how-
ever insignificant are considered to bring the subject of the
cremation to Svarga. But this last means of entry is not looked
upon as so certain as the two others ; the Balinese say of several
princes that, although they were burnt with all proper ceremony,
they still wander upon earth in the form of animals (frogs, snakes,
&c.). Another obstacle is the judicial power of Yama, who
judges the dead with strict impartiality in the lower world
(Naraka). Perhaps the postponement of cremations for two
months, and sometimes for several years (up to twenty), has
reference to this preliminary judgment by Yama, and the punish-
ments imposed by him ; here, however, as is the case in most
religions, there is a palpable inconsistency : it is believed that the
souls of the dead, provided that no outward observances have been
neglected, pass immediately into Svarga ; it is at the same time
believed that the soul must first be judged by Yama (the Indian
Pluto, and the lord, Jehovah, who, according to the Old Testa-
ment, will punish Jewish sinners), and must, for all sins, perform
penances which will last millions of years, and which hinder them
from reaching Svarga. Doceant thcologi meliora !
Baruna (Sansk. Varuna} is the god'of the sea; the temples on
the sea-coast are dedicated partly to him and partly to Siva.
Vishnu also is stated to be a sea-god, or god of water (represented
in India on this account with the inverted trisida], but he has no
THE ISLAND OF BALL 1 1 1
temples either in this or in his other capacities. The subordinate
Baruna, however, for the same reason as Indra, is not erased from
the list of the gods who are worshipped. Baruna is sometimes repre-
sented as a youth, in the same manner as Indra. He is then dis-
tinguished by the pasa. This should really be a sling, with which
he catches up the dead bodies, &c., and Yama the souls ; in Bali,
however, the pdsa, contrary to the meaning of the word, is a long
.dart, round which a serpent winds, and which ends in three teeth
(like the trisula). Baruna has usually a monstrous figure with the
head of a Raksasa, from which a serpent's tongue projects, and a
huge serpent's tail turning upwards, the rest of the body being
human. This serpent-form indicates his nature as sea-god. The
deva agun sagara (agun must be agung; Crawfurd, On the Island
of Bali, " Asiat. Res." vol. xiii#) is the same as Varuna.
Yama, the god of death and of punitive justice, is^ identified
with Siva (and Kala) ; he has no special worship, but Siva is also
worshipped under the name of Yama. He is called sang hyang
Darma (justice); and Prctardja (the prince of the departed).
His severe punishments (e.g. the boiling of a soul in a copper
kettle for thousands of years) are very well known. He is regarded
as a god, not as a Raksasa, although he has teeth on his cheeks
and his forehead (dangstra, comp. Ganesa], and, besides this, the
well-known tusks of the Raksasas (siyung or chaling) his symbol
is the gada, club.
The rest of the gods of the cardinal points have, as it were, no
worship. Surya is the same as Siva, and the worship of the
priests is addressed to him ; masuryasevana (worship of the sun).
Chandra, the moon, in India an attribute of Siva, did not become
known to me as such in Bali, but all principal feasts are regulated
according to the full and new moon. To insure success, all great
undertakings (offerings, cremations) must be carried out in the
first, or white, half of the moon. Anila or Vayu (Bal. Bayit) has
no worship at all, but is regarded as the vital principle ; in fasting
(brata) the doer of penance shall live by the vayu alone. The
vayu in its various forms come also under notice in the healing of
the sick. The physician causes his vayu to pass into the sick
man's body. The Padandas are, therefore, in great request for
curing sick people, because their vayu is particularly holy.
Kuvera, the god of riches (Bal. Kubera}, known only from the
myths. Still less is known of Nirriti and Isarii. Agni, finally,
the god of fire, is frequently invoked in the Veda-worship in Bali,
as in India, but has no temples, and is not worshipped by the
people.
To the family of Siva, but still as subordinate gods, belong
Ganesa and Kartikeya or Skanda, his sons. I have not yet
become acquainted with the latter in Bali ; of Ganesa (commonly
called Gana\ on the entrances, there are images of stone (Ganes'a
ii2 AN ACCOUNT OF
images are introduced into Bali by Kabu Ayu ; compare Abdullah
in the T. v. N. /., 2, 161 seq.) and paintings, all tolerably rude.
The stone images are sometimes to be found in the temples of
Siva, and Ganesa then appears publicly as a subordinate, attendant
deity, very much resembling the Raksasas that keep watch.
Special worship and offerings for him are unknown. In literature,
however, Gana is as important in Bali as in India. He is the god
of learning and of cunning (also of the orang dagang and of
thieves). He is called Ganapati, lord of hosts (he is the leader
of the yet inferior gods) ; Vindyaka (the leader, in reference to
the same idea) ; Sarvavigna (from san>a, all, and avigna, without
obstacle i.e., he who overcomes all obstacles); Vignakarta (utgna,
obstacle, karta probably from krit, to cut, to solve he who
removes all obstacles). In the drawings which are made of him, he
holds in his right hand the pustaka, a book of lontar leaves as a
symbol of learning. Of the rest of his (Indian) attributes, with
the exception of the moon, I have seen or heard nothing. He
has an elephant's trunk (tulali) and elephant tusks (gading}, and
also the dangstra (which we have found in Yama), on his cheeks
and forehead. It is only in Ganesa and Yama that these mis-
formed parts do not indicate a demoniacal nature ; they belong
otherwise only to the Biitas and Raksasas, and to gods who by
choice assume demoniacal forms. All the rest of Ganesa's body is
entirely human, and he has onjy two arms, the reason of which we
have already seen above in Siva's f case. Independently of the
preference given to the worship of Siva, the sole and supreme god,
the absence of homage to Ganesa may also be explained by his
shape. Everything monstrous is regarded in Bali with a certain
repugnance ; it is true, the numerous representations of demons
conflict with this statement, but then these are the hostile spirits,
which are overcome in great measure by the beneficent gods. It
is especially remarkable that the form of an elephant is looked
upon as unlucky, although I cannot assert that this was always the
case. A former prince of Boleling kept an elephant and used to
ride out upon it ; his conduct was universally condemned, and to
this is ascribed the fact that this prince was punished by the gods
with the loss of his kingdom. The abhorrence of the tiger is more
natural ; if tigers make their appearance in a kingdom of Bali, it
is believed that that kingdom will speedily fall (come under the
dominion of the evil spirits).* The rhinoceros, on the contrary,
enjoys great honour, although not in life. The Deva agung and
also the prince of Lombok have asked the Netherlands Govern-
* In the greater portion of Bali tigers do not occur ; they swim over to Jtm-
brana from ffattyuwanng, and remain in that nearly uninhabited district and
in the mountains of Tabanan and Boleling. The high cultivation of the country
prevents them from spreading further. "Their appearance is a sign that men
must depart.
THE ISLAND OF BALL 113
ment for one of these animals for great sacrificial festivals ; they
use the blood and fat for the preparation of various offerings, and
the excrementa as medicine.
Kama, the god of love, and his wife Rati, have, so far as I
know, no special temples, but yet are much honoured among this
very erotic people. Kama also is again, as it were, a form of Siva
(see " Us. Bali," p. 275). Other names of Kama are sang hyang
Smara (about equal to the Latin citra, for " to be in love ") ; also
Anangga, the bodiless, for, according to a myth, Siva has burnt
his body ; and Manobu, born in the heart.
Vasuki y the Indian serpent-king, nearly coincides here with
Ananta (eternity), or Antabhoa, the serpent upon which Vishnu
rests. Vasuki also belongs here to Siva's retinue ; he dwells in
the various principal temples in which Siva is worshipped (in the
sadka/iyangans). After the time of worship in Basuki, at the foot
of the Gunong agung (which place is named after the same ser-
pent), it is supposed that he goes through the air to Ulnvatu, the
sanctuary on the point of the table-land (in Badong\ and so round
to the other Kahyangans as well. He is then to be seen as a
fiery streak in the sky. The brightness comes from the precious
stones with which his immense head is adorned. Many apparently
meaningless fables exist about Vasuki (in Bali always called
gasitki). As yet I have not succeeded in extracting the " sachen
aus diesen sagen." I have long thought of a former serpent-
worship, especially as a Padanda is also called Bujangga (serpent),
and in the " Usana Java," Siva, Buddha, and Bujangga are called
sons Sang of Haji (adia, as it seems to me, " the first "), a circum-
stance which indicates very strongly r the existence of three different
forms of worship viz., the purely Sivaitic, the Buddhistic, and the
serpent-worship (?). Of purely Indian myths, that of a former
destruction of the serpents, in which Taksaka, Vasuki, and a third,
at the prayer of a penitent, alone were spared, is known in Bali
from the books. This sacrifice of serpents (Sar'payajna) was
accomplished by king Janamejaya, the great-grandson of Arjuna
Vijaya (compare the Raksasa- Yajila of Bagavan Vasistd). The
bird of Vishnu Garuda is frequently represented in Bali, always in
monstrous Raksasa shape, with a beak and wings, but at the same
time with tusks; in other respects it has a human body. His
parents are Kasyapa and Vinata ; Aruna is his brother. The
most inferior persons of the Indian mythology, such as Kinnara,
Kiinpurusha, Uraga (serpents), Detya, JDanara, Pisacha, and
others, are known in the literature of Bali. For the most part,
however, we meet only with the Gftndaroas t the Vidyadars, and
Apsaras, in Indra's heaven, and the Detyas as giants of antiquity;
the Raksasas and Bhiitas (real beings, evil spirits) as enemies of
mankind and opponents of the beneficent gods, dreaded, yet
always to be propitiated.
SECOND SERIES. VOL. II. I
ii4 AN ACCOUNT OF
In the last-mentioned we clearly recognize the principle occur-
ring in all religions, of a good and an evil supreme power, the
conflict between which in the Hindu doctrine, it would seem, is
never, and never will be, decided. A union, however, of the
two powers is clearly apparent in the fact that Kala and Durga,
the heads of the Raksasas and Bhittas, are regarded as no other
than Siva and his wife, since the gods possess the power to change
the latter into Raksasas.
The accounts relating to the Bhiitas are confused ; there are a
great number of names for a few of them e.g., buta Vilis, buta
Larehan:* they are also named after the shape which they assume
e.g., Imtit hulu asu, " the buta with a dog's head " (asu, Sansk. srci,
dog), buta hulu Icinbu, " the buta with the head of a cow." A col-
lective Balinese name is dagan ; their haunts are chiefly burial-
grounds and unclean places, and at night they break into the
houses which are not protected from them by means of offerings.
With the people in general the Liaks 1 are still more common than
the Bhutas. The former are human beings, who, by the knowledge
of certain mantras (magic formularies), can alter their shapes and
also render themselves invisible, a bright light, proceeding from
the place of the tongue, alone remaining; they are obliged to feed
on carcases, and chiefly haunt burial-grounds and the places where
corpses are kept for cremation. They also take out the entrails of
sleepers, so that the person thus robbed must die in a short time.
Their mistress is Rangda ning Gt'ra, the widow of Gira, whose
history is found in the Chalon-Arang (a JBabad) ; she lives on the
Gunnng agung, where the Liaks hold their assemblies. Fire-flies,
which are very numerous and of large size in Bali, are sometimes
said to be Liaks ; moreover, the accusation of being a Liak often
affords reasons for declaring a person to be innocent.
THE CREATION.
According to the Brahm&ndapurdna, where the world is
created from an egg (anda\ four beings come first into existence,
through the penance performed by Brahma ; two of these are
Siinandti and Sanatkumftra, and the other two are not named.
Then the heavens, the rivers, the sea, the mountains, plants,
shrubs, time, &c., come into existence. He (Brahma) further
creates the Dcvarshi (the divine Rishis}, Marichi, Brigu,
Anggira, Pulastya, Pitlctha, Kratu, Daksha, Atri, Vasishtha. It is
not till he has done this that he creates the Paramesrara, the
Supreme Lord, a name for Siva ; the latter is at once regarded as
Brahma's grandfather ! He is called Bava, nature ; San>a, all ; fsa,
lord; Bima, the terrible one; Ma/nidcva, the great god. His body
* Also Kiika., servant of Dnrga (in India Kalika is another name for Durga).
Xlika was probably originally the same person as Durga.
|\\'ilken, 1. 1. p. 22: and R. van Eck, " Balineesch Woordenbock," s. v.
THE ISLAND OF BALI. \ 1 5
consists ( i ) of Aditya-Sartra, sun-body ; ( 2 ) Vcscirira, water-body ;
(3) Bay it sari ra ; (4 ) Agtiisarlra, fire-body 5(5) Akda t the higher
invisible heaven; (6) Mahapaudita, the great Pandita ; (7) Chan-
dra, the moon ; (8) Batara Guru, the teacher. He is therefore
called Ashtatanu, with eight bodies. He must be worshipped
through the Sadhaka, the full priest, or the performer of the cere-
monies. Kalpa, form, and Dharma, justice, are children of
Brahma, sprung from his spirit. It is stated here from which
parts of the body of Brahma the Devarshis carne forth.
The lord then created the gods (Devas), the Asuras (evil spirits),
the Pitris (shades), and man. He also creates out of his own
body the Yaksha, a sort of demon ; Pisacha, lemures ; Uraga, the
serpents which are worshipped ; Gandharva, heavenly musicians ;
Apsaras, the heavenly female dancers; Gana, the hosts, whose
leader is Ganapati ; Kinnara, elves ; Rdkshasa, demons ; and,
finally, the animals, pasu.
Then the four castes are created : the Brdhmans, out of
Brahma's mouth ; the Kshatriyas, out of his arm ; the Vesyas ( Vais-
yas), out of his thigh ; and the Stidras, out of his foot. Then
Dharnia (the Dharma already mentioned above), and Ahingsa
(husband and wife), jus fife and the sparing of everything that is
alive ; these are the two principal virtues of the Buddhists. It
appears, therefore, that in the ancient combination of Brahmanism
and Buddhism in Java, Buddhistic doctrines crept even into the
Brahman dapurdna. I cannot say for certain, however, whether
these are not found in the Indian Brahmdndapurdna as well, this
work being hitherto but little known to Europeans. (There are
many repetitions in the manuscript.) Then Svayambhuva Manu,
(" the first Manu" the Manus govern the world during a certain
period, kalpa, " from the one creation to the other ") and batari
atarfipa are created ; they beget the batari Rati, the wife of
Kama, the god of love. Then follows the genealogy of the race
of Svayambhuva Manu and their relationships with the nine
Derarsis. The descendants include twelve Yamas, and Lakshml
(in India, Vishnu's wife). Buddha is the son of Buddhi, the
understanding. The rest of this race are attributes of nature, of
the mind, the heart, and also the body. Nilalohita (Siva) has a
thousand children, the Sahasra Rudra. Sri is the daughter of
the Devarshi Brigu, married to Vishnu; their children are Bala
and Buddana. Sarasvati is the wife of Purnamdsa, the full moon.
Agni is the son of bagavan Anggira and Smriti, tradition ; among
his descendants is Parjanya, the god of rain. The holy Pulaha
is the ancestor of the Kshatriya Daha, the warrior-caste vtDaha or
Kcdiri : this royal family, represented in Java in Majapahit, and
in Bali in the race of the Dcva Agimg, thus derives its origin from
a Devarshi, and ought properly to belong to the Brahman caste.
They are the grandchildren of Kamya and Priyabrata.
I 2
n6 AN ACCOUNT OF
The 60,000 Balakhilyas, who are all Brahmachari i.e., students-
of the Vedas (compare the 60,000 Sagarides in the Balakanqa of
the Indian Ramayana}, are children of Kratu and Sannati.
Besides a great number of other mythological personages who do
not appear in the religious worship, but are sacred to the Brah-
mans, we find here the seven Rishis, who existed in the time of
Uttama Manu, the second successor of Svdyambhuva Manu : they
are Raja, Batra (Badra), Urddabdhit, Lawana, Anaya, Satapa, and
Saka, and are descended from the Dei'arshi Vasishtha.
We give here a few examples from the original :
Agre sasarjja bhagavan manasam utmanah samam.
" In the beginning the holy one created the soul, which was
like unto himself." This is explained thus : mayoga batara Brahma
m'tu tang Resi patang siki sang Nandana Sanatkumara. " Batara
Brahma held the yoga (was sunk in meditation), thereby originated
the Resis, four in number : sang Nanda7ia and Sanatkumara^
The other two are not named ; according to the Indian tradition,
however, Sdnanda (sic) and Sanatkumara are two of the four first-
born sons of Brahma. It still remains to be investigated whether
the sons of Brahma are representatives of the (world-) soul, which,
with the means at present at my disposal, I am not in a position
to do.
Tatwasrijat punah Brahma.
" After that Brahma created again." Here follows in the Bali-
nese commentary the creation of Siva, and an enumeration of his
different characteristics and bodies. T/u's, then, appears to be an
interpolation into the original work : the Sanskrit text says nothing
about it, and this creation is certainly not in its proper place. But
with a sect such as that in Bali, Siva could not be created later on
iv ithout losing importance ; therefore advantage was taken of^he
Sanskrit words, which scarcely any one understood, to glorify Siva
by means of an incorrect commentary. Tatwa has been formed
here out of tato-asrijat ; the a has not been elided, but has been
altered with the preceding o into wa ; in the Sanskrit the word
should be tato-srijat. This Tativa occurs again in the next half-
sloka, but there it can only be explained by tatha "so," " in the
same manner."
Tatwa devasurapitrin manushyako-srijat prabhu.
" So also the Lord created the gods (Dei'a\ the Asuras, the
Pitaras, mankind."
In Manushyako (see above) there must be an error ; manusyaka
is a " multitude of men," but this does not suit either the sense or
the metre in this passage ; had it to be brought into the metre, it
would be nianushyakam asrijat prabhuh, which gives a syllable too
many ; moreover the fdung (the sign of the long a) is wanting in
THE ISLAND OF BALI. \ 1 7
-the manuscript. We have altered the Sanskrit text into manush-
yansclia "and the men (mankind)," which agrees admirably with
the sense, and leaves the metre undisturbed. Our passage is re-
produced in the Balinese commentary thus :
Muwah mayoga batara Brahma, mijil sang deva sura pitr;
man usa, tuhun sangkanika mijil saking manah hikang deva, ring
lambung sangkaning pitrO, ring pasva sangkaning manusa, jagana
*sangkaning Asura.
"And further was batara Brahma sunk in meditation; thereby
came into existence the Devas, Asuras, Pilaras, and mankind ;
the place whence they sprung (out of his body) was : the Devas
came forth out of his mind (manas) ; the Pilaras out of the hollow
under his arm ; man out of his side (for pasva we must read
parhva} ; his thigh was the place whence the Asuras came forth."
The time also when they came into existence is stated, and this
certainly has an astronomical meaning, since in India also the
Devas and Asuras play an important part in astronomy (see
Bentley on the Indian Astronomy). The Asuras are born at noon,
man in the morning, the Devas at midnight, and the Pita-Pitarah
(the worshipped shades of the departed) in the evening.
We will add a brief sketch of the survey of the world as it
appears in the BrahmCindapurana.
The world is divided into seven Dvtyas (islands) : Pushkara-
dv'ipa (the lotus-island), Kronchadv / ipa,Kusadvlpa,Sangkadv'ipa (Skt.
SdkadiApd), Sahnalidvlpa, Plakshadvlpa, Jambudv'ipa. The last,
the island of the famfat-frnit, is India and the adjoining countries.
All these islands are surrounded by the ocean, and lie round about
Merit, the centre of the earth, like lotus-leaves. The islands are
under the rule of the grandsons of Svayambhuva Mann, the first
of the seven Manus, who govern the earth for the space of a kalpa,
a long period of time (see below). They are named Savana,
futitnan, Vapushmanta, Medhatithi, Gomeda, Avya and Agmdhra.
The last rules <y*vc Jambudrtpa. Jambudv'ipa itself is divided into
nine varshas, more or less distant from Meru, which is sometimes
to be regarded as the North Pole, and sometimes as the centre of
the earth ; great ranges of mountains, mostly fabulous, divide
these varshas from each other. The children of Agmdhra are
.(i) Nabi, (2) Kimpurusha, (3) Harivarsha, (4) llavrita, (5) Ram-
yaka, (6) Hiranyaka, (7) Kuril, (8) Bhadra, (9) Ketumala. Most
of these nine also give their names to the varshas, over which they
preside, (i) The land (the varsha} of Nabi is to the south of the
Himavan (Himalaya'] this is therefore India proper. (2) Between
Himavan and Hemakiita (the golden peak) is Kinnaravarsha,
under the prince Kimpurusha. The Kinnaras or Kimpurushas
(literally " what sort of man ") are barbarous nations in the north
of India, represented with horses' heads ; they are a sort of
centaur, and indicate the nomadic equestrian nations of the north
nS AN ACCOUNT OF
(in Tartary}. The Hemakuta is a mountain range to the north of
the Himalayas, in Tibet ; it is not loftier than the Himalayas, but
is represented as being so. In this range also is the fabulous
Kailasa, Siva's seat. (3) Naishadavarsha : NaisJiadJia is a range
of mountains to the east of Hcinakiita and south of Ilircrita. A
country iu the south-east of India also bears the name of Nlshadha r
well known through its prince Nala (Naishadha, in the Maluibha-
rata the episode is published by Bopp). Here, however, we
have to do with the northern Nishadha, of which Harivarsha is
prince. (4) llin'rita (or Hilavrita) the name of prince, country,,
and mountain range. This is the highest and most central range
on the earth, according to Indian ideas; Merit, the highest
mountain and the seat of the gods, is situated in it. (5) The
region north of Ilavrita and Nilaparvata, called Ramyaka or
NllararsJia. Nilaparrata, the blue mountains, are as mysterious
as Ilarrita with Merit ; we do not find either in the position indi-
cated by the Indians, but we may perhaps suppose the Aryans
(Arya, the Indian), who descended to the plains of India from
the mountainous regions of the north-west, to have brought with,
them some knowledge of the ranges (Altai, Caucasus) of northern
and western Asia. (6) Hiranyakavarsha to the north of Sreta-
parvala, the white mountains, northwards of Nilaparrata-
(7) Kururarslia to the north of the Sringavcin range ; this is the
Uttara-Kuru. the most northern and the coldest land, but the
land whence the Aryans appear originally to have come. Accord-
ing to the Zendmyth, it was in ancient times a fertile, inhabitable
land, and was changed into a cold wilderness by Ahriman (the
evil spirit). In this may, perhaps, be recognized a trace of a
remembrance of the changes which have taken place in the
climates of our earth, such as that to which the fossil bones found
in Siberia bear witness. (8) Bhadravarsha, to the west of
the Malyaran (the ilowery mountain), which itself lies to
the east of the Mem. Bhadrararsha is thus by the side and
to the eastward of the great chains of mountains. (9) Ketumciln.
dwells to the east of mount Gandamddana (delightful to the
smell). This is the country lying to the west of the other Varshas
(Persia?). We, therefore, have (i) India proper, as far as the
I [imalayas ; (2) the region between the Himalayas and ffemakfita
(Tibet); (3) the region between Hemaknta and Jlarrila (with
A/em); in our enumeration this is No. 4 (Tartary, Mongolia?) :.
(4) the region to the north of Nllaparrata, north-east of the J/ivv/
(Eastern Mongolia, Manchuria?) ; (5) the region to the north of
Xi ] etaparvata, north of the previous country (Eastern Siberia ?) ;
(6) Uttara-Kuru (northern and western Siberia to the Arctic
Ocean). These six form almost a continuous series from south to
north, with the Merit as a centre. Then come to the east Nilavarsha.
(China ?) and Bhadravdrsha^ and to the west Ketvm&lavarsha (Per-
THE ISLAND OF BALI. 119
sia and the western regions). Concerning the rest of the varshas
nothing further is mentioned ; the narrative proceeds to enumerate
the princes who ruled in India proper, descendants of Nabi,
the eldest son of Agmdhra. Nabi is an Emperor, the universal
ruler ; he is also a Kshatriya. From Nabi and hir, wife Manudwi
are descended successively (i) Rishaba, (2) Bharata, who dwells
south of the Ifimaran in Bliaratavarsha, (3) Siunati, (4) Tejasa,
(5) Indradyumna, (6) Parameshthi, (7) Prati/iara, (8) Pratiharsha,
(9) Unnata, (10) Bhava t (n) Mudgifa, (12) Prafastavi, (13)
Vibhu, (14) Prithu, (15) Nakta, (16) Gtf/tf, (ijjfara, (18) F//,
(19) Dlman y (20) Ma/ian, (21) Bochara, (22) Toshta (Tushta\
(23) Viraja, (24) 7fa/tf, (25) TV? (!). These are the Kshatriyas
(princes) in the second great period of the world's existence, Tre-
tayuga. Several of these names are those of the Jaina deities ;
fat Jainas are disguised Buddhists, who still exist in India. These
names give us a hint towards the comprehension of Buddhism in
Java. Were the Buddhists of Java Jainas ; and have we to
attribute to that sect the union of the Buddhistic and Sivaitic
religions and doctrines in Java and Bali ? The Jainas at any
rate worship the Brahmanical gods besides the Jinas, and have
even retained the institution of the castes in order to protect
themselves from the persecutions of the followers of Brahma.
Under Svayambhuva Manu there are further eleven Rudras
(see the feast of Ekadatarudra), twelve Adilyas (the twelve solar
months), eight Basils (Sanskrit Vasum Vasudeva\ twelve Sadhyas,
ten Visvadevas, two Sanggis (?), twelve Bargawas. And further
there are the Devas, Asuras, Gandharvas, Yakshas, Pisachas,
Rakshasas ; these are immortal, it is true, but yet are born again.
Their lifetime and that of man varies according to the different
yitgas (Satya, or Krita-, Treta-, Dvapara-, Kaliyuga\ and gradually
decreases. In the Tretdyuga man attains the age of 188 years, at
the close of the Tretdyuga 147, in the Dvaparayuga 126, at the
close of the same 105, and in the Kaliyuga only 84. The lifetimes
of the inferior gods differ in the same proportion. The angulas
(inches) are given as the measure of time, but at present I am
without any explanation of their astronomical meaning. Hereupon
the work speaks of the Manva?itaras, the periods of time in which
a Manu governs ; these are also called kalpa, and they contain seven
chaturyugas ; according to the work this appears to be the time of
one Mann, but the latter is really but one chaturyuga (a combina-
tion of the four Yugas, a Kalpd] ; there are, however, seven
Manus, and the seven chaturyugas are, in fact, the time of the
duration of the world. After each kalpa-cx chaturyuga the world
is destroyed and created again.
Here follows a description of Chaos. Manu alone was in
existence. He (not Brahma) then creates a series of beings ;
Deva, Rishi) Asura, Pitri, Manusha, Bhuta, Pisacha, Gandharra,
120 AN ACCOUNT OF
Yaksha, and Rakshasa. Mann is called here Sumantia (?) and
sa/ig Sista (the instructor) He instructs* the beings who through
his penance have come into existence, in the SadachCira (the right
conduct); this consists of lobhadeya (lobha, "greedy desire,"
sensuality, covetousness, and adeya, probably from ada, sumere,
tollere ; thus " the putting away of greedy desire ") 1 ; kshamd
(patience), satya (truth, truthfulness), Tidy a (knowledge), ijyd (the
making of offerings), dana (alms-giving). The attributes of the
Sad&ch&ra are seven in number {Saptakani charitrani}.
The stages in the life of a Brahman are also given : ( i ) Brah-
ma chari, he who lives as a pupil with his Guru; (2) Grihastha,
the head of a family, the married man, whose duty it is to exercise
the dharma (right) ; this consists principally in rearing a son, who
must make offerings for his forefathers (Pitara/i), and in hospi-
tality; (3) Vaikhanasa, the hermit in the forest-hermitages (asrama);
(4) Yati, an ascetic, who has brought into subjection all that is
sensuous, and only occupies himself with yoga (meditation). They
are also called in India Satiny a si, and are saints. The Yati is
further called Sadhaka ; this word, which is also used as a title of
the Padandas, is not found in Wilson; but Sadhana means, among
other things, accomplishment, and good works, or the moral doctrine
and the ceremonies of the Hindu religion, as a means for attaining
purity and release from the transmigration of the soul. Sadhaka,
then, is " some one who performs these good works." In our work
the name Sadhaka is explained as "he who exercises the asrama-
dharma (the right or custom of the hermits)." We have, more-
over, explanations of D harm a and Adharma (right and wrong).
Further, Srnti (revelation) and Smriti (tradition) are mentioned.
The former is like Dharmasastra (the book of instruction in
right), and is said to be dh'iragnihotrav'ijya (rijya, derived, begin-
ning fiom; dhlra, agni. hotra, the maintenance of a continual
fire) ; it thus lias reference also to the fire-worship. The smriti is
the raruasrama-acJiara (vanjasrama is here perhaps an error for
vanasrama, forest-hermitage) ; thus " the mode of life, the example
of the forest-hermitages/' To these hermitages the ancient tradition
was handed down. The seven Rishis shall teach the Dharmadvaya
(the double right ; Dharma and Adharma (?) or rather Sruti and
Smriti} on the earth in the time of Ptirva Mann ( = Svayam-
bhuva .?). The Dharmadvaya is the contents of the Chaturveda
(the four Vedas), according to which the Sadhu (the good man)
regulates his life. It contains also the Daiidaniti (the regulation
of punishments), the Trayi (the three Vedas without the Atharra-
rcda ; we had just now all four Vedas, but these inaccuracies often
recur), and the rarnah-ama (here also varna takes the place of
vana). Does Tarna perchance convey the idea of a caste ? and
* He is here the Gnnt ; the deity (Siva) appears in Java and Bali chiefly in
this character. 1 [ ? ].
THE ISLAND OF BALI 121
the vartjAframa mean the regulation of the life (the four periods)
of all castes? 1 In general the passage before us speaks only of
the Brahmans.
The word Sruti is explained by/, "ri denian kinatuturan,"
because she is learned, she is called Sruti. Smriti, makanimitta ri
kangen-angenira matangian Smriti ngarania, on account of her
amiableness she is named Smriti. The name Smara (love, the god
of love) and a signification of the root smri, "desiderare, to
desire, to like," have obviously led to the latter explanation.
The characteristics of the Satya-BrahmacJiari (of the true
Brahman pupils) are : Tapah (penance), Mono, (Sanskrit mauna,
to be silent), Yajna (to make offerings), Daya (the receiving of
alms), Kshama (patience), Alobha (freedom from desire), Dama
(subjugation, i.e., of sins), Sama (repose of the soul), 'JitatmCi
(victory over the atma, the passions), Dana (almsgiving), Anamah
(not to greet ; this is forbidden to the Brahmans, because they
would lower themselves by bowing their heads, &c.), Advesha
(freedom from hate), Araga (freedom from affections). He is
rirakta (freed from sensuousness), tyfiga (relinquishes all earthly
power and gives them away), vijndna (knows the differences, has
things to discriminate, or shrewdness). These characteristics are
called dJiarmapratyangalaksJiana, the names (or signs) of the
organs of the Dharma.
Iti-uktang tatvancha sarvang, dharmapratyangalakshanang.
So is the whole Tatva (dogma) related, which contains the
names of the organs of the Dharma.
This is not the place to speak more fully of the contents of the
BraJunaudapurana ; it is to be hoped that in a short time they
may be fully worked out. We will conclude with a note on the
contents of the Vedas.
The three Vedas (here also the fourth is not mentioned), Jtfig,
Yajuh, and Sama, contain four Stotras (stotra is praise), namely,
(i) Dravya- (Wilson, "elementary substance;" there are nine
kinds) ; (2) Guna- (the three qualities which penetrate all beings,
viz., Satva, reality, truth; Rajas, passion; Tamas, darkness,
ignorance, badness) ; (3) Karma- (the works, actions, the practical
part, the offerings) ; (4) Bijana-Stotra (this must be abhijana,
"family, race" (Wilson). It is thus the genealogical part).'-
RELIGIOUS CEREMONIES AND OFFERINGS.
The five daily offerings which the head (grihastha] of every
Indian family has to make, are not found in Bali. The Pamlitas
read (or mumble) the Vedas once a day ; the people make their
offerings and say their prayers on certain days. The religious
1 [The duties concerning the castes and stages of life are meant.]
- [Probably meant for vijndna, discernment, or the intellectual part.]
122 AN ACCOUNT OF
ceremony consists in bringing offerings, which are offered with a
sembah and deposited in front of the small temple-houses ; in
the ordinary temples this is mostly done by women, to whom
is also entrusted the preparation of the various offerings. The act
of offering is connected with mantras (forms of prayer), which r
however, are not said aloud. On great feasts the offerings are
presented by "tandakking" persons inspired by a deity, ivawalen
or prakitlit (see " Us. Bali," p. ^^balian-waivaleii). These
offerings are coupled with " tandakking " J with the creese, which
indicates that the offerer is ready to offer up his life. The offerings
are very numerous, and are specially appointed for each feast and
each deity ; one usually sees various preparations of rice, cooked
meats, fruits, sirih, kwekwe, money (Chinese " cash ''), clothes,
and also drinks.
Bloody sacrifices are made to Durgd and the Biitas ; they are
usually confined to hens, ducks, and young pigs, although in
great sacrifices, buffaloes, goats, deer and dogs are slaughtered.
The persons offering the sacrifice eat the flesh of the three first-
mentioned animals, and only offer the skin and bones and portions-
of the flesh, boiled or roasted ; they also dress the flesh of the
dogs (sasatt}, but, according to their own statements, they do not
eat any of this, the whole animal being intended for the Juntas.
On the feast of Bayakala every family kills a pig, and the skin
and blood are deposited in the courtyards of the houses for Kala
and the Bfitas? The flesh, however, is in this case also used as
food by the persons making the offering. Besides these offerings
the temple is sprinkled with holy water (toy a tirtci). This water-
is bought from the Panditas, who consecrate it by reading the
Vedas. The persons offering also use this holy water to purify
themselves, to rinse their mouths and to sprinkle their faces. \\'e
have in Bali, it is true, the toy a Sindu (water of the Indus), at the
foot of the Gunung agung in Basuki, and also a Ganggd in
Tabanan, but the priests do not regard this water as holy. They
know that these rivers are in Kling (India), and that they cannot
obtain the water thence, but this want is supplied by mumbling
the Vedas. Besides the Gangga and Sindu, all the great Indian
rivers are known in Bali : the Yamuna, Narmadd (Nerbudda ),
Kaveri (near Seringapatam\ the Sarayti, the river of Ayodya-
(Oude), and others. The holy water produced by the Panditas is
called Mrita or Amrita (Ambrosia, immortal, life-giving food).
This water lias Kusa-grass soaked in it. The Padanda also strews
A/A&7-grass soaked in holy water over the persons who make the
offerings. The ATwjfo-(or -ter/;tf-)grass is also used to lay the
offerings upon. Ghrita (Ghee) is likewise known here, but, for
want of milk, is made of gooseberries. The Panditas, however,,
also use sometimes the milk of cows to prepare this; it ought
1 [/>., dancing with gesticulation.] - [Sanskrit Mfifas,]
THE ISLAND Ol< HA 1. 1.
really to be made of the milk of a lembit putih, a white cow with a
hump ; these animals are sacred ; the common Balinese cow is
not sacred, although it may not be eaten by the faithful. In
Badong the Balinese are forbidden to kill cows, although it is done
secretly, but in other States they are slaughtered openly. In order
to be able to use the milk of these Balinese cows for Ghrita, the
beast is tied up for a fortnight, and is given nothing but Kitsa-
grass to eat ; this food renders the milk sacred, and especially
purifies it from the smell of musk which the Balinese cows have.
27/rt-grass is also used, but is not known to the ordinary Balinese
(see a passage in the RCunayanci). Madhu (honey) also occurs in
the offerings, and likewise perfumes, such as m'nyan, maja gawu,
and chandana (see " Us. Bali").
The bloody sacrifices, as we have said, are chiefly dedicated to
Kala, Durga and the Butas, seldom to Siva. In India Siva
demands blood almost entirely, but in Bali, as supreme deity,
he has taken rather the mild form of Vishnu (and Buddha). The
offerings, which on great feasts are numerous and of some value,
do not remain in the temples, nor are they all given to the
priests. The Padandas who conduct the service receive part of
the clothes and money, and the offerings of eatable things, after
being presented, are taken back with prayers and eaten at home
as Amrita, life-giving food, by the persons presenting the offering.
The Panditas besides this receive considerable sums of money
for cremations and for being present at the offerings made by the
princes ; this money is presented to them as an offering itself
(this also characterizes the Panditas as a kind of god upon earth).
Most of the Padandas return a large portion of the offering made
to them. The Panditas do not take any active part in the offer-
ings ; they indicate from their -books, before the ceremony takes
place, all the usages to be observed the quantity and preparation
of the offerings, and the way in which they are to be presented ;
but during the ceremony they sit motionless, as they do in the
domestic worship, mumbling the Vedas. By mumbling the
Vedas they draw the attention of the gods to the offerings made
by man, and cause the gods to look graciously upon them. The
place where the Pandita sits is a high framework of bamboo,
under a roof, and he has all the utensils for the Suryasevana by
his side. To the form of worship belongs also the Pradakshina,
the marching round the temple towards the right hand, which is
done three times ; it is only performed by the Panditas in the
spirit, while their bodies remain all the time motionless. I have
not yet observed this custom at offerings either, and it seems not
to have penetrated to the popular worship of Bali, the Polynesian
tandakking* appearing to have taken its place, while the priests
* Tandak) however, appears to be an Indian word ; tandti is one of Siva's
doorkeepers, and master of dancing and mimicry; tandaka is a juggler; tanda-ca
"dancing with violent gesticulation," as executed by siva and his followers.
i2 4 AN A CCO UNT OF
have retained a remembrance of the original form of worship with
Pradakshiua.
DRESS OF THE PANDITAS.
At the great festivals of the princes, at the cremation of persons
of high rank, and at the domestic worship on full and new moon,
the Padand^ has a special dress, very much resembling the
articles with which the Javanese idols are adorned. They
are clothed in the ordinary Balinese manner, the upper part of
the body being naked (see the idols of the Bataviaasch Genoot-
schap van Kunsten en Wetenschappen) ; the garment which
hangs from the hips is on these occasions white. On his head
he wears a red cap, which, however, may also be white ; this is
called Jatd (see the catalogue of Indian Antiquities, p. xxv.) ;
Jatd is otherwise the head-dress of Sivaites, but not a covering
for the head. Thisjata has some resemblance to the head-dress
of Dagawan Trinawindu (No. 145), and still more to the Glung
K lining of the Balinese representations of Siva. It is wider at
the top than at the bottom, and goes down lower at the back
of the head than in front. I counted thirteen annular divisions,
formed from the bottom upwards by narrow stripes of gold, which
run round the flannel covering of the top. Upon asking whether
these divisions and the number of divisions had any meaning, I
was answered in the negative ; but nevertheless I am disposed to
believe that they have some connection with those which I have
observed in the Merits (see above) in the temples. In the Merits
I have never as yet counted more than eleven roofs ; can it be-
that the Panditas assume still greater sanctity by having a greater
number of storeys in their Jata^ which I venture to regard as a
kind of Merit ? This, added to their name (Siva), does not seem
altogether improbable; their own Merii^ the Meru of the Brah-
mans as it were, would then be higher than that of the gods
adored by the people.
Along the edge of the fata, across the forehead, runs a band,
called in Bali Kesabharana, ornament of the hair, and in India
Mnkuta ; it is covered with gold, and ornamented at short
distances with S&ryakdntas * (according to Balinese ideas a sort
of precious stone). In the centre of this Kesabharana, over the
forehead, there is a Lilian, in the form usual in Bali, and of
crystal (or glass). At the extremity of the Jatd there is a ball,
supporting a Lin&a (ball and Linga being of the same material as
* Suryakanta (beloved by the sun), a. precious stone, often meaning crystal,
but really said to have proceeded from the rays of the sun. The Balinese
Xuryakanta are of crystal or glass, and will be formed of pitsakas of Majapahit.
The fact that this precious stone is used for the SAryasevana i.e., the worship ol
the sun is a further indication of the importance of this heavenly body. Another
precious stone of fabulous origin is the Chandrakdnta^ proceeding from the
condensed rays of the moon.
THE ISLAND OF BALL 125
above). The fact that we here find the ball ornamented with the
Linga r is perhaps again a sign of the intermixture of Buddhism
and Sivaism, in which, however, the Sivaitic symbol has the
upper place. The Pandita wears in his ears the Kundala ; these
are shaped like an egg, and are attached to the ear by a gold
ring ; it is sometimes of crystal.
For nearly all the remaining articles of his attire the name
dbharana is used, whereas in India each article has its particular
name. We must not confound Karndbharana (ear-ornaments)
with the earrings ; they are fastened behind the ear. tnu\-
bharana (ornaments of the dtmd, the soul or the breath), a short
band, worn round the neck, and on the breast on each side, with
gold clasps of considerable size. Vdyitbiiarana (ornaments of
the wind ; i'Ayu t the wind, is the vital principle ; the form vdyu-
l) ha ran a is incorrect, and should be vdyvdbharand]. These are
double ; they are worn round the neck, and are longer and fall
lower down on the breast than the Atmdbharana. Hastdbharaya,
as the name indicates, are worn round the wrists.
The guduha (or guduha genitri) is a kind of rosary ; we find
it in several of the Javanese idols j it is called in India akshamdlCi
(string of beads). The name guduha does not appear to be
Indian. The Pandita has two or three of these, and uses them
on great festivals while he mumbles the Vedas, telling them after
the manner of the Roman Catholics. All these ornaments, in-
cluding the Brahman-band, which is called sampat, saivit, silimpit-
are strings of the fruit of the plant genitri (black balls about half
an inch in diameter) ; their interstices are inlaid with gold, and at
intervals they are set with Suryakdntas.
The Brahman-band (not called here by an Indian name;
perhaps it would be a profanation of this sacred ornament to
make known to the people the Indian name Upavlta or Yajno-
pavlta .?) is threefold, but is not formed of three cords interwoven,
and not of Hfunja-grass, which name is unknown here. There
are three strings, passing over the left shoulder and under the
right arm, and fastened together on the shoulder. At the place
where they are joined there are three lingas (again of crystal).
The ends hang down, and are ornamented with large precious
stones one with a red stone, Puala (doubtless Upala, opal) ;
the second with a large (white) Suryakdnta ; and the third with
a black transparent stone, found in the mountains of Bali, and
called manik girang-girangan. These three colours have re-
ference to the three gods of the Trim Art i t Brahma, Vishnu, and
Siva (or Sada-Sira, Pramd-Siva and Mahdsiva). Red is the
colour of Brahma, white of Siva, black or dark blue of Vishnu
(and also of Kala). Besides this Bralnnan-band, the Pandita
has also a band of white cloth, six yards long by three inches
wide, which, like the other, is worn over the left shoulder and
i26 AN ACCOUNT OF
under the right arm. This band is also called sampat, sawit,
or silimpat.
The Pandita, finally, wears on his hands several gold rings
with costly rubies. The ruby is the favourite stone in Bali,
almost more prized than the diamond ; seven kinds are known
in Bali, each possessing a special supernatural power. On the
upper part of the arm and on the ankles the Pandita wears no
ornaments, but the gods do. The gods also have many of these
ornaments in the form of serpents, which is not the case with the
Panditas.
DRESS OF THE GODS.
The head-dress is very complicated, especially that of Siva.
Siva alone of all the gods has the Glung kurnng, a nearly
spherical cap, much resembling the Jata of the Panditas, whose
example Siva is. This cap covers the back of the head. On
the forehead is the Glung chandi, a round, lofty head-dress, nearly
in the shape of the chandis in the temples. Papudnkan are
ornaments next to the glung chandi ; Garuda mnngkur above
the papudukan. Patitis (among the Panditas Kesabharana]
is the forehead-band, set with precious stones ; in front is
the chtidamani. Mangle wijaya (perhaps mangala vijaya, happy
victory), the plain stripe above the Patitis. Kuudala, the
earrings. Skar taji\ the pointed ornaments immediately behind
the ear. Ronron, the ornaments behind the s'kar taji. Apus
kupak, the ornaments on the shoulders. Glangkana, the bands
on the upper part of the arm. Glang, the bands on the wrists.
Glang batis, the bands on the ankles. Bapang, the small garment
which hangs in two capes from the shoulders on to the upper part
of the arms. Kalung, a short collar ornamented in silk with
serpent-heads. Sampat, Silimpat or Sawit ^ also Genitri (after
the name of the plant of the seeds of which it is made), the
Brahman-band, worn, like that of the Panditas, over the left
shoulder and under the right arm. Babcdatti, the breast-band,
worn round the breast under both arms, and fastened in the
middle with a jewel ; this is also found on many Javanese idols.
Naga wangsul (the serpent of Bali), a large band hanging from
the shoulders to the stomach, formed of a serpent, worn in this
way as the Brahman-band. Babadong (badong, the name of a
kingdom, means, like bandhana, the Sanskrit name of it, "con-
nexion "), the upper band round the hips. Tambcdaua, two other
bands worn round the body, below the babadong. Kambcn, the
cloth (not a sarong), unsewn, and fastened above the hips by the
bands just mentioned. Saniir, an end of the cloth, rolled up and
sticking out above the Tambcdana. Jawat, the end of the cloth
which hangs down in front between the legs. Muchcr, two ends
of the same hanging at the side, dialer, short breeches (these
THE ISLAND OF BALI. 127
are never worn by the Balinese, and are only seen in represen-
tations of mythological personages). iva and all the real gods
sit or stand on a lotus-bed, padmasana (see, above, the Padmdsana
in the temples as a seat for Siva). They have also a glory, sur-
rounding the whole figure, called here simply trans * c j a i " the
bright lustre." All these ornaments can only be clearly explained
by means of drawings, and I therefore caused the principal
Balinese gods to be drawn by a Brahman ; the first of these
drawings which I sent were lost on the journey; those which I
sent on the second occasion at any rate reached Batavia.
The costume just described is that of nearly all the gods ;
Siva wears only the Glung kurung in front. This monotony also
indicates that all the gods are merged in the one Siva. In the
female images we find other ornaments for the ears, and the cloth
hangs down lower, as it is worn by the Balinese women. Unta,
Siva's wife, has ornaments stuck through the lobe of the ear,
named Subong; they are similar to those of most Balinese girls,
which are made of lontar-leaves rolled together. She also has
the Glung chandi of Siva. Sri, Vishnu's wife, has the same
ornaments, but a simple head-dress, called *Mengure glung.
Vishnu's earrings are called rumbing, and are round and without
a cavity.
It is especially to be remarked here, that the names of the
articles of the Panditas' dress are Sanskrit, while the names of
those of the gods themselves are Balinese. The gods are thus
less sacred than the priests ! But the deity may not and cannot
really be represented; drawings and images have no value but
for the ignorant.
FEASTS.
These are partly feasts of the various gods and temples, and
partly expiatory feasts. Those of the first kind are dedicated to
the more beneficent gods, the others to the Butas and Riiksasas,
with Kala and Durgd at their head. Especially grand feasts are
celebrated in the sadkahyangs, the six most sacred temples in
Bali. In the temples of Bdsuki (or JBesakt) at the foot of the
Gunung Agnngj the feast-day is every lunar year on the full moon
(Puniama] of the month Kapat or Karttika. This feast-day is
really valid for all Balinese ("Us. Bali," pp. 273 seq. and 346) ;
but on account of the existing hostilities, Badong and Tabanan
have not for a long time performed their worship, because the
temple is in / Karang Asem. In Basuki, Mahadeva or sang
Purua Jaya (Siva), whose seat is the Giinung Agung (or Meru),
is worshipped.
In Badong the feast-day in the great sanctuary of Vluwatu is
on the 2ist day of the new (Balinese) year, anggara kaliwon, in
the week Madang Siha. On this day the princes of Badong
128 AN ACCOUNT OF
worship there, and to do this have to travel over the mountains
to that holy place by wild and rocky paths. The holiness of
this temple is ascribed to the fact that the ship (the pra/ni)
of Dewl Damt^ Mahadevtfs sister when she came from India,
stopped at that place and turned into stone. This temple, how-
ever, cannot be so old as that of Basuki : it is not named in the
" Us. Bali," p. 320 ; and it must further be taken into consideration
that the feast clay is Jioced^ not according to the Indian, but according
to the Balinese year; while, on the contrary, in Basuki it is fixed
according to the Indian calendar, and this seems to indicate that
the temple of Badcng is of later date. The same thing applies
also to the temple of Sakennan or Serangan, dedicated to Indra.
Here the feast is on Rediti Manis, in the week Langkir, the
twelfth day of the Balinese new year.
Each temple has likewise a feast on the date of its foundation.
This is called wedalan, commonly pronounced odalan, or anniver-
sary. General feasts of the gods and Pitaras are celebrated on
Gahmgan, and in the five succeeding weeks (see "Us. Bali"). At
this time the gods are supposed to dwell on the earth, and the
Pitaras especially return to the bosoms of their families ; hence
the constant offerings and the incessant games and amusements
which are regarded as necessary less for the living generation than
for the Pitaras and gods sojourning among them ; hence also the
cessation from work and the disinclination to intercourse with
foreigners during this period. Trade and foreigners are not agree-
able to the Pitaras, who desire to see old institutions and usages
faithfully preserved. The princes also have feasts on their birth-
days, and on the anniversaries of their temples. The number and
the preparation of the offerings is minutely prescribed for each
deity and each temple.
The expiatory feasts, however, are those most worthy of atten-
tion. They are, in great part, celebrated, not in the temples, but
in the inner portion of the houses (iiatar), or in places arranged
for the purpose. We have already mentioned two great feasts :
Bayakala (see " Us. Bali," p. 323 sqq.) and Panchavalikrama. This
feast is not kept on fixed days in the year, but on great occasions.
It occurs after the conquest of a State, for the sake of the con-
quered population, who are thereby delivered from the evil
influence of the demons, who have power over conquered places ;
it also takes place before the abiseka (anointing) of the Sovereigns,
and it is celebrated by all the princes and the men who bear arms.
It is further observed after a contagious disease e.g., the smallpox.
It is necessary that five Padandas should be present, four seated
facing each of the cardinal points and one in the middle : one of
the five must be a Padanda Buddha (a Buddhist priest), who sits
facing the south. We have drawn attention above to this pheno-
menon, and may here observe that although Buddha is also repre-
sented at this feast, he plays but a subordinate part. To the north
THE ISLAND OF BALL , 2 >
is Vishnu, whose colour is black ; to the east Maheh'ara (white),
to the south Brahma (red), to the west Mahadera (yellow), and in
the middle Siva, with mixed colours. (The offerings are of cor-
responding colours.) In the middle is the Padanda of the
supreme Siva, and he / naturally has the chief place. There are
three other priests of Siva besides, but only one Buddhist.
Another expiatory feast is called Ekadasa Rudra (the eleven
Rudras ; Rudra is a subordinate form of Siva ; eleven Rudras are
also mentioned in Wilson). The origin of this feast, however,
seems to be known to but few Balinese. (See " Tijdschrift van
N.I.," VIII. vol. iii. p. 242, in which passage the name yajna,
through a printer's error, is written jadjoeja.} The offerings
enumerated there are the greatest known in India, but, according
to the descriptions which I have obtained, they are not organized
in Bali and Lombok precisely in the Indian manner. I was told
that there were no bloody sacrifices in the feast in Lombok (Sep-
tember 1846) ; simply large sums of money were distributed
among the priests and the people, and the ordinary offerings were
presented.*
It appears, therefore, that only the names of these feasts have
survived in the memory of the Balinese, and that the latter, in a
spirit of vain-glory, apply these sacred names to their curtailed
feasts, which they have not the means to make very grand.
It is a surprising fact, however, that by indirect inquiries I have
convinced myself that a few human sacrifices have actually taken
place in Bali ! A former prince of Karang Asem, who was defeated
in war, put one of his slaves to death in a forest, and then placed
his body, concealed by clothes, among the other numerous offer-
ings, as a means of imploring the gods to restore his power. The
deed was discovered, for while the Pandita was mumbling the
Vedas, a wind arose and uncovered the body. A curse fell upon
the presumptuous prince, and he never regained his power.
Another instance is known from Gyanyar, where the prince (pro-
bably the first Deva Mangis, said to have been changed into a
serpent on account of his atrocious acts) had set aside a slave f to
be sacrificed ; he intended to murder him in the darkness of the
night, but killed by mistake his own son, and roasted and offered
him as a sacrifice.
These two facts cannot well be gainsaid, for I have obtained
the accounts of the former not only from Badong, but also from
Mengtii, which is allied with and has long been friendly to Karang
Asem. The people of Karang Asem utterly deny it. Of Gyanyar
* For the reader's convenience I will repeat the names : \.Ahamedhayajna J
the horse-sacrifice ; 2. Gomedhayajna, the cow-sacrifice ; 3. ManusJiyayajna ,
human sacrifice ; 4. RAjasftyayajna, the sacrifice of the universal prince (i.e.
offered by him) ; 5. Devayajna\ 6. Rishiyajna; 7. Mutayajna ; 8. JKcwyasa-
yajna', 9. Rajabusanayajna.
t Gnling buntitt is the name for human sacrifices.
SECOND SERIES. VOL. II. K
130 AN ACCOUNT OF
it is a common saying. How far this barbarity has gone, and,
perhaps, still goes on in a thickly populated country, where the
common man is a slave and of no value, I will not venture to say.
The burning of widows, and also the amok on the fall of a State,
must be classed among the human sacrifices.
The general name for these expiatory feasts is prayaschitta,
commonly pronounced inaccurately prayas-tista. The word means,
according to Wilson, expiation, penance ; and thus is very appro-
priate. The Panchavalikrania is expressly included under this
term. To these also belong the purification of a house in which a
corpse has lain. The feast of the Sarasvati, on Sancschara, in
the week of Watu gunnng, has something in common with this ; on
that day the collective books of the princes are carried into the
temple and purified for the coming year, through the priest reading
the Vedas. Another feast is for the weapons, the ceremony being
the same as in the previous one, and at this a bloody sacrifice to
the Butas is also necessary. Yet another feast is observed for the
welfare of domestic animals, cows, horses, pigs, fowls, &c. Among
the feasts belong also the cock-fights, not only as an amusement
of the people, but also as a religious ceremony. At the feasts of
the great temples e.g., the temple of assembly of the Gusti Pam-
chuttan every one belonging to the congregation must send at
least one cock, and must make it fight, either himself or through
a deputy. This custom is based on the supposed incarnation of
Vishnu as a cock (Silingsing] in Bali, but we can explain Vishnu's
assumed incarnation more satisfactorily as an apotheosis, not de-
rived from India, of the cock-fights which are so popular in all
these islands.
FURTHER DETAILS OF THE RELIGIOUS WORSHIP.
The mangkii (see " Usana Bali," p. 267 sqq.} is the guardian of
the temple, who superintends the edifice, and partly performs,
partly conducts the presentation of offerings ; he must know cer-
tain mantras e.g., patikelaning genta sapta and sastra sangha, and
must employ them when offerings are made. Both men and
women can fill this post. (The Brahmanic women those sprung
from the Brahmanic caste, not the concubines from the lower
castes are likewise acquainted with the Vedas, and perform some
religious functions instead of the men.) The mangkus can further
be of various castes, and not Sitdras alone ; I know several Gnstis
( Vesyas] in fiadong, who fill the office of mangku : usually, how-
ever, these latter hold the post in the chief temples, such as, for
instance, the domestic temples of the princes. I have not yet
found a Brahman acting as a mangku ; although perhaps there
are some who do so. The Brahmans generally aspire to the dig-
nity of a Padanda, and therefore look down upon the position of
THE ISLAND OF BALI. 131
a mangkn. In order to become a mangku it is necessary, or at
any rate usual, that a deity should pass into, and thenceforward
speak from the body of the person selected. Two cases have
become known to me in which young Gustis appeared for a time
almost to have lost their reason, speaking an incomprehensible
language, performing none of their actions in the ordinary manner,
and sleeping in the temples. These Gustis were thereupon
observed by the persons who had been longer inspired by a deity
(wawalen or prakulif), and after due inquiry, acknowledged as
also inspired. Such individuals are supposed to be either wawa-
lens or actual madmen. When they are thoroughly penetrated by
the deity, and have become calm again, they are true wawalen,
and are able to state what deity it is that inspires them. They
are then regarded as the most devoted servants (inaugku) of the said
deity. They become madmen if their minds do not become calm
again or rather if their deception is not properly carried out ; for,
of course, these wawalen must be regarded as idle impostors, who
choose to lead an easy life, chiefly at the expense of the credulous
populace. (Compare on this subject " Usana Bali," pp. 268 and
335-)
These temple-servants, however, detract nothing from the im-
portance of the Panditas. The latter, by their life and the study of
the Vedas, are identified both spiritually and materially with the
supreme Siva, whereas the wawalen have merely, as it were, given
their bodies to be dwelling-places of the deity. In the wawalen
an unconscious, in the Padandas a conscious, unity with the deity
has commenced. In the great temples, therefore, the mangku
and wawalen are of little consideration ; all ceremonies at the
feasts are regulated by the directions given from the books by the
Padandas, and the latter mostly use for this purpose mantras
which are unknown to the maugku. Moreover, the Pandita alone
(and not the wawalen} is able to call down the gods by mumbling
the Vedas. The sayings of the wawalens, when, at the offerings in
the temples, they give utterance to the voice of the god which
dwells in them, are compared with the statements of the sacred
writings of the Padandas, after which a decision is arrived at as
to what must be done in important cases (sickness, wars, &c.).
Thus, in these cases also, the Padandas have an opportunity of
exercising their power ; in the first place, they can counteract, by
quotations from their sacred works, the utterances of the wawa-
Icns, if they do not meet their views ; in the second place, they
can easily make known to the wawalens, whom we are disposed
to regard as deliberate impostors, what they must say, and, by
their approval, elevate these utterances into laws for the guidance
of the people.
Archa (Sanskrit archa, an image) is an idol, usually of stone, in
Bali always of rude workmanship. Sometimes it is supposed that
K 2
132 AN ACCOUNT OF
the deity passes into such an image, and this fact then induces the
faithful to bring their offerings. (See " Us. Bali," p. 274.) This
belief, however, is for the most part held only by the common
people. The priests, and also a large portion of the people,
attach little value to the images. "Does not the deity dwell in
heaven ? " was an idea expressed by a common man in Boleling.
The idols, as we have already said, are called togog or tongkok,
which means neither more nor less than "doll." We find them
principally as watchers, in the form of Raksasas and Biitas ; for
the same purpose there are also small watch-houses, called Tugu,
before the temples, in order to prevent the evil spirits from enter-
ing. We also find images, such as those of Ganesa and Vasukij
and also of Hamiman and Gariida, in some of the small temple-
houses, representing as it were the retinue of Siva. Garmla is
always monstrous, with Raksasa tusks. The Nandi is very well
known to the Balinese. but I do not remember to have seen repre-
sentations of it. The worship of images therefore has, upon the
whole, penetrated but little into the belief of the Balinese, and we
have here an obvious contrast to the creed now existing in India,
where the highest value is attached to idols. I have not yet seen
any representations in stone of Siva, although I have met with
some rude images of Vishnu (the subordinate incarnate deity). In
pictures we also see Siva as fsvara and Mahadeua (when he
appears as a youth), and as Arjuna, but these pictures are not
worshipped, and, like those of Vishnu, as Rama, &c., have refer-
ence rather to the myths in the Kavi works, than to religion.
The offerings are called banten, charu, or aturan. Religious
fasting, which the priests and those who desire to become Rishi
practise daily, is called mavinten. The cap of the Pandita, jata,
is also called bawa. The flowers which he uses in his domestic
worship are chempaka putih, chempaka knning (Sanskrit champaka ;
the a is here also altered by the nasal pronunciation into a short
e),Jepon, Kenyeri, Ergani, Jenpiring, &c.
The vilva fruit and the salagrama stone, which play so impor-
tant a part in the Indian worship, I have not yet met with in
Bali.
The lotus has really no worship, but it is found planted in pots
in the Padandas 1 houses, and also in the ponds which the princes
are accustomed to have in their palaces.
The Brahman-band (upavlta} belongs in India to the three
upper castes, which are called on this account dvija, "born
twice" (the second time through adopting this band). In that
country it is of different kinds, according to three different castes.
In Bali it is found only among the Padandas, and then only if
they are in full dress. But the Kshatriyas and Vesyas, and even
XttJras, who have obtained permission from the Padandas, also
wear a protective band, a sort of amulet, in war, called by the
THE ISLAND OF BALL 133
same name (Sampat} ; they only wear it, however, in time of war.
In Badong I have never met with it. Chandra or unilan tu-
juanggal (the waxing moon) is in the head-dress of Ganeta, as in
the Javanese images of that god. The skull (mw.nfa] above the
fhandra is not known to the Balinese ! They have a milder form
of religion, with which skulls and chains of skulls (found with the
Indian Siva and Gaye4a t and also in Java) are incompatible.
Siva's name, Kapalaltfirit, "he who wears a skull," seems to be no
longer clearly understood in Bali. The mark on the forehead
found in some gods (eg. Ganesa) not to be confounded with
the third eye of Siva and Indra is called chuundung ; it is also
marked on the foreheads of the princes when they adorn them-
selves for a religious service ; its meaning is unknown.
The King Siva-Buddha is named in the Ravgga Lawe as ruler
of Tumapcl. His kingdom was overthrown by the people of
Daha (Kediri}. This King's name cannot well be anything else
but an indication of the state of the religion. At that time
Buddhism and Sivaism must have been completely blended together.
The fall of the kingdom, then, appears to represent a reaction,
brought about by the Sivaites ; and it is the more remarkable that
this kingdom, with a mixed religion, was conquered by that of
Kediri, where the flower of the Sivaitic priests and learned men
were to be found under Aver Langgia and Jayabaya, although
the latter (fayabaya} especially also tolerated the Buddhists (see
under the article Literature]. From Kediri also the orthodox
Sivaitic Brahmans in Bali trace their descent. We have already
drawn attention in various places to the intermixture of Sivaism
and Buddhism. We will give here a few facts bearing on this
point. We found an image of Buddha in a temple of Siva in
Bolelcng ; on the jata of the Padandas there is a ball under the
liuga ; and, further, inverted pots or glasses are found on the
small temple-houses ; in the " Usana Bali " we have among the
gods, and also in Jayabaycts retinue (in the Bdrata- Yudda], the
Jtisi Seva Sogata (Saiva belonging to Siva, and Saitgata to Sugata
or Buddha), ffisis /holy beings partly in heaven, partly on earth)
of the worship of Siva and of Buddha ; in the " Usana Java" we
have Siva, Buddha, and Bujangga as sons of sang Haji (the
original one).
The Buddhists, the Balinese assert, came later than the
.Sivaites, and if this refers to their arrival in Java, it agrees with
the accounts given in Java, where the Buddhists are also said to
have come later than the Brahmans. What has happened in
Java is very often confounded, even by the priests, with occur-
rences in Bali. These nations are particularly weak in chron-
ology and geography. In Raffles, Appendix K, p. ccxxxix.,
there is an account of Bali by a Muhammedan : it states : " The
religion of Buddha (under which he evidently includes the entire
1 34 AN A CCO UNT OF
Hindu religion) is divided into Sakalan and Niskalan. The first
division will include all earthly things, and the second the re-
ligion." This division exists in the writings (/;//?/?-) of the
Balinese, but it has no reference to religion ; sakala is that which
belongs to time, and niskala that which belongs, not to time, but
to eternity, to the period after death. 1 The former, therefore, is
correctly interpreted by the narrator, but the latter has a wider
meaning.
The following are names of the ornaments of the gods :
Anting-anting, like kuudala, earrings.
Glaiig Kupak, instead of alang, bracelets.
GitduJia pawilangan, the rosary (wHang, to count).
Parmaia, a golden girdle (Incl. mekhald).
J\ilat balm, the necklace (with the Padanda atmabliarana).
Babandong, the longest band round the neck (vdyubhara^a).
Chechandian, Chandi repeated (the glung chandi}.
The chakra is also to be found in the place of the geuitn (gu-
duJia genitri ), a substitution which in some cases can easily occur
with the Javanese images as well.
The temples in Bali are of the same class as those of Majapa/u't,
or of the third period : in Cravvfurd, " Ind. Arch.," vol. ii. p. 205,
' temples constructed of brick mortar." The bricks of which they
are built are not of first-rate manufacture, and consequently not
very durable. Stone carvings, which are plentiful in the ruins of
Majapahit, are only to be found rarely in a few ancient temples
and palaces (the best in Mengiti ; and in Klongkong 2X& Gvauyar}.
This art has evidently not advanced in Bali, and there are now
very few men who can work in stone, and even these no longer
produce anything of beauty. The art of carving wood, ivory, and
whalebone has been tolerably well preserved ; neat images of gods
and Raksasas are carved out of the last material.
Crawfurd and Raffles maintain that the religion of the people
in Bali cannot be called Hindu ; but, from all my inquiries into
the worship of the common people, which at first I was inclined
to regard as a remnant of the purely Polynesian age, I have con-
vinced myself that this also is 7 Hindu, and that the low estimation
in which the temples of the Siidras are held by the priests has
merely obscured the objects and corrupted the mode of this
worship. These small temples are regarded as puuggawas (sub-
stitutes) for the large and principal ones, just as the Sndras who
pray in them are vassals of the men of rank who pray in the great
temples.
Another difference finds expression in the saying, that not only
in such temples, but also in the small sanggars of the upper
] [R. van Eck in " Tijdschrift voor Nederlandsch Indie," 1879, i. p. 57-1
THE ISLAND OF BALL 135
classes, Ci Siva's children " are worshipped. Who these children
are, however, is not clear; someBalinese names e.g., san^ Kasufiun
Kidiil (meaning he who is " worshipped in the south"), BraJiithi
or Mahadeva-&Q not afford us any explanation. It would appear
that here also we must have recourse to the difference of the
castes ; as the insignificant temples alone belong to the Sudras,
so they have no claim to the worship of the Supreme Siva. Only
the subordinate manifestations of Siva, called for want of a better
word, " Siva's children," are intended for them.
Besides these children of Siva, the Pitaras, the shades of the
dead, are also worshipped in the small temples, and we shall show
that these may be regarded as identical, as it were, with " Siva's
children." We have already seen in connexion with Indra, that
it is not easy to reconcile the decrees as to becoming a Pitara
with the punishments to be imposed by Yama. According to
some, all the persons who have undergone cremation dwell as
pitaras in Svarga or Indraloka, and there enjoy eternal happiness ;
according to others, they wander about for a long time before
reaching that place, and assume various forms upon earth ; and
finally it is also said that, although they enter Indraloka, they are
obliged to descend again to the earth as human beings ; it is not
said when this obligation ceases, and their state in heaven is called
indiscriminately deva, hyang (god), or pitara. According to the
popular belief, Brahmaloka or Vishnuloka (which are identical and
higher than Indraloka], and Sivaloka, the highest of all, are not
attainable. The Brahmans, however, appear not to share this
belief, and lay claim also to these supreme worlds and the moksha
(deliverance from the transmigration of the soul). Among the
"children of f Siva" we have also to include such ascended souls,
dwelling in Sivaloka, although it is surprising that they have not
become identical and one with Siva (who here is the same as the
Indian Parabrahma}, as the word moksha indicates.
According to the accounts of men of the lower pastes, it is
supposed that, in the worship of the Panditas, not Siva himself,
but Bagawan Byasa, enters into the Padanda ; the latter then
becomes like Byasa, and possesses the divine power of the
DevarshL Upon the whole, this statement is almost identical
with the opinion which we have already expressed. Byasa is, as
it were, the same as the deity (Siva), and, as we have already seen
(under Literature), is called Yogisvara, a name for Siva himself.
Into the wawalen, to whom we have before alluded, there
passes, according to the same accounts, z.Buta kaparagan (an em-
bodied Buta\ who is said to be the confidant of the god and to
know his secrets. This spirit, therefore, and not the deity itself,
speaks from out of the wawalen. We may be surprised that a
Bftta should be called here the confidant of the deity, yet the
conceptions of a Buta, a Raksasa, and a Deva are so mixed up
i.?6 ./.V ACCOUNT OF
m tin- munis el the lower order;,, th:it we cannot attach mm h
ini|i(irt;iii i- to llir, !.i< t , .1 IK I. bcsidd this, the /'///^v and /
ully as rlosely jelatcd to ea< h other ;is /\',i/,i to A/Vv/,
.mil the Klves to the l-'ays.
The ,v,/,i',//V/M aie al:,o the pi i halian t/ifkiin);* when
they a< t in this < -apa< ity, they re< He Miintnis (>na\y,a}, moving
their bodies as the I'anditas do when they nmiiiMe the \
In addition to the nuintnis, the A/r/y (the wind) is the remedy
employed by the halian ; he causes his breath to pass into tin-
bodies of the sick.*
Me who performs ta[ni lives only on the -.wr//, without any
other food. In the " I saua Km," liowever, we have seen that /<//><!
(penamc) is now no longer performed in llali.
HIS.
Religious i ites analogous to the |oine:,lir worship of the J\n/un
d(U are peiloimed also by certain pnmrs and other DcVtU and
(iiisfis, ill order to attain the diynitv of a /\'/.\/. These \>-
also peilorm a BCrvicC eVCrV momm:-, on a la:, tin;- :,!oina< h, USUlgi
however, not the Vedas, DUl the Mantras, namely the Mantra
f'asiti/i/ati (tlie pnrilyinj; Mantra}. They peiloim ablutions with
holy water, wash out their mouths, ptirily their teeth, and wash
then han ; not until they have done this do they dress and appear
iu public. When, by means ol this daily BCfvicC, which must be
< 01 1 pled with a \ cry i <-"iil.ir mode ol life ( t hey ma y not lie, n
or do anything <lejralin^, iVe. ), they have atlained a <eitam holi
they beiomc l\i\i\. I lus BtatC <>l A'/\/ has much m
with the third peiiod of the hlelime ol tin- Indian:; a< < ordiii:',
to Mann, wheie they live loi medila! ion alone, and, withdrawn
liom all earthly allaus, pass a peaceful lile, pleasui;', in ( iod, in the
fores! hei milages. In I5ali,also, a piinee who has In-come a /\W
must reliinjuish his kini'/lom to his children, and, intent alone
upon his heavenly state, sepaiale himself from all human < oiu
panionslnp. ll SCCmi, however, that al present love lor earthly
i. nl more weij'.ht than < are about heaven. The last prince
but one of runichnttan in /V//Av/;;, Ngitni/i d"</f ranic/inttan had
be. ome a AV\/, yel he ivtauied his kini-dom until his death. The
old A'li/ti Ka^iinan has peiloinu-d the said BCTVice (\\hih LI in
COlTeCtly ailed also i/uti'fi/tt or //lasiin'tiscrtnia} lot
ind yet he has not become a /\'/A/, and .',emr, lo keep that
the piinccs of rannimttan and
1 [T; 'nkiin upplic-s IM l>otli lexei, (lie I '.iy.il. A;.'.,;-/ !< i
uiiy. See i i;mk-l.iiiii\ " Grammatik," p. 109 I
In llir In, .iiinl; ninl /././', innli-isliiiitlin
tie three propertf . .'// (or // ////.// //) ii m out.
ol It.
////: /S/ AND OF KM. L
/will no longer acknowledge him as their guardian. The
wife of ;i AW can alone follow him into the state ol srp.natioii.
|>in\ided that she, too, has performed the d.iilv MfVicCi and, like
linn, has remained lice from all misdeeds and sins.
This MTVke makes the I'.rahman (Mr) a I \uhim1n (where.
however, other regulations also come in), and the ks/nitri\:r and
Wtsyai A'/.v/ ( A'<r/<rrs//t\ royal /v'/.v/, to be distinguished liom the
' v/'/\, di\me AWs, and ihe />Vw/////^/'\///v, the Urahmanical
A'/.v/H A .sW/'<7, finally, becomes through this servi. e a />///////.
'I he Trimiirti or Trii<ikti (trinity) is contained in the word
r'//,V (really Indian <>///), which is Immed of three letters, </, //, ;//,
or, as tlu- r.alme'.e say, of (ii/x, ////;;, ///////v, meaning .SW</.-
/(//v//// ( r.S'/Vv/, and ..Miilki\i",w ; or Brahniii, Vis/iuu, and .S'/Vv/,
also represented as */.v/// (lire), AM,/ (\vat<-r), and fl^rVi (air). In
this combination Siva is cooidmated \\illi I'.rahma ai.d Vishnu,
but he is called .!A///<M/Vw, the -real Siva, and the oilier two gods
bear his name, and must be regarded, as it were, as t '.\-/>rf<siiws <>/
himself. Siva, a I . \ : being in the centre,
with ilrahma on his right hand and Vishnu on his left.
This trinity (also called triiiikti) repeats itself throughout
Nature. We have the /n'M//r<t//<r (the three worlds), consisting
of />////;-, bhnvah, and svah (earth, air, and space). And further,
lnnit) //V////77, and mini/tn (heavt n, earth, and hell). Three
kinds uf human beings: piiriiinpinvan, /<r/ci, and hanji (woman,
man, and hermaphrodite). Tlie last kind also appears in Siva as
Arddkanare&VQrt (Siva with his wife in one body).
In the month of AV^w//.^/, in which the lunar year of months
begins, all the llalinese keep the feaxt of Mn/twitr or l.abu
rini/tt : A'tt.wHyr is also called by its Indian name, (Vis/Hr, and,
as in Jmlia, begins the year.
CREMATIONS.
( )ne of the most important religious ceremonies is the cremi
tioil of the dead. Only through the burniii:; of the bod
jiossible for the soul to enter the heaven of Indra /), and,
thence, that of Vishnu and the supreme heaven of Siva (I'islniu-
/t>/c<f and 5nM/0Xv/). The doctrine of the h lion of the
soul exists in IJali, although the 15aline.se do not speak of its
details; the- titintif>mtan*ti is a work on this subject (explained
as kunif*nlnn bmliin, the assumption of various bodies ; according
to the Indian words it seems to be "glorification of the soul," a
jx>int which its contents leave in uncertainty). This work, how-
ever, is a popular one, written in K'uiun^ but it does not appear
138 AN ACCOUNT OF
to contain the whole of the purely Indian doctrine of the trans-
migration of souls. Other writings of the Brahmans exist on this
subject, but they are secret; it is according to these that the
ceremonial at cremations is regulated.
It is believed that cremation, and the offerings which precede
and follow it, exempt a man from any further change of shape
at least he remains for a certain time as a pita in the heaven of
the Devas, and as such demands the worship and offerings of his
surviving relations. Cremation requires a considerable outlay, 1
and therefore every family is not in a position to show this
honour to its dead. Instead of being burned, they are in that
case buried, the body being inclosed in a case of bamboo, which
completely conceals it, and covered with a number of clothes;
it is carried out upon a bier amid the singing of naenia, which,
among the common people, are trivial compositions. Arrived at
the grave, the body is stripped of its clothes, and let down with
the bamboo covering ; a few small coins are then thrown into the
grave (in order that the dead may provide himself with food), and
it is closed. By the side of the grave a bamboo is fixed in the
ground, on the top of which there is a sort of three-cornered
hutch of lattice-work, in which offerings of small value, chiefly rice
and flowers and fruit, are deposited immediately after the funeral,
and subsequently at certain intervals. These offerings are for
the purpose of propitiating the Biitas (the demoniacal beings
who infest places of burial especially ), lest they should attack the
soul of the deceased. The grave is then surrounded with a fence
or hedge.
Those who are buried in this way cannot enter heaven ; they
then assume all sorts of shapes, and it seems that the Balinese
especially believe that the numerous dogs which wander about
half wild are metamorphoses of Siidras (the lowest caste) ; for
this reason they hardly ever kill a dog, and these ugly animals
increase and multiply enormously, and are a pest to the European.
It is now the duty of a member of the family (son, grandson, &c.,
sometimes even the third generation, if he has become rich enough
to afford a cremation) to cause the bones of all his relations who
have been buried to be exhumed and to burn them together.
At most cremations, therefore, one sees a number of bodies at
once, each in a special coffin, many of which have usually been
buried for years. In times of general calamity, contagious
diseases, &c., all bodies, even those of princes, are buried and
not burned ; it is not permitted then to keep any corpse above
ground, for in such times no work pleasing to the gods can be
undertaken, and the influence of the demons is in the ascendant.
At these times, also, the Galungan is not celebrated.
1 [According to Jacobs, 1. 1. p. 49, from 800 to 1,500 florins, in the case of.
princes 10,000 florins.]
THE ISLAND OF BALL 139
The long periods that corpses are kept is also noteworthy,
although I have discovered that Crawfurd's statements (" As. Re-
searches," vol. xiii. p. 136 ; " Ind. Arch.," vol. ii. p. 253) as to the
length of time are inaccurate. The duration of the impurity of a
house in which a dead bodyjias lain is more precisely fixed, being
more than a month for a Siutra, only eight days for a Brahman,
and an intermediate time for the second and third castes. The
.time during which corpses are kept varies very much, and the
day of the cremation does not at all depend upon the date of the
death.
There are bodies in Jladong which have been kept twenty
years ; on the other hand, the body of the prince of Gyanyar was
burned about forty days after his death. But the statement
(Crawf., " Hist. Ind. Arch.," p. 244) that a month and a week is the
time that bodies must be kept for cremation is also inaccurate, for
this reason, that it is not every day that is suited for a cremation ;
it must be a lucky day, and in order to obtain such a day all good
omens must concur ; it must also be in the first half of the lunar
month (with a waxing moon), and thus can hardly ever happen at
one and the same interval of time after the date of the death.
A series of ceremonies belong to cremation. Immediately after
death the body is embalmed by the relatives, and in the case of
a prince, also by the junior rajas of the friendly States ; for this
purpose spices are used, with which the body is entirely covered,
and these again are covered on the principal parts of the body
with small coins over these come clothes, mats, and a covering
of split bamboo. This sort of embalming 1 does not protect the
body from partial decomposition ; the emanations ~ from it are
collected in a vessel underneath the Bale 3 on which the body lies,
and are poured away every day. If the body remains so long
unburnt, it becomes dried up in about six months. It is watched
the whole time, and if emanations still run from it, they are caught
in the vessel referred to ; offerings and "holy water are also pre-
sented. In its mouth the corpse has a gold ring, set with a ruby ;
this protects it from the power of the demons, and it is the most
heinous sin to steal such a ring (a thing, however, which happened
not long since). Three days before the cremation the corpse is
stripped of its coverings, and the relatives look upon the dead for
the last time. Meanwhile, the spices have penetrated into the
body; the latter is then washed and again enveloped in split
bamboo, mats, and clothes. Instead of a ruby, five small metal
plates inscribed with the word ong, and with mystic formulas, are
1 [Called nglclct in the high dialect, corresponding to the Javanese Ngoko,
and nggiihin^ in the low or Krama form.]
- [Called banych, and in the case of a Brahman corpse, pringet. ]
3 [The Bale, or more correctly bale pandung, is a bier or kind of tent, often
of considerable dimensions, for the reception of the body until its cremation.]
MO AN ACCOUNT OF
put into the mouth ; the five plates signify the five principal gods
(Siva, Brahma, Vishnu, Indra^ and Yama?), comprised in the
words Sa, ba, ta, ha, i (Set = Siva, Ba - Brahma, 1= Indra ; ta
and ha are not clear to me, but ha seems to indicate Hari, one of
the principal names of Vishnu) ; * the plates are of gold, silver,
copper, iron, and lead. These plates, which keep the body, as it
were, under the protection of those five gods, are taken away
immediately before cremation, when this protection appears to be
rendered unnecessary by the reading of the Vedas and the pour-
ing out of holy water. Houses in which dead bodies are kept are
unclean, and, in the case of princes, the palace is not occupied by
the successor until after the cremation, and is merely guarded by
the people belonging to the deceased. This is occurring, at the
present moment, in Dm Passar (one of the three residences of
princes in Badong\ where the bodies of five illustrious princes are
kept in the great palace, and the reigning prince lives for the
present in a small house of little pretension. The first ceremony
mentioned above, that of washing before cremation, is called
pang&skaran or pabrissian (purification) ; askara seems to be
Sanskrit, but it is not found in Wilson (first edition). 1
The preparations for cremation take much time and much
money. A bridge is built on each side of the wall of the palace
for the conveyance of the body to thefiade. 2 The Bade, or funeral
pyre, is, in the case of princes, very gorgeous. It rests on a basis
of bamboo, concealed by handsome hangings, under which the
bearers place themselves, to move the structure along. On tins
foundation there is a pyramid of woodwork or bamboo, in from
three to eleven storeys. I have only seen the latter number at
the cremation of Dewa Mangis in Gyauyar, and this agrees with
the number of storeys in the Merus of the temples ; it certainly
has a religious meaning, reminding us of the Buddhist pyramidal
buildings and of the philosophical ideas connected therewith.
This pyramid is ornamented with more or less splendour, according
to the means of the relations of the deceased; the covering is
made of little balls of cotton, fastened on to the wooden frame-
work in fringes, and forming heads of elephants and Butas round
the bottom and at the corners of the different storeys. These
balls are of all the colours mentioned in the "Usana Bali": white,
'* According to Wilson, T/ia is Siva; J, Kama; Ha, Siva and Vishnu. See
also " Usana Bali," p. 328. These letters are called PanchAkshara, the five
(sacred) letters.
1 [Askara is a mutilated form of the Sanskrit sanskara i.e., the ceremony
/car' f^ox-nv. See v. d. Tuuk in " Hijdiagcn," III. vol. v. p. 212. Pambrtsihan,
horn in\'sih, " pure," bcrsih in Malay and Javanese.)
- [More correctly isadah. In Javanese and Sumianese the word means any
receptacle. Jacobs, 1. 1. p. 227 ff., gives a full account of two of those gorgeous
structures he saw at Tabanan.]
THE ISLAND OF BAI.f. 141
rtd, dark blue, and yellow. The whole of the lower portion, as
well as the corner of the pyramid, is covered in this way. The
outer sides of the different storeys are ornamented with tinsel
and red, and inside they are fitted up as rooms with mirrors and
furniture. Persons are stationed on each storey to attend to the
regular progress of the machine and to keep the little rooms clean.
The corpse is laid in the highest storey, and covered with a great
number of white cloths, which hang down the sides of the
pyramid.
Special places are made ready for the cremation of distinguished
persons. In Gyanyar a square of about 400 paces in length and
breadth, surrounded by a wall of pillars of masonry with trellis-
work between them, was prepared above the palace, on the
summit of the hill up which the dessa extends. In the centre
there was a Bale of masonry inlaid with a sort of mosaic of glass-
work and boards, after the manner of the Bales used for offerings
in the temples. The Bale consisted of two storeys, and was
surrounded with trellis-work. Above it rose, on four pillars,
another chamber and a roof, the chamber ornamented with mirrors,
and the whole covered with tinsel. The pillars were covered in
red. This place was intended for the cremation, and upon it
stood a figure of a lion (singha). This figure is only used at the
cremation of reigning princes (chokorda)'* other distinguished
persons are burnt in the figure of a cow men in that of a black,
and women in that of a white lembu. The common man, generally,
has only a simple square wooden coffin to be burnt in, but figures
are sometimes used also by the lower orders e.g., Gajamina, a
monster, half elephant, half fish. These figures of beasts are
ornamented to a certain extent (in Gyanyar very splendidly) ;
the erect tail and the back are taken off, like a lid, when the body
is let down into the figure, and are replaced after the ceremony of
letting do\vn the body. In Gyanyar, outside the square, in the
centre and at the sides, there were several large and ornamental
Bales for the distinguished spectators, and especially for the
very numerous women. In addition to these, there were the
places for the cremation of the three Belas, on the left. For
several days before the cremation of people of rank, one or more
Panditas reside with the relatives ; they manage all that has to be
done, indicate the mode of making the offerings, prepare the
metal plates, and invoke, by means of the Vedas, the success of
the important ceremony about to be undertaken. It is they also
who conduct the Bade to the place of cremation. If the deceased
belongs to the second caste, the Kshatriyas, the Bade is fastened
to a serpent (nciga or nagabandha, serpent-band) ; this serpent is
* It has been said, however, that this custom did not come from India.
\Chokorda, in Balinese paduka.]
142 AN ACCOUNT OF
ornamented in the same manner as the foot-piece of the Bade,
and has wings ; the body is quite thirty fathoms long, and is
carried by men. Before the procession starts, the Padanda
descends from his palanquin and shoots from the four cardinal
points at the head of the serpent, whose evil influence is thereby
destroyed. He uses for this purpose wooden arrows with white
flowers attached to their ends, the flowers alone being discharged
at the serpent.* This peculiar custom of fastening a serpent in
front of the Bade, and of the killing of this reptile by a Pandita,
points to a fable of former quarrels between the earthly gods (the
Brahmans) and the princes. A Dcra ailing was in the habit of
ridiculing the Brahmans ; he especially threw doubts upon their
supernatural power. Once upon a time, when a powerful Pandita
was at his palace, he caused a goose to be put into a well or pit,
and the latter to be closed. He then asked the Pandita what
animal was in the well, and the Pandita replied, a naga. There-
upon the prince wished to ridicule him, but when he opened the
well, a terrible serpent came indeed forth. The king then,
astounded and terrified, was rescued in his need by the good
Pandita, who slew the serpent, and ever since that time, at the
cremation of all Kshatriyas, a serpent has been fastened to the
Bade, killed by the Pandita, and burnt with the corpse. When
this serpent has been figuratively slain, it is wound round the
Padanda's seat, the tail remaining fastened to the Bade ; in this
manner the Padanda conducts the corpse (or, in reality, according
to their notions, the soul) to the place of cremation, and so to
heaven (sva rga\ where it (the soul) is admitted among \hQpitaras.
The procession at cremations is very long, even with the lower
orders. In Gyanyar, where the princes and many armed men,
besides the bands of musicians and actors from all States of Bali
(excepting Banglt) t were present at the celebration of the feast,
the procession extended for upwards of a "paal." First came
always the coolies bearing the wood for the cremation. In Gyanyar
they carried pieces of wood of uniform shape, coloured black and
gold. It is said to be customary to use sandal-wood for princes ;
but this did not appear to me to be the case, for the pieces were
too large, and would have been too costly for the means of a
small Balinese prince. Then came the music and the men armed
with muskets, then all the articles of personal use belonging to
the deceased, and then holy water and offerings for the Jlulas,
carried by women on their heads. The order of march is : twenty
or thirty men armed with lances, and then the bearers of the
Upachara, walking two and two (upachara means in Sanskrit
service and also present ; to be understood here as the articles
* This is almost the only case in which a bow and Arraio are used in 15ali ; with
this exception, we only meet with their use in some dramatic representation?,
where they appear in accordance with the Indian myth*.
THE ISLAND OF BALL 143
belonging to the service of the deceased* These include everything
of domestic use in Bali : clothes, rings, and other jewellery,
mirrors, and articles of the toilet, the sirih-box (in the case of
princes, of gold), the water-bottle (also of a precious metal), the
umbrella (payung ; umbrellas are also held over the jewellery by
those walking nearest) ; and, finally, the horse of the deceased,
richly caparisoned, led in the procession by men, and sometimes
ridden by a young son or grandson of the deceased. All these
are again followed by armed men and musicians ; the bands of
music separate the divisions of the procession. The second part
is devoted entirely to holy water (toya tirta}\\ more than a
hundred women carry small vessels of water on their heads ; this
water comes from the most sacred places in Bali, and is solicited
and bought from the Padandas who stand in especially high
esteem. Here, again, the intermixture of Buddhism is note-
worthy ; we have here not only holy water from the Padanda of
S/7Y7, but also from the Buddhist Padanda, and these waters are
mixed together, ,J It appears, therefore, that the Sivaites also
require the help of Buddha, and that Buddhism is still, as it
were, an integral part of the religion of all Balinese. This water
is called toya or toya tirta, " water of a holy place ;" tirtha is Ind.
" a holy bathing-place," and also " a chapel ; " our holy water comes
from both such places, from the sacred waters (see toya Sindu in
" Us. Bali," p. 337), from the temples, and from the houses of the
Padandas, who have consecrated it by mumbling the Vedas.
The offerings for the Butas, banten dagan, form the third
division of the procession. Dagan is the Balinese name for
Buta, although the latter word is used quite as often. These
offerings consist of all sorts of meats, cooked and half-cooked, of
rice, fruits, flowers, leaves ; all these vegetable offerings must be
of five different colours : white (the colour of Siva or Mahadeva),
red (Brahma), yellow (Buddha and also Siva in Bali, black or
dark blue (Vishnu ; also Kala), and brumbu, a mixture of the four
colours ; the last is the colour of the supreme Siva, who combines
in himself the natures of the four already named. These offerings
are made to propitiate the Butas, who are supposed to haunt the
burial-grounds. The offerings for the gods and pitaras (the
shades), which are offered the same day, are left at home.
* According to Gericke's Javanese Dictionary, which has reached me too late,
Jut Packard means tokens of distinction and of honour, state-decorations, insignia,
ceremony, pomp, splendour. This agrees with our explanation.
t This water is as holy as that of the Gauge?.
The mixed holy water is called toya pangcntas, and is especially purifying
and a protection from danger ; leaves of gold (mas fripi} and a costly gold ring
(which the Pandita keeps) with a ruby are placed in it; and also flowers, balung-
balung, or ground sandal-wood; powdered rubies; whole rice (bija}\ cut flowers
and leaves (samsam}.
Comp. Pitdmbar*) "with a yellow garment."
i 4 4 AN ACCOUNT OF
Here usually follows the Padanda, carried in a palanquin, and
drawing after him, when a Kshatriya is burnt, the dead naga with
the Bade attached to it. At several points, especially at the
corners of the road, before the house of the deceased, and before
the place of cremation, the Bade is turned three times, the
Padanda being carried before it and leading the movement. The
lofty structure of the Bade sometimes renders it necessary to
employ more than 400 men to perform these manoeuvres. In
Gyanyar the prince, the son of the deceased, preceded the Bade,
followed by a great number of other princes and the armed men
of Gyajiyaf and Badong. In other cases I have always seen the
relatives following the deceased ; the reason why the prince
preceded the corpse was this, that the Belas with their procession
followed the Bade, and that the princes considered it beneath
their dignity to go behind the Belas, who in this case were all
Gw.n/iks (concubines from the fourth caste). During the whole
march A^zv-songs are sung, chiefly by those who carry the Bade ;
they were not songs of mourning, however, but favourite portions
of the Rctniayana, Bdrata Yudda, and other works well known
to the people.
When the procession has arrived at the place of cremation, and
the Bade has turned round thrice for the last time, the body is
carried down from the top of the Bade by a staircase made for
this purpose, and up by another small staircase to the place of
burning, where it is laid in the coffin (the lion or cow). The
corpse is then only in the covering of bamboo referred to above ;
the clothes and ornaments remain on the Bade, and some of them
are taken home again, the remainder, as well as all the ornaments
of the Bade, being given up to the people as plunder after the
cremation. The Pandita then,, mumbling the Vedas, sprinkles
on the body the different kinds of holy water a ceremony which
generally lasts more than an hour. At last, after all the vessels
of holy water have been emptied, the Padanda quits the spot,
and the bearers of wood now kindle a fire under the figure of the
animal, which is consumed but slowly, on account of the copious
wetting which it has received. Sand is strewn on the foundation
of wood or masonry, to prevent the fire from spreading. When
the corpse is consumed, the bones are collected by the watchers,
laid on the purified place of burning, and covered with clothes, to
which are added also the utensils for adornment and small offer-
ings. These are carried the next day with great state to the sea
and thrown into it, together with money and offerings. Before
this is done the Padanda again reads the Vedas on a lofty erec-
tion, making frequent use of the bells, which stand by his side.
Opposite the Padanda stand rich offerings of all kinds, and
especially quantities of clothes and money. The Padanda re-
ceives these offerings. He is also offered a tolerably large sum,
THE ISLAND OF BALI. 145
U P to ^"50, for performing the ceremonies, but most priests
take only part of this and return the rest to the relatives of the
deceased. Other ceremonies follow at intervals of a fortnight, of
some months, and of a year after the cremation. At these we
have again a procession, as at the cremation itself; there is a
Jtadc, upon which the corpse is represented by flowers; these
flowers (ptispa) are also thrown into the sea. During the whole
of this time numerous offerings are made for the deceased.
After the ceremony, which takes place a year after the cremation,
the deceased is regarded as actually admitted among the pitaras
(the shades) in Indra's heaven, and as sharing in the offerings
which are presented to the latter in the domestic temples.
Great princes sometimes celebrate a feast to the memory of the
dead even after the lapse of some years. 1
The burning of widows has attracted most attention from Euro-
peans. Wonder is expressed at the great number of women sacri-
ficed in this way in Bali in former times (Crawfurd, " As. Res."
vol. xiii. p. 135); yet it should be borne in mind that in Bali
polygamy is carried further than in any other country that, e.g.,
the great-grandfather of raja Kassiman, prince of Ngnirah Sakti
Pam 'chutan, had 500 wives and thus it cannot surprise us to see
so many women burnt, particularly since the women who offer
themselves gain the highest distinction, and the sacred writings,
from the Parvas (divisions of the Mahabhdrata] downwards, re-
present this act as praiseworthy and almost necessary. That in
India but one wife follows her husband in death, is naturally
explained by the fact that very few Hindus have more than one
wife. But little is yet known of the burning of widows among the
princes of India, and as but few independent princes now exist who
are worshippers of Siva, the silence on this subject is easily
accounted for.*
In Bali the burning of widows is confined almost entirely to the
princely families. The princes are now either Kshatriyas or Vesyas
(Devas or Gustis), and so far Crawfurd is correct in saying that
only the second and third castes observe this custom. The true
reason of this, however, lies in the outlay which is requisite. The
Sftdras are seldom rich enough to bear the expense of a splendid
cremation ; their women, also, are less under the influence of the
priests, who do not trouble themselves much about such common
people ; and finally, the wife's respect for the deceased husband is
not so great, because he was of the same rank as herself. Among
1 [Greater details concerning cremation are given by R. van Eck in
"Tijdschrift voor Nederlandsch Indie," 1879, vol. i. pp. 104-24. See also
Jacobs, 1. 1. pp. 49-53-]
* A man of high rank in Nipal, Bhngtee Thapa, was followed into the fire
by his two wives. H. T. PRINSEP, Transactions of the Marquess of Hastings,
vol. i. p. 170, note.
SECOND SERIES. VOL. II. L
146 AN ACCOUNT OF
the Brahmans the burning of widows is still rarer, but yet a case
which occurred some years ago in Badong has come to my know-
ledge. The Brahmans do not regard it as necessary ; it is not
prescribed in the Vedas, nor is it practised in India by the
Brahmans, who have remained free from the worship of the
sect.
The names given to the burning of widows are known from
Crawfurd : Satya and Bela. This explanation of them must be
modified. Satya is the burning of a wife, who from a platform
erected for the purpose, throws herself into the same fire with her
husband, committing suicide with the creese at the same time. These
may be either lawful wives or concubines, and the latter is not
unfrequently the case. Bela, on the contrary, is the burning of a
wife, who is burnt in a separate fire, not with her husband, jumping
into it alive, without using the creese. This latter method is not
thought so much of, but is the more common. These women, who
are also called improperly Belas, are mostly concubines, but I have
also been told of cases in which lawful wives have chosen this
method of self-immolation.
The names Satya and Bela are Indian the former unmistake-
ably, the second in all probability. Satya is truth, fidelity ; the
wife who dies in this way is called Satyavatl, a true, faithful
woman, who has performed in all things her duty to her husband.
Bela is explained by Crawfurd as "retaliation," but so far as I am
able to reconcile the Balinese explanations with each other, it is
rather the Sanskrit wela "sudden and easy death" (Wilson). 1
This is rendered clear by the way in which it takes place ; and
further by the fact that the sacrifice of the followers of a prince
defeated in war, who then die together in an amok, is also called
Bela. In general it means in Bali " dying with the man of higher
rank " (the wife with her husband, the slave with his master, the
followers with the prince).
That female slaves were murdered by men appointed for the
purpose, and were afterwards cast into the fire, which would seem
to be a compulsory human sacrifice, must have been a fact 200
years ago, and was observed by Mr. Zollinger in Lombok* but I
have not found any trace of it in Bali. The women who sacrifice
themselves are indeed, as a rule, slaves, for they are mostly from
the fourth caste (the Sddras), at least all the concubines are ; but
1 [The meaning given in R. von Eck's " Balineesch Woordenbock " is " faith-
ful unto death." In Javanese, the word conveys the meaning " to die or suffer
with or for another." See the numerous examples in Vreede's edition of
Roorda's Dictionary, p. 1093. For other explanations of the terms satya and
wela, see Wilken, " Het Animisme," p. 85. Satya is evidently an abbreviation
of Satyavati.]
* See the report of the Dutch envoys to Gelgel, in Crawfurd, " Ind. Arch."
vol. ii. p. 244 sqq.
THE ISLAND OF BALI. 147
their immolation is voluntary, 1 and neither in the Satya or the
Bda are they touched by a man. From the moment that they
declare themselves ready to be burnt alive, they are holy persons ;
they enjoy all the honours of the Pitaras. They may no longer
tread the ground, but must always be carried. Offerings are even
made to them, and all their wishes are satisfied. Nevertheless,
the successor of the deceased simply regards such concubines as
the servants who will be required by his father after his cremation.
The women themselves are excited by religious ideas ; a female
priest always accompanies them till the moment of death, and
describes to them in glowing colours the happiness of life in
Svarga, the rise to a higher caste, and how they will thereby
become the lawful wives of the deceased. To these deceptions
must be added the honour and the advantages which the woman's
relations expect to receive from the prince's successor ; the men
of her family have a claim to offices ; they are made chiefs of
Dessas, Pamb'kcls, &c. They are therefore the only persons who
sometimes use means of compulsion to prevent the women from
retracting. They accompany the victim of the family ; they keep
up the fire, and, if the woman hesitates, tip up the plank on which
she stands above the fire, so that she falls in against her will.
These cases, however, are of rare occurrence. Deception of the
imagination and the use of opium have generally made the victims
quite indifferent, and they jump into the fire as if it were a bath.
Eight days after the death of a prince or noble his wives are
asked whether any of them wish to follow him into the fire ; those
who then state their willingness are accepted, and, during the
interval before the cremation, are shown the above honours.
They cannot easily change their minds ; the opposition of their
relatives, and the shame which would attach to them, as well as
1 [In reference to this we quote an appropriate passage from the work of
Dr. Julius Jacobs ("Eenigen tijd onder de Baliers," Batavia, 1883, p. 230),
whose account of the inhabitants is at the same time the most recent and the
least biassed : "It is perhaps too much to say that a wife does not follow her
husband to the funeral pyre of her own fret will. There are actually women
who, to all appearance, voluntarily throw themselves into the fire, or, in the
case of the husband having fallen in battle, stab themselves on his corpse (Ma).
A mother may also follow her child to death ; a betrothed, her lover ; a child,
its mother ; nay, even a friend his friend. Supposing that there exist, to some
extent, a genuine sorrow, a holy attachment for the beloved one, it is also in a
great measure a fact that those who sacrifice themselves are heartily convinced
that in requital of that noble deed they are immediately after death received
into the Satyaloka, where they are made partakers of unutterable joys, so that
thereby at once their sins are wiped out and they attain to eternal bliss. Add
to this the influence of the priests who here, as elsewhere, are powerful agents
with women, and try to induce them to self-sacrifice by holding out to them
all sorts of promises, and who bring them by dint of fasting, prayers, and
stupefying drinks, to a state of mental imbecility that makes them no longer
accountable for their actions, and you will know what is meant by their sacri-
ficing themselves of tJieir own free w///."]
L 2
148 AN ACCOUNT OF
the deceptions of the priests, hold them back. The women who-
may wish to give in their names after this period of eight days
are not accepted, perhaps in order to avoid the appearance of
compulsion. The women who have offered themselves to be
burnt lead thenceforward a life of pleasure, and enjoy much
greater honour than they ever knew before. This again is a reason
why they should not change their minds during the interval, some-
times a long one, before the cremation.
We will here add some details of a cremation, at which women
were burnt, in Gyanyar, of which we were eye-witnesses. On the
2oth December, 1847, tne prince of Gyanyar, Dewa Mangis, was
burnt ; we have already spoken above of the way in which his
body underwent cremation. The corpse was followed by the three
wives (concubines), who became Belas. A procession went before
them, as before the body: (i) Upachara, (2) Toya, (3) Banten
dagan. They, like the body, were seated in the highest storeys
of the Bades, which, like that of the prince, were carried by men,
but had only three storeys. After the body of the prince had
arrived at the place of cremation, the three Belas in their Bades
each preceded by the bearers of the offerings destined for her,,
with armed men and bands of music were conducted to the
three fires.
Their Bades were also turned round three times, and were
carried round the whole place of cremation. The women were
then carried down steps from the Bades, and up the steps of the
places erected for their cremation. These places consisted of
a square of masonry three feet high, filled with combustibles r
which had been burning since the morning, and threw out a glow-
ing heat ; the persons appointed to watch them fed the fire, and
at the moment when the women leaped down, poured upon it
a quantity of oil and arrack, so that it flared up to a height of
eight feet, and must have suffocated the victims at once. Behind
this furnace stood an erection of bamboo, in the form of a bridge,
of the same width as the square of masonry, and about forty feet
long, and from sixteen to eighteen feet high ; steps of bamboo led
up to it in the rear. In the centre there is a small house, afford-
ing a last resting-place to the victim, in which she waits until the
ceremonies for her husband are finished and his body has begun
to burn. The side of the bamboo scaffold nearest to the fire is
protected by a wall of wet Pisang-stems. Upon the bridge lies a
plank smeared with oil, which is pushed out a little over the fire,
as soon as the time for the leap draws near. At first there is
a door at the end of the bridge, and this is not removed till the
last minute. The victim sits in the house on the bridge, accom-
panied by a female priest and by her relatives. They all speak to
her of the happiness which she will now shortly enjoy with her
husband. She then makes her toilet j her hair especially is
THE ISLAND OF BALI. 149
combed, the mirror used, and her garments newly arranged ; in
short, she arrays herself exactly as she would for a feast. Her
dress is white, her breasts are covered with a white slendang ; she
wears no ornaments, and after all the preparations to which it has
been subjected, her hair at the last moment hangs quite loose.
When the corpse of the prince was almost consumed, the three
Belas got ready ; they glanced one towards another, to convince
-themselves that all was prepared ; but this was not a glance of
fear, but of impatience, and it seemed to express a wish that they
might leap at the same moment. When the door had been
opened, and the plank smeared with oil and pushed out, each
took her place on her plank, made three Sembahs by joining her
hands above her head, and one of the bystanders placed a small
dove (titiran) 1 upon her head. When the dove flies away, the
soul is considered to escape. They then immediately leaped
down. There was no cry in leaping, no cry from the fire ; they
must have been suffocated at once. One of the Europeans
present succeeded in pushing through the crowd to the fire, and
in seeing the body some seconds after the leap it was dead, and
its movements were caused merely by the combustion of the
materials cast upon the flames. On other occasions, however,
Europeans have heard cries uttered in leaping, and in the first
moments afterwards.
During the whole time, from the burning of the prince till the
leap of the victims, the air resounded with the clangour and noise
of the numerous bands of music. The soldiers had drawn up
outside the square, and contributed to the noise by firing off their
muskets. Besides these, some small cannon were discharged.
There was not one of the 50,000 Balinese present who did not
show a merry face ; no one was filled with repugnance and disgust,
except a few Europeans, whose only desire was to see the end of
such barbarities. The Balinese look upon this cremation simply
as the consummation of their religious and domestic duties ; no
one sees any cruelty in it. Yet, as the all-powerful priests attach
little importance to it for their own caste, with their aid the
diminution and abolition of these human sacrifices among the
other castes might be effected. The priests' interest in maintain-
ing it is, alas ! a monetary one and therefore no help can be
expected from them.
The number of women burned in Bali itself is not considerable ;
in Lombok, however, where only 9000 Balinese reside (the rest
are the Mahomedan Sesakers), this ceremony is more frequent.-
1 [It is called Katitiran in Malay, and a kind of sacredness attaches to it.]
2 [A full and interesting account of the self-sacrifice of the wife of a Gusti is
.given by Zollinger in his article on Lombok, "Journal of the Indian Archi-
pelago," vol. v. p. 529 ff. The Journal of the Dutch Embassy to Bali
j(" Bijdragen," II., vol. i. p. 52 ff.) gives a description of the funeral ceremonies
150 AN A CCO UNT OF
During the time I spent in Bali, five or six cremations of Gustis
took place in Lombok, and on each occasion from four to seven
women offered themselves up. This preciseness in following the
ancient usages also shows itself in great offerings, which cost con-
siderable sums, but at the same time add greatly to the fame of
the prince who makes them, and strengthen him for war (e.g., the
offerings in September 1846). The explanation of this lies in
two circumstances : first, the Balinese of Lombok, and especially
their prince, are much richer than those of Bali, where even the
reigning princes, through the great extension of their families,
possess but little means ; and secondly, the Balinese of Lombok,
and particularly the present prince, are looked down upon, and
said to be ignorant in their religion and their customs, by the
Balinese of Bali, and especially by the Dei' a agiing, whom the
people of Lombok do not acknowledge. Now to refute these un-
favourable opinions, they show themselves to be much more
precise in the performance of their religious duties than the
majority of the Balinese in Bali. During the two years of my stay
in Bali, I only knew of one case of widow-burning in Gyanyar. In
all the southern States also none took place ; intelligence from
Karang-Assem and Bolcleng is uncertain and irregular, but I heard
nothing of the practice in these States either. It cannot be
asserted, however, that the custom has entirely disappeared.
Widow-burning is considered a necessary adjunct to the crema-
tion of a great prince, and in the last thirty years a large number
of women have actually been sacrificed in Badong.
Nine Bdas were burnt with the prince of den Passar, NgruraJi
Made Pam'clmttan, Kassimaris father, and three with the last
prince but one of Panfchuttan, Ngrnrah Gdc. Pani'cJmttan (the
father of the principal wives of the present Raja of Pain'cliuttan}.
One very young wife, who threw herself tandakking into the
flames, was burnt with the last prince of Panichuttan, Anak Agitng
Lanang (father of the present prince).*
Raja Kalcran Kanginan was followed by two wives ; one killed
herself and became Satyavatt (see above), after the example set
by the wife of the raja Salya, one of the heroes of the MahabJia-
rata, known here from the Sa/yaparra. Raja Kalcran Kawan.
was followed by three wives. A young wife followed raja Haiti
Kassiman, a nephew of the old Kassiinan, who died very young.
Some women also followed Kassimaris brother in Belalitan. If
the dead bodies in Den Passar, which have already lain unburied
at the burning of the king's mother on March 21, 1633, at which many of her
female slaves were stabbed and burnt. At the funeral of the two royal princes,
forty-two and thirty-four of their wives respectively threw themselves into the
fire. Princesses of royal blood are not previously stabbed.]
* Tandakking on this occasion does not permit another person to iandalc
also.
THE ISLAND OF BALL 151
for fifteen or twenty years, are burnt, it is doubtful whether women
will follow them. The principal wives seldom follow, and in this
case the deceptions cannot have the same effect upon the concu-
bines as when they are practised upon them shortly after the
death. Raja Kassiman prevents this burning for political reasons,
as it might possibly deprive him of his prestige ; another reason
is the poverty of the present prince of Den Passar, whose revenue
. has been very much diminished by Kassiman, and who will not
for years be able to amass the sum required for such a grand
cremation. 1
CASTES.
We know from Crawfurd that the four Indian castes exist in
Bali ; we will hereafter give reasons which seem to show that caste
has also existed in Java.
The names given by Crawfurd are Brdhmana, Satriya, Wisiya,
Sudra. The names Brdhmana and Sudra are correct ; Satriya
is the corrupt pronunciation of KsJiatriya, which is found in good
manuscripts {Brahmdndapurdna, Rdmdyana, &c.) ; Wisiya has
arisen, through an oral mistake, out of Vesya, as the Balinese call
the third Indian caste ; the Sanskrit Vdisya cannot be rendered
otherwise in Balinese ; in the Balinese letters the four words
correspond closely with the Sanskrit names. We have shown in the
" Usana Bali" (1. 1. p. 254) that the Balinese taling has a mark above
the line in words where it represents the Sanskrit ai ; in the word
Wesya, however, the idea that this taling answers to an at, which
is unpronounceable, appears to have been lost. I have never yet
found the mark in the word Wesya in any manuscript.
The Indian names, however, are not those in common use in
Bali to indicate the difference in caste. The three upper castes
have names with meanings, which are used both in speaking to
and speaking of each member of the caste.- These titles are :
1. Ida for a Brahman ; this is the Balinese pronoun of the
second and third person in the high language. It is also used for
the titles of Deva and Gusti ; but when it stands alone, it always
means a Brahman, who is called He or Thou. A male Brahman
is called Ida bagus, a female Brahman, married or single, Ida
hayu, or, contracted, Idayu (see below as to the word Padandd].
2. Deva is the title of a Kshatriya, both for men and women
(e.g., Deva agung putri}. Deva means in Sanskrit god, and in the
language of the stage king (Wilson) ; but in the lists of names of
royal houses in India which we meet with in Tod's " Annals of
Rajasthan" and in several inscriptions published in the " Journal
1 [According to R. van Eck, the last widow was burnt at Klongkong in 1862
(see " Tijdschrift voor Ned. Indie,'' 1879, vol. i. p. 124) ; but this is doubtful.]
- [See Jacobs, 1. 1. p. 36. P. L. van Bloemen Waanders, 1. 1. p. 125.]
152 AN ACCOUNT OF
of the Asiatic Society of Bengal," we find also the title of deva (or,
corrupted, dew or deb) added to all the names, and we thus dis-
cover that the title of deva is used for all Kshatriyas, not only in
Bali, but also in India, for every one in that caste can become
king, and they are also all called Raja in India.
3. Gusti* is the title for the Wesyas, or third caste ; this name
is not Indian, at least not clearly. In India the third caste is not
of much importance, and we are therefore not surprised that it has
not a special title.f It is quite otherwise in Bali (and Java); but
few Kshatriyas have come here from India ; there were more
Wesyas, originally merchants, agriculturists, and artificers. Since
the Kshatriyas alone were too insignificant in number, greater
honour naturally fell to the more numerous Wesyas. They
even became kings probably in Java, and certainly in Bali,
although they are always regarded as subject to the Deva agung, a
pure Kshatriya. For this reason a title of honour was also given
to the Wesyas, and this is " Gusti." Gusti is never interpreted by
the Balinese as anything but a name for a Wesya. In Java it is
now used before the name of God, " Gusti allah" and before the
names of princes of royal blood ; Gusti, used before allah, indicates
that the word is equivalent to tuwan (lord). The fact, however,
that the princes of Solo 23\& Jokyokarta are called " Gusti" seems
to indicate that this family is also of Wesyan descent, and that they
retained this title of honour, although the name of the caste,
through Mohammedan influence, had been almost lost.
4. The fourth caste, very naturally, has no title of honour.
They are the born servants, and can make no claim to marks of
honour. Courtesy, however, assigns to them in address the hypo-
critical name of " father " and " mother " (bapa and incme). In
speaking of them, the term which is also applied to all foreigners
(Buginese, Chinese, and Europeans) is used kahula, slave,
dependent, or wang ( = orang), man ; they represent the ordinary
man, while the three superior castes trace their descent from gods
and demi-gods.
Mixed castes do not exist in Bali, whereas, even in the very
ancient Indian law-book of Manu, a large number of mixed castes
are enumerated. This is to be explained by the fact that too few
people of the three principal castes came over to Java and Bali,
* GoshtM is in Sanskrit "assembly, meeting; conversation, discourse;
family connexions^ but especially the dependent or junior branches." [This
derivation appears very doubtful in spite of Favre's qualifying concession ; see
his " Dictionnaire Malais," s.v. If tenable at all, it could only come through
a supposed adjective, goshtkint\
f Arya, it is true, is the name for a Wesya in India ; but what does it
mean ? The name Arya for the Kshatriyas oljava (and Bali) is to be read with
a long a (arya), and then means "one of the nation of the Aryans (an Indian,
a Persian}. According to Wilson: Arya^ "a master, a man of the third
tribe ; adj. excellent ; arya, of a good family ; respectable, venerable; apposite,
proper ; a master, an owner.''
THE ISLAND OF BALL j 53
and that these probably brought with them too few women, to
maintain the purity of their descendants, and therefore, to prevent
the extinction of the chief families, all the children remained in
the caste of their father, although the mother may have been of
common extraction.
In India the mixed castes arose in two ways : first, through the
marriage of a man of a higher caste with a woman of a lower one ;
secondly, through the marriage of a woman of high caste with a
man of inferior birth. The first case is not considered a disgrace
to the mixed caste ; the second degrades the offspring, because
this union is a disgrace ; and thus we have the most despised of all
Indian castes the chaijdalas f (Parias), the offspring of a marriage
of a female Brahman with a Sudra. In Bali the first kind of mar-
riage causes no mixture of caste ; the children follow the caste of
the father, but are inferior to the children of high-caste women,
and in matters of succession and inheritance are dealt with as of
lower standing, even though they be older than their more noble
brothers and sisters. The second case cannot occur, for, in Bali,
all marriages of high-caste women with men of lower birth (pro-
vided they are acknowledged) are punished with death. The
guilty woman is burnt alive a hole is made in the ground and
filled witli combustibles, into which the woman is cast; this
punishment is called labuh gni (to fall or be cast into the fire).
The man is weighted with stones and drowned in the sea ; this is
called labuh bat it. 1 This penalty, however, especially the burning
of the woman, is not always carried out so relentlessly. In several
cases which came to my knowledge, both the man and the woman
were drowned ; in another case, where the guilty man had escaped
vengeance by flight, the woman, at the command of her father
{a Gusti in Kutta), was killed with the creese by a relation, her
mother's brother, after having been adorned with flowers and fine
clothes, and rendered fearless by opium and strong drink.
There are, then, no mixed castes in Bali. In the " Usana Bali "
(1. 1. p. 262) different classes of Sudras are mentioned viz.,
Mandesa, Gaduh, Dangka, Batu-haji, Pasek, Kabayan, Ngukuhin,
Talikup ; these, however, are not different castes, but are all
Sudras, some of whom have been degraded to this caste from
that of the Wesyas, and still maintain, in a political sense, a
higher position than the common Sudras?
Chanddlas, nevertheless, exist in Bali, but they are not
chandalas by birth ; there can be none such in Bali. They are
afflicted with malignant and contagious diseases, and especially
.leprosy. Such diseases are ascribed to the curse of some deity or
of a Padanda. These persons may not reside in the kampongs,
] [Jacobs, 1 1. p. 34.]
- [K. van Eck in "Tijdschrift voor Ned. Indie," 1859, vol. i. p. 129.]
154 AN ACCOUNT OF
and are found in such places as the frontiers of two hostile States
(Mcngui and Tabanan), where they live in misery and without
intercourse with the rest of the people. I have seen others on
the high roads ; their employments are named by Crawfurd.
BRAHMANS.
All Brahmans in Bali trace their descent from Padanda Wahu
Rawuh, who, according to the Brahmans of the present time, lived
in or was descended from Kediri ; his descendants went thence
to Majapahit } and from Majapahit to Bali. According to the
Javanese accounts, a number of vSiva-Brahmans made their
appearance in Majapahit^ shortly before the fall of that place,
from India, and fled upon the destruction of that kingdom to the
East and to Bali. The statements of the Balinese upon historical
and geographical subjects are extremely inaccurate. According
to some, Kediri was situated in Baratawarsa (the holy land of
India). If we accepted this as true, we should be able to recon-
cile the accounts of Java and Bali respecting the arrival of Siva-
Brahmans at Majapahit ; but there are many reasons against this.
Jayabaya is said by some Balinese to have been a king in Bara-
tawarsa; undoubted Javanese accounts, however, state that he
ruled in Kediri in Java, and all the Kavi works of most im-
portance were composed under him and his predecessor, Ayer
Langia, The Brahmans, who composed these works, must there-
fore have come from India at an earlier period and acquired the
Kavi language in Java. We must bear also in mind here the
localization of Indian places in Java.
WaJm or Baku Rawuh means "the newly-arrived" (bahu is
the Malay bahrn ; rawuh, Jav.-BaL, is equal to the Mai. datang).
He is also named Bhagavcin Dvijendra, the holy Brahman prince
(dvija is a Brahman, one born twice the second time through
initiation into religion and sacred literature, and through assuming
the sacred band, Uparitd}. Indra is generally king, prince ; thus
we have Gajendra, prince of elephants, Rakshasendra, prince of
the Rakshasas, namely, Ravana. His two names, therefore, are
not family-names, but merely indicate his position ; he was the
leader of a colony of Brahmans which came from India. Concern-
ing his sojourn in Kediri or in Majapahit and Bali, the accounts,
as we have seen, are uncertain ; but, according to the accounts of
the priests themselves, the five existing subdivisions of the
Brahmans in Bali are descended from him and his five wives.
We have :
i. Brahma ua Kauiemi. (Kamenu is in Gyanyar.) They are
descended from Wahu Rawuh and a Brahmanic woman (Brah-
mana-Brahmani) ; according to Indian notions, these alone would
have the full rights and dignities of Brahmans, yet we find that, in,
THE ISLAND OF BALL 155
spite of their purer descent, they do not enjoy greater honour than
their brethren who are of inferior birth. The rank of the Brah-
mans depends upon their ability, their personal appearance, and
their conduct, so that even in youth the qualities of a powerful
Pandita may be recognized. The supernatural power attributed
to them, and not their descent, therefore, determines the choice
of the domestic priests of the princes. Political reasons also have
their weight. Brahmana Kamenu are found in Boleleng.
2. Bralunana of Gclgcl. (Gelgel is the ancient seat of the
Deva-agung, already mentioned in the account of the Dutch
envoys in 1635, in Crawfurd, "History of the Indian Archi-
pelago," vol. ii. p. 244 seq. ; it is also referred to in the Malayan
Manuscript of Abdullah " Tijdschrift voor Nederlandsch Indie,"
Jaargang 7, 2 dc Deel, p. 166 where it is erroneously written Gila
gila in the translation.) They are descendants of Wahu Rawuh
and a Kshatriya woman ; they are called Brahmana-geniten, and
include most of the Brahmans in Klongkon& Mengui, Bangli, and
those of Sanor in Bado?ig. The last-mentioned place is chiefly
inhabited by Brahmans ; none of those at Badong, however, are
of high rank, but the Padanda Agiwg of Somawati, and the
Padanda Made Aleng Kacheng in Taman Intaran, both belonging
to other subdivisions of the Brahmans, are the domestic priests
of the princes. Some Brahmans in Bolekng, also, came originally
from Sanor.
3. Brahma na-Nuaba. These are descended from Wahu Rawuh
and a Kshatriya widow (Bain manis, that is, "a short time
married "). Their original seat is the kampong Nuaba in Gyanyar;
hence came those of Sindhu in Karang-Assem, and of this
family is the Fadanda Agung in Sindhuwati, near Taman Intaran,
who, some twenty years ago, left Karang-Assem for political
reasons, and was received with gladness in Badong* He is the
chief domestic priest. in Badong, and his brother or cousin in
Gyanyar.
4. Brahmana -mas : descended from Wahu Rawuh and a
Wesya woman.t Their original seat is the kampong Mas in
Gyanyar. The family of the Padanda Made Aleng Kacheng in
Taman Intaran came from that place. He is also a domestic
priest, and although he is younger than and became a Padanda
long after the Padanda Agung was already a domestic priest, still,
* He had correspondence with Gyanyar at a time when Gyanyar, allied with
Badong, was at enmity with Karang-Assem ; he therefore lost his position in
Karang-Assem, and, together with his brother, was invited by the princes of
Gyanyar and Badong, received with much ceremony, and appointed to the
office of domestic priest.
t In general, the Brahmana-geniten, descended from a Kshatriya woman,
seem to be sought after by the princes in the States governed by Kshatriyas
and the Brahmana-mas, descended from a Wesya woman, where the "Wesyas
rule on account of their connection with these castes.
156 AN ACCOUNT OF
on account of his learning, his morals and the sakti ascribed to
him, he is held in the highest honour in Badong and Gyanyar.
There are also Brahmana-mas in Tabanan.
5. Brahmana kayit sfinya (ka\u snnya is literally empty /ra",
this seems to be an allusion to the position of the woman from
whom they are descended, for she was a slave, and was therefore
without education and learning). They trace their descent to
Wahu Rawuh by a slave. To this subdivision belong part of the
Brahmans in Mengui.
The Brahmans are very numerous in Bali ; their position as
regards the prince depends upon whether they are ordinary Brah-
mans (Idas) or Padandas i.e., learned priests. The former are
dependent upon the prince, must follow in war, may be employed
as envoys, and may be banished from the country, if they do not
strictly obey the prince's commands. Nevertheless, they are of
higher rank than the princes, and can marry daughters of princes,
while the princes may not marry Brahmanic women. On account
of their large number, a considerable portion of them live in
extreme poverty, and they do not disdain to cultivate the rice-
fields, to engage in the fisheries, and to do manual work for money.
The Padandas are Brahmans who have received a complete
education from another Padanda (their Guru). They must be
thoroughly acquainted with religion and with literature.
In order to become a Padanda, they undergo all kinds of tests,
to show their knowledge and their submission to the Guru. Thus,
for instance, they place their heads under the Guru's foot, and
drink the water that runs off his feet during his ablutions. Many
other ceremonies precede their consecration, and a certain amount
of wealth is required to defray the expenses, so that the dignity of
Padanda mostly remains in a family which has already grown rich
by means of this position. Others are supported and helped to
become Padandas by the Rajas. The mark of the dignity is a
staff, danda^ which they receive from the Guru, and which gives
them power to guide and to punish men in all things relating to
religion. After this staff they are called Padanda, that is, " bear-
ing a staff." Their other name, Paiidita, merely indicates their
knowledge ; Pandita is " learned," and is explained in Bali by
prajna, Sanskrit, and///V?r, Mai.
The domestic priest, Purohita (Sanskrit idem), is chosen from
the Padandas by the prince ; or sometimes the prince helps an
Ida whom he thinks clever and upright to become a Padanda, and
then makes him his domestic priest. He is then the spiritual
teacher (Guru) of the prince, who becomes his sishya or pupil,
makes the sembah * to him, and thereby shows his inferiority to
the priest ; he always sits on a raised seat (see Crawfurd, "Asiatic
1 [The Sanskrit danda is written danda in Javanese and Palinese.]
* The sembah is not an ordinary token of courtesy in 13ali. The upper
THE ISLAND OF BALL .57
Researches," vol. xiii. p. no). The domestic priest is consulted
in all religious and political matters, and even in the ordinary
affairs of life, in taking up arms, in choosing wives, &c. He alone
teaches the worship of the Pitaras and conducts the cremations of
the princely families. In all offerings, both domestic and for the
State, the advice of the domestic priest is acted upon ; he is pre-
sent in his elevated place opposite the offerings, and blesses the
ceremonies by means of prayers from the Vedas.
Sometimes the word Purohita is applied generally to all priests
who carry the staff (Padandas). It is these whom Raffles heard
spoken of under the name of Maperwita or Mapurwita. Purwita
is the corrupted pronunciation of Purohita, which latter word I
first recognized in good manuscripts ; mapurohita (or mapurwita)
means the being a purohita, or the collective Purohitas*
Guru loka, " teacher of the world," is a name applied only to
a few chosen Purohitas or Padandas, who present offerings for an
entire kingdom ; there are one or two of these in each of the
different States in Bali. They are the special Gurus and coun-
sellors of the chief prince, and are also the Gurus of the loka, of
the world, or the subjects of a State.
The prince can also call other Padandas to perform less im-
portant religious functions, and the smaller rajas (who are of
inferior birth) can likewise choose a Guru out of the rest of the
Padandas.
The present Guru lokas in the southern States, who exercise
great influence upon the actions of the princes, are :
In Klongkong
i. Padanda Wayahan Pidada, who is a Brdhmana-nuaba
2. Padanda K'tut Ngrurah, a Brdhmana-gcniten. He lives
in the kampong of Dawan.
In Gyanyar
i. Padanda Wayahan Kakeran, a Brdhmana-nuaba, cousin
of the Padanda Agung in Badong. He lives in the
kadaton of Sindhuwati, in the kampong of Kramas.
In Budong
1. Padanda Agung, Brdhmana-nuaba. He resides in Sin-
dhuwati (or Somawati)^ near Taman Intaran, and is the
father-in-law of
2. Padanda Made Aleng Kacheng, Brdhmanamas ,'m Taman
Intaran.
castes only make it to the prince, and to the Padandas who are Guru loka.
The reason of this is, that to the sembah is added an inward formula of
reverence out of a Mantra or the Vedas ; by making them to other persons of
rank they would humble themselves too much.
* The explanation formerly given ("Tijdschrift voor N. I.," Jaarg. 8, Deel 4)
of Purwita, by means oipurva and ila t thus falls to the ground.
158 AN ACCOUNT OF
In Tabanan
i. Padanda Jumping Brahma na-gcniten, in Pasekan, to the
north of the kadaton in Tabanan.
In Mtngui
i. Padanda Putu, Brahman a kayu snnya, resides in Kaba-
kaba. On account of their low birth the Brahmana
kayu snnya appear to receive much less respect than the
rest ; yet we find a Guru loka among their number,
although the men of Badong assert that the people in
Mengui are grossly ignorant. But Badong and Mengui
are old enemies.
The Brahmans also have many wives from among the people,
but the children always remain Brahmans.* There is not a single
one of pure blood, but, at any rate, care is taken that a Brahman
does not have too much Sudra blood among his ancestors. If, in
three generations, no woman of high birth has married into the
family, the descendants lose all rank, and are treated by the princes
as Sudras, and are obliged to perform service as vassals. It is the
same with the other two upper castes.
The Brahmans' wives of low extraction, especially if they have
children, are ennobled by the husband ; their rank in life, it is
true, is much inferior to that of women of high birth, and their
husbands give them nothing (they have to maintain themselves
and their children), but, after death, they are burnt as Brahmanic
women, and enjoy the honour of the Pitaras.
The women of high birth share in all the privileges of their
husbands. They are also instructed in the Vedas, themselves
present offeri)igs with the mumbling of the Vedas, and assist the
Belas at cremations. They are also called Padanda, with the
addition of istri, which is the highest title for women in Bali. (In
the Sanskrit strl only means " woman," in Bali " princely woman/'
compare pittri.}
KSHATRIYAS.
In India the Kshatriyas, the second caste, are, according to
law, those who, alone, bear arms and defend the country. The
princes are of this caste. But, in the present day, there are no
longer any pure Kshatriyas in India; even the Rajaputras of
Rajasthan are not regarded as of pure extraction. The profes-
sion of arms has thus come into the hands of the whole people.
The same thing has occurred in Bali. The rajas and their fami-
* Raffles, App. K., p. 238, says that the children of a priest by a woman of
lower rank are called Bujangga, but this nowhere came to my knowledge. See
below respecting the word Bujangga [the note in the section on Caste in Java].
In the "Tijdschrift v. N. I.," Jaargang 7, vol. ii. p. 172, subdivisions (Poma/i,
Anggana, &c.) of Brahmans are mentioned, whose existence I did not discover,
in spite of my repeated inquiries.
THE ISLAND OF BALL 159
lies, at least, are said to be Kshatriyas, but this is but partially
the case. The highest prince, the Deva Agung, is a Kshatriya,
but most of the other princes are of the third caste, the Wesya,
The Kshatriyas no doubt came to Java only in small numbers.
In Java the Usana Jawa enumerates Kshatriyas of Koripan
(Panjis-seat), Gaglang, Kediri, *&&Janggala. The chiefs of the
court of Jawa or Kediri, who were Kshatriyas and Wesyas, are
mentioned in the Rangga Lawe. This, the largest kingdom in
Java, did not contain many Kshatriyas; they are called Mahisa
or ICbo (buffalo, to indicate their strength), and Rangga (Jav.
rotiggo] ; their names are as follow : Mahisa Bungalan, ICbo
Wilalungan, ICbo Siluman, ICbo Jcrang, ICbo Kanigara, 1C bo
Chaluk, ICbo Tki, ICbo Taluktak, Ki Mahisa Sapati, ICbo Mun-
darang, and further Rangga Smi, Rangga Mayang, Rangga Palana,
Rangga Ralengsong, Rangga Pasung, Rangga Wirada, Rangga
Rabete, Rangga Swnbi, Rangga Sanipana, and Anurangga Smiting.
These are all the Kshatriyas who existed in the largest kingdom
of Java. A particular sort of creese is attributed to each of them,
and these creeses have crossed over to Bali through Majapahit.
The Kshatriya families themselves, however, have not crossed to
Bali, with the exception of the Deva Agung and his half-brothers,
Arya Damar and six others. The pure Kshatriyas were probably
exterminated in the numerous Javanese wars, and in the destruc-
tion of Majapahit, and the royal family of the Deva Agung also
seems to have once (either in Bali or Java) been on the verge of
extinction, for the reigning prince Taruna (a youth, unmarried)
had no children. Here, however, a new race was raised by a
priest, Dang hyang Kapakisan, out of a stone, batu henggong (see
" Us. Bali," p. 344). The race thus sprung from batu henggong
reigns in Bali at the present time, and from it the Kshatriyas are
descended. The descendants of the half-brothers (Arya Damar
and the others) were in later times degraded to Wesyas. Thus
all the Kshatriyas now existing in Bali trace their descent from
the Deva Agung a fact which would surprise us, for their
number is considerable, were it not that there are but few of them
in Badong, Tabanan, Mengui, and Karang-Assem ; and the rest
can have descended from the one Deva Agung (who lived 400
years ago) just as well as 800 Gusti Pam'chuttan in Badong,
counting only the heads of families who have houses of their own,
and are married, may have sprung from the Raja Ngrurah Sakti
Pam'chuttan (who four generations ago was prince of all Badong,
excepting Jambe). At the present day the Kshatriyas are still
reigning only in Klongkong, Bangli, and Gyanyar; formerly there
was also a Kshatriya dynasty in Boleleng, descended from the
Deva Agung, and its descendants now live in Badong. This
dynasty was expelled by Wesyas seven generations (?) ago. The
same thing took place in earlier times in Karang-Assem.
1 60 AN A CCO UNT OF
Two hundred years ago (1633) the Kshatriyas and all the
princes of Bali seem still to have been subject to the Deva
Agung, who is called prince of Bali by the Dutch envoys of that
time. The authority of the Deva Agung was very much weak-
ened by a war with Karang-Assem about a hundred years ago, in
which the ancient seat of Gclgel (to the east of Klongkong, near
the sea) was destroyed. Since that time Karang-Assem and
Boleleng, which was conquered by it, have no longer paid homage
to the Deva Agung. Soon after this, also, an insignificant Deva
in Gyanyar raised himself to the rank of a great prince by expelling
the relatives of the Deva Agung from the various provinces which
form the present Gyanyar. The new dynasty of Gyanyar, how-
ever, returned to the sovereignty of the Deva Agung. Bangli,
where a Kshatriya prince also rules, acknowledged the Deva
Agung until recently, but now has no connection with Klongkong.
Tabanan has withdrawn from the supremacy of the Deva Agung,
because Karang-Assem and Boleleng do not pay him homage.
Thus there remain only Mengui, Badong, and Gyanyar which
acknowledge the Deva Agung as Sovereign of Bali. Karang-
Assem and Boleleng use the name of the Deva Agung as their
Sovereign whenever it suits their purpose, but they pay him no
homage (sembafi) and send no presents (or tribute) to Klongkong.
Badong also yields him but little, and, in fact, has always been
opposed to his interests, although openly it pays him homage,
sends envoys to him, and contributes a little to great offerings
and feasts in Klongkong.
Dessak, Pradeva, and Pungakan are names of Kshatriyas who
have much Sudra blood in their veins.
WESYAS.
This caste, from a political point of view, is at present the most
important in Bali. To it belong the princes of Karang-Assem,
Boleleng, Mengui, Tabanan, and Badong, and also the prince of
Lombok. It is much more numerous than the Kshatriyas. The
race of the princes of Karang-Assem, Boleleng, Mengui, and
Lombok is descended from Patih Gaja Madda, the second
general of Majapahit, who, together with Arya Damar, conquered
Bali ; he was a Wesya of Majapahit, while Arya Damar, the chief
conqueror, was a Kshatriya, and a half-brother of the prince.
Arya Damar was the ancestor of the princes of Tabanan and
Badong; these, however, are now Wesyas, having apparently
been degraded to this caste, about 300 years ago, by the Deva
Agung. The reason of this degradation is said to have been
that these Kshatriyas wore their hair after the manner of the
Wesyas. In the present day there is no perceptible difference
between the Kshatriyas and Wesyas in the mode of wearing
THE ISLAND OF BALL i(>i
the hair ; the Deva Agung wear it exactly as the ancient Kassi-
man did, and the young Kshatriyas and Wesyas both wear
theirs sometimes loose and sometimes bound up (in the Sivaitic
manner) at the back of the head. The true reason was no doubt
political ; it was desired to humble the powerful race of Arya
Damar, and the rest of the Kshatriyas, who were descended from
the Deva Agung, and were already very numerous at that time,
endeavoured to obtain more power. This object was not attained,
but the princes of Tabanan and Badong have remained Wesyas.
We have seen above that the Kshatriyas of Dahaand Majapahit
bear the titles of Mahisa or K'bo and Rangga. Patih, Deniang,
and Tuinenggung are given as names for the Wesyas. Mantris,
who in Java now occupy the lowest position among the native
chiefs, can, according to the Balinese, be of either caste ; this is
explained by the original meaning of the word Mantri ; in
Sanskrit it means " Minister" and is thus applicable to any one
who fills this position, whether he be Kshatriya or Wesya. Patih
also was a much higher rank in ancient Java and Bali than it is
in Java now : Gaja Madda, who is stated to be the ancestor of
four princely families in Bali, and is regarded as the incarnation of
Vishnu, bears the title of Patih. And further, the first Deva
Agung (see " Usana Java ") appoints the conqueror of Bali and
governor of Tabanan, -Arya Damar, to be Patih or first Minister,
who must be consulted on all occasions. Of the Javanese titles
we also find that of B*kel in Bali ; it belongs, however, not to
members of one of the three upper castes, but to Sudras, and is
equal to mandur in Java. In Bali they are called Parb'kel, Pam-
bkel, or PraVkel ; the original name is Prab'kel, which, like
Pragusti and Pradeva (usually pronounced Pergusti and Perdeva)
means the assembly of the B'kels (Gustis and Devas).
Of the principal Wesyas of the Court of Daha (Kediri), the fol-
lowing are named in the " Usana Java " : Mantri Bawong, Kala
Mudong, Tumenggung Parungsari, D'mang Drawalika, Gebob
Basah, Lobar (the creese of this man's shape is still used by the
princes in Karang-Asseui, according to the Pusaka in Ngalihaii),
Kala Limpung, Buta Wilis, Bubar Baleman, Jalak Katengeng.
From such Wesya families, as well as from the real brothers of
Arya Damar, a great number of Balinese have sprung ; but all,
except the descendants of Arya Damar and Patih Gaja Madda,
are of no importance, and most of them have become Sudras. A
few still bear the title of Gusti and have followers, but the rest are,
in all respects, like the Sudras. The reason of this is, that their
forefathers in Bali were conquered and displaced by the races of
Arya Damar and Patih Gaja Madda.
The Wesyas were originally intended for commerce, agriculture,
and the exercise of arts and handicrafts. This is known in Bali,
but the principal Gustis despise these occupations, and they are
SECOND SERIES. VOL. II. M
1 62 AN A CCO UNT OS
only disposed to carry on trade for the sake of obtaining the money
required for opium-smoking and cock-fights. Trade, however, is
not solely in the hands of the Wesyas ; all the other castes, also,
take part in it.
In order to become better acquainted with the present position
of the Dewas and Gustis in Bali, we shall give here an account of
the princely families and their descendants.
THE PRINCELY FAMILIES.
We have already referred briefly to the fact that all the Kshatriyan
princes, and all the present Kshatriyas, trace their descent to the
Deva Agung. The princes and Gustis of Tabanan and Badong
are descended from Arya Damar ; and, finally, the princes and
Gustis of Mengui, Karang-Assem, Boleleng, and Lombok derive
their pedigree from Patih Gaja Madda. This carries us back to
the time of the conquest of Bali by the Javanese of Majapahit.
An ancient connection between Java and Bali is indicated in the
" Usana Bali." Bali was in the possession of evil spirits, or giants,
that is, the Balinese were not yet Hindus (comp. Abdullah, in the
"Tijdsch. voor Neerlands Indie," Jaargang 7, vol. ii. p. 160, sqq.).
A few Hindus from Majapahit had settled in Bali, and had a
temple in Bazuki (so-called after the Indian serpent-king Vasuki
who, in the Indian and Balinese mythology, accompanies
Siva and plays an important part. They were oppressed, how-
ever, by the infidel princes and people. The account of the
descent of the gods and the defeat of the Maya Danawa and the
demons indicates the triumph of Hinduism. The " Usana Bali"
does not tell us by what earthly means this religion was established,
and the reason of this seems to be, (i) that, in order to attain its
full sanctity, the religion must be introduced by the gods themselves ;
(2) that it was desirable or necessary to spare the feelings of the
conquered people (the original Balinese] by representing them as con-
quered, not by men, but by gods. The " Usana Bali " is intended
only for the people.
In the " Usana Java," however, we find traces of the true con-
quest. It is represented, here, as taking place immediately before
the crossing over of the Devj, Agung, the chief of Majapahit ; but at
that time Bali (according to* the same " Usana Java") had already
become a province of the kingdom of Majapahit, and is merely
subdued a second time after a revolt of the governor. The insti-
tution of castes and the Hindu religion evidently existed in Bali
previously to this, as is clear also from the narrative of Abdullah;
but the revolt and the defeat of the Governor of Bali afforded an
opportunity of dividing the land among the nobles of Majapahit,
and the prince of Majapahit, or his son, came to reside in this
island after the fall of the kingdom of Majapahit. This destruc-
THE ISLAND OF BALI. 163
tion of Majapahit was effected, according to the Javanese accounts,
by Muhammadans ; according to the Balinese, the kingdom and
city were deserted in consequence of a disease caused by a buta
(demon).
According to the " Usana Java," Arya Damar and Patih Gaja,
Madda were sent from Majapahit in the capacity of generals
against the rebellious Bali. Arya Damar conquers the north,
.while Patih Gaja Madda remains inactive in the south; but, on
the approach of Arya Damar, the latter portion also submits to this
victorious general. The crossing over of the prince of Majapahit
is caused, according to the " Usana Java," by the appearance in
Bali of a demoniacal king, Mraja Danawa (another infidel, there-
fore !) ; the latter is of the family of Maya Danaiva in the " Usana
Bali," and refers to the event that forms the subject of that writing.
At that time Arya Damar was in Majapahit, and on the receipt
of the intelligence that this Raksasa Mraja Danawa is exercising
his power in Bali, the prince of Majapahit himself sets out against
Bali with Arya Damar and his whole army ; after defeating the
Raksasa prince, who, when he can resist no longer, flies away
through the air, the prince of Majapahit establishes himself in
Gelgel. This account is obviously improbable, and was perhaps
invented to conceal from the original Balinese the manner in which,
and the reasons for which, the prince of Majapahit, or his son,
left his kingdom to settle in Bali.*
What would seem nearest the truth in these accounts is this :
Arya Damar had subdued rebellious Bali, and again compelled
respect for the prince of Majapahit ; a short time afterwards the
kingdom of Majapahit fell to the ground (through war or other dis-
asters), and the surviving prince, or one of his sons, came over to
peaceful Bali. The Balinese naturally regard this arrival as an
honour, and look upon the loss of Majapahit as of little moment,
for they say that place (and all Java) became infested with evil
spirits. The princes do not seem to have so easily forgotten the
loss of their great kingdom in Java ; hence their continual wars
with Blambangan, and even in Passuruan (Raffles, vol. ii. p. 200,
sqq., " History of Java "), whence, however, they finally had to
withdraw. Blambangan (the country near Banyuwangi) for a long
time still belonged to Bali. The wildness of this country is partly
owing to the wars with the Balinese, who were unable to hold it.
It is remarkable that the opposite side of Bali, Jembrana, is also,
to a great extent, desolate ; here, as in Majapahit, the reason of
this desolation is said to be that the dwelling of a king of de-
moniacal form made the land unsafe. But both Jembrana and
Blambangan were really laid waste by the long wars between Java
* The " Usana Java" does not give the name of the prince who became the
first Deva Agung in Bali. According to other accounts, his name was Deva
Agung K'tut, and this is given by Raffles and confirmed by the Balinese.
M 2
1 64 AN A CCO UNT OF
and Bali, and, even now, are little cultivated, more for security's
sake than from fear of the demoniacal king. The longing of the
Balinese to regain their lost country has shown itself in the expe-
dition of the Bolelengers against Banyuwangi under the English
rule. They have also attempted to gain in the East what they
have lost in the West hence the conquest of Lombok and the
attack on Sembawa, where they were stopped by the Dutch
Government.
After the settlement of the Deva Agung in Bali at Gelgel, the
land is divided among the chief men in the army and the Court.
Arya Damar received the great land of Tabanan, and became a
Patih, first Minister of the Deva Agung. The prince could not
undertake anything without consulting him, and this privilege
descends to his offspring, and forms the ground of the present
grievances of the princes of Tabanan and Badong, who never
forget this ancient privilege, and, as the Deva Agung does not
keep the old promise, no longer consider themselves bound to
him. Badong, however, preserves, for political reasons, the ap-
pearance of subjection. Arya Damar also obtains the title of
Arya Kencheng (Kenjeng or Kengjeng is the title for princely
Javanese invested with authority, and is also given to the Re-
sident, the Government, &c.). The number of his men is said to
have been 40,000. Smaller governorships were also given to
Arya Damar's brothers ; to Arya Sento, the countries of Pachitng,
the present Marga, belonging formerly to Mengui, but now to
Tabanan : to Arya Beleteng, the country of Pitiatih, since con-
quered by the princes of Badong, but still a separate kingdom ;
to Arya Waringin, that of Kapal in the present Mengui ; to Arya
Blog, that of Kabakaba in Mengui ; to Arya Kapakisan, that of
Habianscnial in Mengui ; to Arya Binchaluku, that of Tangkas in
Klongkong. Besides these brothers of Arya Damar, Arya Man-
guri is mentioned as governor in Dawuh in Karang-Assem, and
the three principal Wesyas, Tan Kuber, Tan Kawur, and Tan
Mundur (names, really, symbolical), also receive a domain. In
the " Pamendanga," a sort of history of the princes and priests,
however, of little value, the governorships allotted to these nobles
are somewhat different, but, at any rate, this work mentions Patih
Gaja Madda as governor of Mengui, a fact confirmed by all
Balinese, but omitted from the " Usana Java."* We thus see Bali,
at the very outset, divided among governors ; these could soon
change from governors into independent princes, such as we now
find. In the year 1633, according to the Dutch Envoys, the
Deva Agung seems still to have been the only prince in Bali,
and it is probable that he was regarded as such and had influence
over the whole of Bali until about 100 years ago, when Gelgel,
* The " Usana Bali" betrays partiality for the race of Arya Damar.
THE ISLAND OF BALL 165
his ancient seat, was destroyed. The countries adjoining Klong-
kong, Bangli, Gyanyar, and also Boleleng, seem to have been
immediately under the Deva Agung, and were then, in course of
time, given as governorships to members of his family. Here
also, after the degradation of the race of the Arya Damar, were
the only remaining Kshatriyas, but even these were partly ex-
pelled by the Wesyas. The history of Arya Damar's descendants
is remarkable only on account of the conquest of Badong and
the founding of this kingdom. The race of Patih Gaja Madda
has much more influence upon the history of the whole of Bali.
This chief, the second general of the princes of Majapahit, had
his seat in Mengui. The palace of Mengui is one of the oldest.
Abdullah (p. 163) even makes the Deva Agung reside in Mengui
from the first; the information I have obtained as to this point,
however, agrees with the " Usana Java," where the Deva Agung
has his first seat in Gelgel ; proceeding subsequently to Klongkong.
To Mengui belonged, besides the present country of that name,
the greater portion of Badong (the smaller, eastern portion of
Badong, formed the kingdom of Pinatih, which in later times was
subject to Mengui); and, further, a portion of the present Gyan-
yar, Kramas, and the land of Marga, which now belongs to
Tabanan. It was thus a considerable domain, and as large, if
not larger, than Tabanan. Arya Damar's brothers had, it is true,
various portions of the present Mengui, but they all appear to
have been speedily subjected to the power of Gaja Madda and
his successors ; the kingdom of Pinatih alone remained under the
descendants of Arya Damar, but it became tributary to Mengui.
The position of Bali at this period (about 250 years ago) may
therefore be thus described : Klongkong, Gyanyar (with the excep-
tion of Kramas}, Bangli, and Boleleng belonged to the Deva
Agung and to punggawas (governors) of his family ; Tabanan to
the descendants of Arya Dama ; Mengui with Badong and parts of
Gyanyar and Tabanan (Kramas and Marga) to the descendants
of Gaja Madda. Karang-Assem was probably still under the
descendants of Arya Manguri (at least partially). A change in
this state of affairs was caused by the princes of Mengui conquer-
ing Karang-Assem, and a Gusti of Tabanan establishing himself
and his descendants in Badong, and forming a separate inde-
pendent kingdom, after being for a time subject to Mengui.
Another and a more important change began about one
hundred yqars ago, in consequence of the war of Karang-Assem
against Klongkong. The cause of this war was the putting to
death of a prince of Karang-Assem, by command of the Deva
Agung. The prince in question did penance after the manner of
the Indian yogis; he gave himself up entirely to contemplation,
and, thereby, neglected all outward worldly things so much that
he grossly transgressed decorum e.g., he allowed his excrement to
166 AN ACCOUNT OF
fall where he happened to be sitting. When he was in Klong-
kong, he conducted himself in the same manner, and thus
offended the Deva Agung and the nobles of the court of Gelgel.
On his return journey, he was killed from an ambush by command
of the Deva Agung. He left three sons, who immediately re-
solved to avenge his death. The penance performed by their
murdered father rendered their power irresistible in the eyes of
the Balinese; the real fact is, however, that the race of Gaja
Madda, which then possessed all the extensive country of Mengui
and Karang-Assem, was the most powerful in Bali. They de-
feated the Deva Agung and destroyed his royal seat in Gelgel.
The Deva Agung retained his territory, however, and seems, from
this time forward, to have fallen into the state of dependence
under the Karang-Assem family in which we now find him.
Peace was restored by marriages, and Klongkong was held in
subjection. The wife of the Deva Agung last-deceased was a
princess of Karang-Assem, and governed the whole land for him
so completely that she even dared to murder another wife of her
husband, a princess of Badong. From this time the decline of
the power of the Deva Agung is principally to be dated. He was
a conquered prince, and, although he retained his territory, and
the conqueror remained in outward appearance his inferior, yet
his prestige among his own people was seriously lowered. In
addition to this, the princes of Kareng-Assem no longer per-
formed feudal service in Klongkong, but simply conceded to the
Deva Agung the title of first ruler of Bali, without paying him
tribute.
This victory had yet other important consequences for the
family of Karang-Assem. The conquerors of Klongkong could
without much difficulty also attack Boleleng, where dynasties had
already changed several times, and which at that time was cer-
tainly in a weak state. They took this country also, and one of
the brothers became king of Boleleng. At that time the most
ancient dynasty of the Kshatriyas had already ceased to exist in
Boleleng ; * the statements of the people of Badong asserting that
it was driven from the throne seven generations ago, retiring to
Badong, where it still lives, subordinate to the ruling Wesyas, but
yet of some distinction. (Its head is the Deva Made Rahi in
Kutta, who has obtained the chief command of that place from
the rajas of Badong.) One of the succeeding princes of Boleleng,
also of Wesyan blood (being descended from Arya Beleteng
in Pinatih), was Pauji ; he, however, did not hold the kingdom
long. Whether he expelled the Kshatriyas is not certain, nor
is it known whether the family of Karang-Assem immediately
succeeded him.
* This is open to doubt. In that case how can Kshatriyas have ruled in.
Boleleng in Crawford's time (1812) ?
THE ISLAND OF BALI. 167
The last exploit of the victorious brothers of Karang-Assem was
the conquest of Lombok. Here, also, one of the brothers re-
mained as prince, keeping five thousand Balinese families with
him, from whom the present Balinese population of Lombok have
sprung.
In the south, the Gusti family from Tabanan had, in the mean-
time, subdued the whole of the western portion of Badong,
namely, Pinatih : the eastern portion was conquered somewhat
later.
About the same time (three generations ago) the family of the
Deva Agung was also robbed of its possessions in Gyanyar, and
an inferior Deva (Pungakan\ named Deva Mangis, founded the
present kingdom of Gyanyar.
A hundred years later, therefore, we have the following state of
affairs in Bali : (i) The Deva Agung in Klongkong, only in name
still prince of all Bali, and with his territory reduced to Bangli and
Klongkong. Bangli, however, had its own princes, who were also
Kshatriyas and descended from the Deva Agung, but of lower
birth than the Kshatriyas in Klongkong. In Gyanyar the rela-
tions of the Deva Agung were conquered by a Deva of insig-
nificant rank. (2) The family of Gaja Madda ruling in Mengui,
Karang-Assem, Boleleng, and Lombok. Mengui, their original
country, had, however, already lost a large piece of territory to
the newly arisen Badong. (3) The family of Arya Damar in
Tabanan and in the newly founded Badong.
Since that time there have still been quarrels without end
among the eight States which we have mentioned. We say eight,
for Pahyangan was not a separate State, but belonged formerly to
Bangli, and now to Mengui, and Jembrana has also, always, or for
a very long time, been subject to Boleleng it was conquered
twenty years ago by Badong, but retaken by Boleleng.
The most frequent wars have been between Badong and
Mengui, with its allies Karang-Assem and Boleleng ; and between
Gyanyar and Mengui, allied with Bangli. The new kingdoms of
Badong and Gyanyar soon became allies, although they have had
a few small wars with each other, when one has been for and the
other against the party of the Deva Agung. In general, they main-
tain friendly relations with the Deva Agung, pay him homage, and
send him a few presents. To show the present condition of Bali,
we will now speak of each State in particular.
i. Klongkong, governed by the Deva Agung, is the smallest,
and is not a rich country. His men are said to number 6000.
Formerly there were members of his family in Nagara, Suka-
wati, and Pejeng (all in the present Gyanyar); the family
also had Boleleng, but it was driven out thence and went to
Badong. In Bangli, also, the Kshatriyan family is no longer
related to the Deva Agung, but, down to the most recent period,
1 68 AN AC CO UNT OF
the Kshatriyas of Bangli were always true followers of, and paid
homage to, those of Klongkong. At present, Bangli and Klongkong
are bitter enemies. Gyanyar, Badong, and Mengui acknow-
ledge the authority of the Deva Agung by presents and envoys.
Karang-Assem and Boleleng acknowledge him as supreme
prince, but pay him no homage, and, although they act in full
harmony with Klongkong, they do so as an entirely indepen-
dent State. Towards the Dutch Government, however, they make
use of the pretended power of the Deva Agung, in order to
represent their acts as controlled by the Deva Agung, and to take
refuge behind him. Tabanan, Bangli, and Lombok do not even
acknowledge the superior rank of the Deva Agung, much less give
him presents.
The name of the present Deva Agung is G'de Putra;* his sister,
the daughter of the above-mentioned princess of Karang-Assem,
is named Deva Agung Istri. The Deva Agung's mother was a
Sudra woman, but the deceased Deva Agung had no male
children by noble wives, and thus the son of a Sudra woman was
obliged to succeed him.
2. Gyanyar. This State is governed by Dei'a Pahan, a son of
Deva Mangis, who died in October 1847. The family is Ksha-
triyan, but of low descent (on account of too much intermixture
with Sudra blood), and is called pungakan (pungakan means
Jalleii}. Deva Pahan's great-grandfather, named Deva Mangis,
was the founder of this State. He was commander of 200 men in
the dessa of Gyanyar, and was under the Punggawas of the Deva
Agung, who were governors in Nagara, Sukawati, and Pejeng, and
relations of the ruling Deva Agung. By deceit, violence, and
poison he gained the mastery over these punggawas, and con-
quered from Mengui the country of Kramas. On account of his
infamous deeds, his poisoning, c., he is said to have changed
after death into a serpent, which was kept for a long time in the
palace at Gyanyar, but disappeared in the last few years. His
success in all his undertakings was probably owing to the fact that
he began in a time when Klongkong was defeated by Karang-
Assem, and deprived of all power. Gyanyar, however, has
submitted to the Deva Agung as the supreme ruler, and sends
* G't/e means the oldest son or daughter of the same mother among other
than royal persons, and among Brahmans the usual word for this is WayaJian
(old) ; Made is the second (really the middle one, Sanskrit madhya) ; Nyoman,
the third, if there be a fourth ; A''tuf, the third or fourth (really the youngest).
If there arc more children, the same expressions are used, but are placed before
the name instead of after it. Thus we have in Den Fassar a raja Ngr-urah
A"/*//, and a K'tnt Ngmrah, both sons of the deceased prince, the brother of
Kassiman. Putra is prince, but ought properly to be applied only to those of
purely noble birth ; in this case, however, the Deva Agung is the son of a
Sudra woman, but, as the only one who could succeed to the throne, he was
ennobled.
THE ISLAND OF BALI. i^
him numerous presents, which cause him to forget that his nearest
relatives are disgracefully oppressed for the former punggawas
are still living in Gyanyar, but they are under the command of a
Pungakan. It thus appears that, all over Bali, noble birth is not
sufficient to protect a family. Here, too, the stronger conquers,
<?ven though he be of the lowest extraction. Such a victorious
family is then again elevated by noble marriages. Gyanyar is
stated to contain 35,000 men, but not more than half this return
can be taken as true. It is one of the most fertile and best-
cultivated districts of Bali.
Gyanyar is allied with Badong, and acts as mediator in the
disputes between this State and Klongkong. Its attitude towards
Karang-Assem is neither friendly nor hostile. In the last Dutch
expedition against Boleleng, Gyanyar, by command of the Deva
Agung, sent 5000 men to assist ; they arrived too late, however,
and were not the best soldiers, but, on the contrary, the refuse of
the kingdom.
Gyanyar has had many quarrels under the three princes bearing
the name of Deva Mangis, chiefly with Bangli and Mengui, but
also with other States. At the conquest of Mengui by Badong,
Gyanyar was allied with the latter, and received a piece of terri-
tory on the frontier, Kadewatan. Gyanyar's friendship towards
Badong is not to be relied on ; hence the new campongs built by
the raja Kassiman on the frontier of Badong.
3. Bangli. The reigning prince is Deva G'de Putu Tangkeban*
His family also is descended from the first Deva Agung in Klong-
kong, but in rank it stands lower than the line of the Deva
Agungs. Formerly this family ruled over two States Bangli and
Taman-Bali. This close relationship was not sufficient to prevent
the prince of the former State from seizing the latter and extir-
pating its princes. The same thing has, however, occurred to the
families of Karang-Assem and of Lombok. Until about ten years
.ago, Bangli was attached to Klongkong ; the prince of Bangli, the
most warlike in Bali, was Commander-in-chief (Senapati). Its
enmity with Klongkong was brought about by the influence of the
princess of Karang-Assem, who was married to the deceased Deva
Agung. In its many wars with Karang-Assem, Boleleng, and
vGyanyar, however, it lost part of its territory in the north, and
lately also Pahyangan, which lies south of Bangli, and is now
divided from Bangli itself by a strip of territory acquired by
-Gyanyar. Pahyangan has been won by Boleleng and ceded by
this State to the Deva Agung, who again has handed it over to the
government of the prince of Mengui, his Punggawa.
Bangli no longer acknowledges the Deva Agung, and has
entered into friendly relations with Tabanan and Badong, which,
* Putu is applied to a person at whose birth the grandfather or grandmother
,is still livinc.
T 70 AN AC CO VNT OF
however, the surrounding hostile States render of little account.
The war with Gyanyar has recently been stopped through the
influence of Badong. The country has also suffered nothing
from Karang-Assem and Boleleng since the first Dutch expe-
dition ; now, however, it is in great danger from the union of
these two States and of Klongkong and Mengui. The men of
Bangli are the bravest in Bali, and it is only by virtue of this
quality that they have been able to hold their own against so
many powerful enemies. Women also bear arms in this country.
There are only fifty firearms in Bangli.*
4. Mengui. The reigning prince is Anak Agung K'tut Agung.
He murdered his elder brother, who was the first prince; the
widow of the latter, yang Agitn^ however, still has much
influence. The families of Karang-Assem, Boleleng, and Lom-
bok are of Menguian extraction. Patih Gaja Madda was
governor of Mengui. According to some accounts, this Patih
Gaja Madda was an incarnation of Vishnu, who has thus also
assumed the body of a Wesya. The story probably originated
when the family had subdued nearly the whole of Bali and
Lombok. Another account says that Patih Gaja Madda vanished
from the earth and left no children, the house of Mengui being
descended from Ki Yasak, who married the granddaughter of
Arya Damar against the will of her father, Arya Yasan. In
ancient times Mengui included the whole of western Badong,
while the eastern portion, the State of Pinatih, also acknowledged
the supremacy of Mengui among the descendants of Arya Bele-
teng. Besides this, Kramas and Kadevatan of Gyanyar and
Marga, in the mountain range of Tabanan, formerly belonged to
Mengui. Kramas was conquered by Gyanyar, as we have seen ;
* The above, like all the rest of this account, was written in 1848. The
third expedition despatched by the Dutch Government against the hostile States
in Bali in 1849 nas ver y much altered the position of several princes. The
author has thought it desirable, however, not to alter his work on this account,
but to describe Bali in the state in which it was when he left it. Now that
Boleleng, Karang-Assem, and Bangli are better known, the public will of
course have fresh information regarding them ; hitherto, however, but little
that is authentic has come to the author's ears, and he therefore gives the infor-
mation which he gathered, to serve as far as possible as historical data. It is
well known that the prince of Bangli, the faithful ally of the Dutch Govern-
ment, has now, in addition to his own country, received from the Government
the State of Boleleng, to be ruled under the sovereignty of Holland ; also thai
Karang-Assem has been given to the prince of Mataram-Lombok, who believed
he had a well-founded claim to it. The authority of the Deva Agung has also
been still further lessened, both because he has been compelled to acknowledge
the sovereignty of Holland, and because two princes, who previously did not
recognize his authority, have now become his powerful and dreaded neighbours,
who will always be ready and able to hold him in check. To avoid repetition,
we need here only allude to the fact that the princes hitherto reigning in
Boleleng and Karang-Assem, and the still better-known Gusti Jelanteg, have
fallen in the struggle.
THE ISLAND OF BALI. 171
we shall speak of the rest in connection with Badong. Mengui
was for some time, about twelve years, under the dominion of
Badong, but is now again free from that kingdom and stands as a
separate governorship under Klongkong. The house of Mengui
is not only of the same origin as that of Karang-Assem, but is
also allied to it by a very recent marriage between the prince of
Karang-Assem and a princess of Mengui. Mengui must obey
blindly the orders of the Deva Agung ; it is, however, hostile to
Badong and Tabanan, and is nearly neutralized by its position
between these two States. The small piece of territory belonging
to Mengui on the sea-coast is very much coveted by Badong,
which, by obtaining it, would touch the frontier of its ally
Tabanan. This piece of territory, however, is strongly defended
by its rocky coast, which only leaves open a small path.
The hostility which exists among the Balinese is shown by,
among other things, the diversion of water ; Mengui, for instance,
has dug a new bed on its territory for a river which formerly
flowed into the sea in Badong, and by this means has left dry the
rice-fields of Grobokkan on the borders of Badong.
5. Karang-Assem. The reigning prince is Ngrurah* G'tfe
Karang-Assem. The family is that of Gaja Madda, and the
country was conquered by Mengui more than two hundred years
ago. A list of the names of the princes of Karang-Assem is given
in the li Pamendanga," a work which we have mentioned above,
but nothing is stated as to their acts. In the " Usana Java," the
governorship of Dawuh in Karang-Assem is held by Arya Man-
guri; it is not certain whether that family had the whole of
Karang-Assem possibly part of it was in the possession of
the Deva Agung. By the conquest of Karang-Assem by the
house of Mengui, Klongkong was cut off from Boleleng, and the
powerful royal family in the conquered State afterwards found it
an easy matter to subdue Boleleng by itself.
No State has waged so many wars as Karang-Assem. We have
spoken above of the victory over the Deva Agung and the destruc-
tion of Gelgel. The consequence of this was the subjugation of
Boleleng and Lonibok t and the family would perhaps have ended
by making itself master of all Bali, Lombok, and SSmbawa (Sem-
bawa was actually attacked, but was spared through the interven-
tion of the Dutch Government) but for its numerous civil wars.
Many of the princes of Karang-Assem and of the conquered
Boleleng were expelled by members of their own family. In
* Ngnirah means something that overshadows, palindongan, a payong, and
also the vault of heaven. The princes of the Wesyan race nearly all bear this
title ; they overshadow and protect the land. The prince of Mengui has not
this title, for his country belongs to Klongkong, and is merely entrusted to
the present prince as a fief of Klongkong : Ngnirah, Angrura/t, Anglurah, is
in Java Lurah, a chief of the fourth rank.
172 AN ACCOUNT OF
Lombok, also, the two princes of the house of Karang-Assem
were at war with each other nine years ago, with the result that
the chief prince, that of Karang-Assem-Lombok, was defeated by
the prince of Mataram, and committed suicide. Of the Karang-
Assem-Lombok family two children were saved, a son and a
daughter of the last prince, and they are now living in Karang-
Assem-Bali.* The rest of the family, including the women
murdered each other in true Indian fashion, in order not to
survive the shame of defeat. They even wished to murder also a
European who had sided with them, in order to go to heaven
(Svarga) all together. This method "of quitting life by the
members of the family murdering each other " is also regarded in
Bali as a J5e/a, and here also the Indian meaning of the word wela
("sudden and easy death," see above) is applicable. Since the
fall of Karang-Assem-Lombok, the princes of Karang-Assem-Bali,
of Boleleng, and the Deva Agung, who is independent of them,
have been enemies of Lombok, and do not acknowledge each other
as lawful rulers. The prince of Lombok, which is also called
Sclaparang, is Ngrurah K'tut Karang-Assem.
Karang-Assem is the most mountainous country of Bali, and
grows little rice, but the dense population is very skilful in manual
work, especially in wood-carving, whereby they gain their liveli-
hood. According to the statements of the Balinese, it contains
50,000 men able to bear arms.
6. Boleleng. The prince is Ngrurah Made Karang-Assem.
The family comes from Karang-Assem, and thus is originally
descended from Mengui and Patih Gaja Madda. The present
prince is a brother of the prince of Karang-Assem. Many
dynasties have ruled in Boleleng. Seven generations ago the
Kshatriyan princes of Boleleng, ^relations of the Deva Agung,
were expelled by a Wesyan family; to the latter belonged
Ngrurah Panji, a descendant of Arya Beleteng. The surviving
Kshatriyas of Boleleng now live in Badong. Boleleng was finally
conquered four or five generations ago by Karang-Assem, but this
did not put a stop to the wars, for the members of the Karang-
Assem family could not agree together. The most profound
peace reigns now, since two brothers have filled the thrones of
Karang-Assem and Boleleng. According to Balinese statements,
Boleleng has but 1 2,000 men capable of bearing arms ; perhaps
so few are returned in order to represent its conquest by the
Government as of little importance.
The well-known Gustijdanieg is a cousin of the prince; his
father was murdered by the late prince ; yet the son now reigns,
whilst the lawful king is but a shadow.
* Perhaps these sole descendants have now been killed by the prince of
Mataram- Lombok ; at all events he was formerly in constant dread of these
lawful heirs of Lombok.
THE ISLAND OF BALI. 173
7. Tabanan. Prince Ratti* Ngrurah Agung. The family is
that of Arya Damar, which is said to have remained pure,
although here this is at any rate improbable, and in Badong is
untrue. Tabanan does not engage in many wars ; it has suffered
defeat a few times from Boleleng, but has never been entirely
conquered. In the war with Mengui, in which the whole of that
State was conquered by Badong and its allies, Gyanyar and
Tabanan, Tabanan received the mountain-district of Marga. The
more recent quarrels with Mengui are of no importance and never
result in anything. The men of Tabanan understand the art of
war much less than the other Balinese. Two men of Tabanan are
calculated to be no more than a match for one of Mengui, and
the people of Badong add to this that one man of Badong is equal
to two of Mengui. f The men able to bear arms are stated to be
100,000 in number (?). Under the prince of Tabanan stands that
of Kediri, a relation of the former, ruling over a tolerably large
territory.
Another Punggawa of Tabanan is the prince of / the Marga
already mentioned ; the latter is not a Wesya, but a Sudra. His
ancestor was a seller of palm-wine (tuak), who managed to gain
the favour of the prince of Mengui, and by him was made Pung-
gawa. When the territory was transferred from Mengui to
Tabanan, the prince of Marga retained his position. (His district
grows most of the coffee in Bali.) This ^instance is the only one
which has come to my knowledge of a Sudra prince, but it indi-
cates the decay of the institution of caste. One often hears the
nobles say that Bali must go to the bad now that the Sudras or
children of Sudra mothers become kings. Compare Deva Agung)
raja Pan? chuttan.
8. Badong. The three princes who together rule this State
are (i) Ngrurah G'de Pam'chuttan, (2) Made Ngrurah in Den
Passar, and (3) Ngrurah G'de {Kassiman). This State has been
formed gradually. The western portion formerly belonged to
Mengui, and the eastern, the State of Pinatih, to the descendants
of Arya Beleteng, who acknowledged the supremacy of Mengui.
Pinatih lies to the eastward of Kassiman, from the frontier of
Gyanyar to Tanjang, opposite Serangan (at the place where the
roadstead is on the eastern side). It comprises Gunung Rata,
Sanor, Taman Intaran, Soong, and the island of Serangan, and is
a very fertile district. The poorer western portion with Gro-
bokkan, Legian, Kutta, Tuban, Jembaran, and the southern
corner of Bali (the point of the table-land called by the Balinese
* The title of ratu is always used in addressing princes, but it is only used
before the name in speaking of especially distinguished princes. The raja of
Tabanan has lately adopted it, and also Saving- Adi in Pam'chuttan.
t Recent experience during the third Dutch expedition does not confirm
this assertion, Badong having, as is well known, been defeated.
174 AN A CCO UNT 0*
bukit, the mountain-range), were immediately under Mengui, to
which State also belonged the P'kcn Badong, a much-frequented
place of trade. The founder of the State of Badong was a Gusti
of the royal house of Tabanan. In a manuscript which was lent
to me, and in which a brief enumeration of the names and mar-
riages of the princes of Badong was written at the end of another
work, he was called Gusti Ngrurah Bola ; he had settled in
Tabanan in the kampong of Buahan (Buah zn&Jambe mean the
betel-nut, pinang), and is therefore called (i) Anak Agung ring
Buahan burnt Tabanan (the prince in or of Buahan in the land of
Tabanan) ; he was the younger son, and sought for a place.
From him to the raja Kassiman, who now has the supreme
government, there are ten generations, but until we come to the
great-grandfather of this man, their history is little more than a
list of names).* He went from Tabanan to the P'ken Badong,
and lived there in the house of Hi Sedahan, a Sudra (the name
Sdah means sirih ; this name renders the matter somewhat open
to suspicion ; was it necessary exactly that the pinang (bush)
should come to the sirih ?). He thus had no palace, nor does he
seem to have had a governorship, nor a fixed residence in Badong,
for his son or grandson again comes from Buahan to Badong.
His sister married the Gusti Agung, the prince of Mengui, but left
no children. The reason of his departure is said to be that he
went to seek a governorship in Mengui, an object which his son or
grandson seems for the first time to have attained.
2. Anak Agung K'tut Mandesa : this prince, the son or grand-
son of Ngrurah Bola, went from Buahan, in Tabanan, to the
Gunung Batur, the mountain which vomits fire, on which Dewi
Danu or Gangga is worshipped. This was evidently a pilgrimage ;
he did penance on the sacred mountain, in order to obtain earthly
power. Thence he came to Badong and lived in the house of the
M'kel (B'kel] Tinggi, to the south of the place of cremation of the
present Pam'chuttan, in the dessa of T'gal. His surname, Man-
desa, is said to have been given him because he resided at first in
the house of a mandesa (it is possible that he himself was nothing
more than a mandesa, a kampong-chief). With the aid of the
M'kel Tinggi he soon managed to gain a large number of fol-
lowers, the result of his penances, and became a Punggawa of
Mengui.
3. Anak Agung Pededekan, son of the last mentioned ; he also
appears to have been a Punggawa.
* The date of a war with Sideman (1582, corresponding with 1660 A.n.) is
found in the manuscript of the Wriga Garga, which was lent to me ; in that
year the men of Karang-Assem (Sideman) with their allies, Mengui, &c., had
attacked and invested Badong, but had not conquered it. This is the only
historical fact chronologically determined which came to my knowledge in
Bali. It appears to have occurred in the time of Gusti Nyoman T'geh.
THE ISLAND OF BALL 175
4. The three sons of the last are Gusti Wayahan Tgch, Gusti
Nyoman Tgeh and Gusti K'tut K'di. The second, Gusti Nyoman
T'geh, is the ancestor of the succeeding princes, and increased the
power of the house. He married a woman of noble family (prami)
of Buahan. He was brave and cunning, and had a body of picked
troops. One of his wife's sisters was married at Klongkong to the
Dal em (Deva Agung), and died as a Satya; and the other married
.in Mengui the Gusti Agung, and became the ancestress of all the
Gusti Agungs (princes) of Mengui. These noble alliances and his
personal qualities added to his influence, and he seems to have
possessed the whole of ancient Badong from Abian-Timbul to
Glogor, Pam'chuttan, and Kassiman. It is not certain when the
wars with Mengui began, but probably he and his son, and grand-
son after him, were Punggawas of that State.
5. Gusti Ngrurah Jambe Mihik (he is named Jambe, because
his mother was from Buahan, or Jambe, in Tabanan).
6. The two sons of the last mentioned, by one mother, are
Anak Agung G'de Galogor and Anak Agung Tlas ring batu
Krotok. Their mother was from Panataran, a place in the present
Pam'chuttan ; the Aryas Panataran were at that time still Wesyas,
but afterwards they were degraded to Sudras. His wife was also
from Panataran, and he seems to have founded Pam'chuttan, or to
have made it his residence. Pam'chuttan is derived from p'chut,
an ox-whip ; the descent of the race of Pam'chuttan from an ox-
driver is found in Abdullah (" Tijdschrift," vii. 2, p. 166 sqq.}. It is
admitted in Badong that the wife of one of the ancestors was of
humble origin, the daughter of an ox-driver, but it does not appear
that the wife of Anak Agung T'las ring batu Krotok was of such
low birth ; had she been so, her son Ngrurah Sakti Pam'chuttan
would not have attained such great distinction and power.
At this time there were princes' capitals in T'gal ; this is the
most ancient, and was founded by the second prince. T'gal lies
to the south of Pam'chuttan the principal family, of course,
resided here ; thence was descended the family in Pken Badong,
which had palaces both here and in Kshatriya, north of Den Passar.
Galogor, to the north of Pam'chuttan, was also a capital, founded
by the elder brother of Anak Agung T'las ring batu Krotok ; the
family still exists, and has probably been spared on account of its
near relationship with the line of Pam'chuttan. The descendants
became Punggawas of Pam'chuttan, and afterwards of Den Passar.
Pam'chuttan, finally, the capital of Anak Agung T'las ring batu
Krotok, was at that time the seat of the younger line, which, how-
ever, was soon to unite the whole of Badong.
7. The sons of Anak Agung Tlas ring batu Krotok are called
putras, princes. They are (a) Anak Agung ring Pam'chuttan,
also called Ngrurah Sakti Pam'chuttan, (b} Gusti Made Tgal,
(t) Gusti K'tut T'labah. Of the last two nothing is known.
176 AN ACCOUNT OF
The first is he who founded the power of Pam'chuttan.
He had obtained that great power by means of a creese
pussaka, 1 called singha thence his name Sakti (supernatural
power). He subjugated the most ancient capital of the
princes of Badong in T'gal, and waged successful wars against
Mengui, from which he wrested the territory from the present
frontier of Mengui to the point of the table-land. He appears to
have been the first who was regarded in Badong as an independent
prince. He had five hundred wives ; the principal ones (prauii)
were from Tangkeban (Bangli, thus an intermixture with Kshatri-
yas), Galogor, and Mengui. From this prince are descended eight
hundred Gusti Pam'chuttan, who, on account of their near rela-
tionship, are regarded as the support and strength of the land.
But where brother is ready to fight against brother, such a strength
is imaginary. These eight hundred Gustis have a common sanc-
tuary in Pam'chuttan, where they must appear once a month,
and in case of absence have to pay a penalty.
Besides Pam'chuttan, the kingdom of the Jambes also existed at
that time in P'ken Badong and Kshatriya (the last being merely
a country residence of the prince). They also gained distinction
by subduing the kingdom of Pinatih, the eastern portion of
Badong. They were still of importance at that period, and really
of nobler birth than the princes of Pam'chuttan. Galogor had
transferred itself to Pam'chuttan as Punggawa.
Ngrurah Sakti's principal sons were :
8. (a) Ngrurah G'de Patrfchuttan, devata di Ukiran (who died
in Ukiran) ; from him are descended the present princes of Pam-
'chuttan, of whom we shall speak hereafter, (b) Ngrurah Mayun,
in the palace of Mayun* on the opposite side of the river to the
east of Kassiman. This palace no longer exists ; all the materials
of which it was built have been taken to Den Passar. (c) Ngrurah
Kaleran, in the palace of Kaleran ; to the north of Pam'chuttan,
which position is also indicated by the name kaleran (northern).
Little is recorded of these princes ; of course they, too, had wars
with Mengui and other States, but they have done no prominent
acts, and are overshadowed by the fame of their father and their
descendants. Pam'chuttan remained and still remains the chief
seat of the family ; the prince of Pam'chuttan alone can obtain
the Abbisheka, that is, can be anointed as prince of the whole realm.
9. The most remarkable of the descendants of the three sons of
Ngrurah Sakti was Ngrurah Made Pairfchuttan, the son of Ngrurali
Kaleran. This chief married the daughter of Ngrurah Mayun,
thereby uniting two portions of the possessions of the Pam'chuttan
1 \_Pusaka, an heirloom.]
* Mayun is ihe same as Made, "the middle or second son (or daughter)."
This name now no longer exists in Badong ; the second prince in the family of
Pam'chuttan is called Made.
THE ISLAND OF BALI. 177
family, and founded the palace of Kassiman. Not content with
this, he attacked the Jambes in P'ken Badong and Kshatriya, and
conquered their territory after a severe struggle. He was supported
by Pam'chuttan and Gyanyar, whilst the Jambes received aid from
Mengui. Great must have been the slaughter in the palace of P'ken
Badong it is spoken of with terror to this day. The fate is also
lamented of the murdered Raja Jambe, who, the people say, was
entirely blameless, and had given no cause for the war. He had
his revenge, however, upon his conqueror, according to the belief
of the Balinese, for he was born again in the family of his foe, as
his grandson, and the one of noblest birth, a circumstance which
was an omen of great misfortune to that family. The conqueror
began to build the great palace of Den Passar, but died before it
was half finished ; he had already taken up his residence, however,
in the new palace, and in him began the line of the princes of Den
Passar : the palace in Kassiman was still inhabited afterwards by
his wife from Mayun, and was finally given up to his second son
(according to birth), the still living Raja Kassiman. This prince,
for the sake of distinction, is called devata di made, <: died in the
middle" 1 (made the middle is here Den Passar,f which, both
from the rank of its prince and from its situation, comes between
Pam'chuttan and Kassiman). His numerous quarrels with Mengui
and nearly all Bali have been without result. Even Tabanan has
once fought against him, at the desire of the Deva Agung, and, to
save appearances, burnt a single kampong; in reality, however,
Tabanan and Gyanyar have always remained friendly to Badong,
but they were obliged for political reasons to assume the appear-
ance of hostility against their ally. Badong has neither gained nor
lost territory under his rule, while it has become an independent
State. The quarrels with the other States were caused chiefly by
the aggressions of the Jambes.
10. The sons of this prince, besides many of lower birth, were
(a) Ngrurah Made Pairfchuttan in Den Passar, (b) Ngrurah
Kassiman in Kassiman, (c] Ngrurah Jambe, who lives near Den
Passar. All three are by different mothers; the first is by a
mother from Pam'chuttan, daughter of the Devata di Ukiran (Ngr.
G'de Pant chuttari) and a princess of Tabanan. This prince was
younger than the prince of Kassiman, but as he was born of a
Raja-woman he took the highest rank among the sons. Kassiman,
the old prince who still lives, is the son of a Gusti-woman of
Pam'chuttan. Ngrurah Jambe is the son of the daughter of the last
prince of Kshatriya, who was forced to marry the conqueror and
murderer of her father. By birth he would be more noble than
1 [Devata, in the sense of dying (lit. being deified), is only used of princes.]
t Den Passar means, north of the Passar, or, still better, on the further side
of the Passar ; thus we have also Den Bukit, " on the further side of the
mountains," as another name for Boleleng.
SECOND SERIES. VOL. II. N
1 78 AN A CCO UNT OF
Kassiman, and equal to Pam'chuttan, but the descendants of a
conquered prince can never again acquire rank in Bali. His
noble birth is acknowledged, but he can make no claim to the
throne.
The prince of Den Passar, called after his death devata di Ksha-
triya, was an ally of Gyanyar and Tabanan. These three began
a fresh war against Mengui, which was carried on more by artifice
than by force of arms ; the Punggawa of Marga, for instance, who
at that time was subject to Mengui, being induced to surrender to
Tabanan. For fear that he would lose all his territory in this
way, the prince of Mengui gave his land in fief to the prince of
Den Passar, and remained in possession of Mengui as Punggawa
of Badong ; he only lost Marga to Tabanan, and Kadewatan, a
small piece of territory on the frontier, to Gyanyar. After this
arrangement, the four southern States were allied together against
Karang-Assem and Boleleng, the old enemies of Badong and
Tabanan, whilst Klongkong remained neutral. This state of
affairs continued until shortly after the death of the prince of Den
Passar, which took place in 1829.
The prince of Den Passar continued the building of the palace
at that place, but did not finish it ; we see it now in the state in
which he left it. Most of the building materials had to be found
by the conquered State of Mengui, where timber was obtained
from the mountains, this article being very scarce in Badong. He
had several noble wives, but his only son of noble birth, Ngrurah
G'de Putra, died a short time before his father. This was the one
already mentioned, who was considered to be the last raja Jam be
born again, which belief his own father shared. He was regarded
as certain to bring misfortune upon the family, and it would seem
that he did not die a natural death. In the compact with Mengui,
the prince of that land had declared himself a vassal of Ngrurah
Made Pam'chuttan and his son Ngrurah G'de Putra. The death
of both without a previous renewal of the compact enabled Mengui
to regard itself as discharged from its obligations towards Badong,
and it soon, in fact, withdrew from them.
After the death of Ngrurah Made Pam'chuttan, Kassiman was
the only prince of importance of the family of Den Passar, and he
thus gained the supremacy in Den Passar and Kassiman. During
his brother's lifetime these two nearly came to blows ; Kassiman
had already placed his country in a state of defence, but this civil
war was prevented by the intervention of Ngrurali G'de Pam-
'chuttan, the then prince of Pam'chuttan.
The eldest surviving son of Ngrurah Made Pam'chuttan was
Ngrnrah G'de Oka. He would have become prince of Den
Passar, but he would not acknowledge the supremacy of Kassi-
man. Kassiman, in conjunction with the prince of Pam'chuttan,.
compelled him, however, to leave the country, and banished him
THE ISLAND OF BALL 179
to Tabanan. This did not prevent him from acting against Kassi-
man. From Tabanan he went to Mengui, and, both here and in
Bangli, gained friends who were willing to support him. To
strengthen his party still further, he released Mengui from its
vassalage to Badong, under the pretext of being his father's heir,
and gave it to the Deva Agung. The latter hastened to make use
of this gift, and was able to do so without scruple, because the
prince of Mengui had only sworn allegiance in the contract with
Badong to the deceased princes Ngrurah Made Pam'chuttan and
his son Ngrurah G'de Putra. The Deva Agung then commanded
that Ngrurah G'de Oka should be received again in Badong, and
this command was obeyed, for this prince had made his appear-
ance with a numerous army from Mengui and Bangli. Ngrurah
G'de Oka afterwards carried off Kassiman's only daughter, and
took her to wife. Old Kassiman again made use of this to con-
firm his power over Den Passar : he was now in fact the prince's
father. Not long after this marriage, however, Ngrurah G'de Oka
died also. The sudden deaths of this prince, his brother, and his
father lead us to suspect unnatural means, but I cannot assert,
nor would I willingly believe, that they were applied by Kassiman.
This old man, it is true, took the best advantage of circumstances,
not only in Den Passar, but also in Pam'chuttan, of which we are
about to speak. The present prince of Den Passar is Ngrurah
Made, who, against his will, acknowledges old prince Kassiman as
supreme prince in Badong, but, nevertheless, is independent and
endeavouring to increase his power.
Pam'chuttan, since the time of Ngrurah Sakti, has been the
chief seat of princes in Badong. The family of Den Passar, how-
ever, has, by its wars, acquired great fame, and under Kassiman's
father and brother has, in fact, held the supreme authority, not-
withstanding the nominally higher rank of the prince of Pam'chut-
tan. We have seen above, that the eldest son of Ngrurah Sakti
Pam'chuttan took up his residence in Pam'chuttan. His name
was (8) Ngrurah Gde Pavfchuttan devata di Ukiran (Ukiran is a
place in Pam'chuttan). He was succeeded by his- son (9) Ngrurah
G'de Pairfchuttan devata di Munchuk ; both were always allied
with their more famous relations in Den Passar, and this friendship
was maintained by marriages. The last prince of noble birth was
(10) Ngrurah G'de Pam'chuttan devata digdong; he was anointed,*
and played an important part in the wars of Kassiman's father
and brother. By command of this prince and Kassiman's brother,
his cousin Anak Agung Lanang crossed the sea with an army to
* The anointing of a prince, Abhisheka (Sans.), is performed by the priests
(the Guru lokas). In Badong it only takes place at Pam'chuttan. In order
to be anointed, the prince must be both of noble birth and instructed in all
religious duties. The prince of Pam'chuttan referred to in the text was a Rishi ;
he had attained the position of a saint by penances (maveda).
N 2
1 80 AN A CCO UNT OF
Jembrana, and conquered this country, which belonged to Boleleng.
Anak Agung Lanang went thither because the prince of Den
Passar wished to banish him from Badong. This fact also shows
the supremacy of Den Passar at that time. Jembrana was soon
lost again, however, and subsequently (n) Anak Agung Lanang
(about 1830), after the death of Ngrurah G'de Panrchuttan, who
left no sons, became prince of Pam'chuttan. He was not anointed,
but yet enjoyed greater renown than his son, the present prince.
He had no sons of noble birth ; he only had by a Gundik (con-
cubine) the two sons now called (12) Ngrurah Gde Pam'chuttan
and Ngrurali Made Pumchuttan. These, during their father's
lifetime, were his Parakans, who carried after him his sirih-box,
&c. After the death of Anak Agung Lanang, a great portion of
the Punggawas of Pam'chuttan would not acknowledge his son
as prince, on account of his low birth. Kassiman, however, who
in the meantime had also obtained the supreme power in Den
Passar, supported the new prince. Some Punggawas (e.g., Deva
Made Rahi, in Kutta) submitted, and a Gusti of Legian fled the
country and took refuge in Gyanyar. Kassiman then established
the new prince in the ancestral palace of Pam'chuttan, and his
authority, by marrying him to the daughters of Ngrurah G'de
Pam'chuttan, the cousin and predecessor of his father. These
women are the principal surviving members of the family, and
their illustrious birth enhances that of the prince, who is himself,
as it were, his wife's inferior. His principal wife's name is Sagung*
Adi, another is called Sagung Made, and a third Sagung Oka.
Sagung Adi has now assumed the name of Ratu, which we have
also seen was done by the prince of Tabanan.
Kassiman's intervention in the affairs of Pam'chuttan gave him
the supreme authority in this part of Badong also. He is re-
garded as the father of the princes of Den Passar and Pam'chuttan,
and uses circumstances very cleverly to keep those princes in
dependence.
The prince of Den Passar, Made Ngrurah, and his brothers
K'tut Ngrurah and Ngrurah K'tut, were all born of mothers of
low extraction, and, had they a brother of higher birth, would not
have the least claim to rule. Besides this, K'tut Ngrurah is in
opposition to Made Ngrurah, and asserts that he has a better
claim than the latter. He has several of the Punggawas on his
side, and Made Ngrurah is therefore obliged to invoke the aid of
his uncle Kassiman to maintain his position. Kassiman, how-
ever, it would seem, does not trust Made Ngrurah, who has
inherited much of his father's ambitious character : he therefore
does all he can to uphold the power of Pam'chuttan, and has
* Sagung means a princess, horn of a princely father and mother. Sayu,
one whose mother was only a Gusti-woman. Sagung is derived from
great ; Sayu from ayu, good, which we have also found to be a name for the
female Brahmans (Ltayu).
THE ISLAND OF BALL 181
enriched the prince of that country and invested him witli a
certain renown. The prince of Pam'chuttan, therefore, dares not
dispute Kassiman's authority ; he is of low birth and a peaceful
disposition, and would also have to fear the house of Den Passar,
which considers itself raised above so low a birth by the deeds
of its father and grandfather. The prince of Pam'chuttan has not
the abhisheka; although he is a man of about forty-five, he is not
. yet sufficiently instructed in his duties and in religion ; he has
performed his duties to the Pitaras, having, with all due ceremony,
burnt his father, and built a new domestic temple, which was
finished a year ago ; yet it appears that he will not receive the
abhisheka during Kassiman's lifetime, and, in the event of his death,
this would also, perhaps, be prevented by the opposition of Den
Passar. The prince of Den Passar has not fulfilled his duties
towards his ancestors ; his father and brother as well as other
chief members of his family are still unburnt, and are preserved in
the palace of Den Passar. On this account Made Ngrurah does
not reside in the palace. 1
This cremation must from the rank of the deceased be very
splendid ; the prince of Den Passar, however, has not the means
to do it, and although, apparently, the whole population of
Badong works for him, this produces very little visible result.
He has to complete the palace which his father left unfinished,
and in which much work has still to be done, and further to im-
prove the roads of his country, which have fallen into a very bad
state of repair since the death of his father, the last prince of
Pam'chuttan who had the abhisheka. And, finally, he cannot
easily raise the funds for a great cremation, and is opposed in his
undertakings by K'tut Ngrurah, and, in secret, by Kassiman. In
Den Passar also, therefore, it is probable that the position of
atfairs will remain unchanged till Kassiman's death. Yet another
reason against the cremation of the late prince of Den Passar and
his son G'de Putra appears to be the above-mentioned super-
stition, that prince Jambe has been born again in the family of
Den Passar in the person of G'de Putra. This second birth
indicates misfortune for the family, which fears either that that
prince will be born a third time, or that the curse which seems to
attach to the family prevents it from performing works pleasing to
the gods.
1 [The position of affairs, 35 years later, is thus described by Dr. Jacobs,
1. 1. p. 1 68 : " Badung, however small, is divided among three princes, each
of whom originally ruled his own portion, and contested the supremacy with
the other two, so that we find there three chief cities, Den Passar, Pam'chu-
tan, and Kasiman. The present Raja Kasiman, though still residing at
Kasiman, has deputed his power to the Raja Den Passar, so that actually the
latter shares with Rajah Pam'chutan supreme authority. Raja Den Passar
exercises sway over the whole of Badung, or is at least looked upon by our
government as exercising such sway, while also Pam'chutan is recognized as
a ruling prince.]
i8 2 AN ACCOUNT OF
GENEALOGICAL TABLE OF THE PRINCES OF BADONG.
( I ) Gusti Ngrurah Bola.
(2) Anak Agung K'tut Mandesa,
(3) Anak Agung Padidikan.
Gusti Wayahan Tegeh. (4) Gusti Nyoman Tegeh, Gusti K'tut K'di.
called Anak Agung di Pule.
(5) Gusti Ngrurah Jambt MihiL.
Anak Agung G'de Galogor. (6) Anak Agung t'las ring batu Krotok,
| in Pam'chuttan.
The princely family of
Galogor.
(7) Anak Agung ring Panic httttan, Gusti Made Gusti K'tut T'labah.
called Ngrurah Sakti Pam'chuttan, T'gal.
with 500 wives.
(8) Ngr. G'de Pam'chuttan devata di Ukiran. Ngr. Mayun Ngr. Kaleran ;
his daughter, Ngr. |
Putu, married Ngr. Made Pam'-
chuttan, devata
di Ukiran.
I I
(9) Kgr. G'de Pamchuttan devata di Anak Agung Lanang.
Munckuk.
(10) Ngrurah G'de Pam'chuttun (n) Anak Agung Lanang. See thelbllow-
devata di G'dong. ing table.
(12) Ngr.G'de Pam'ch. Ngr.MadePam'ch.
Daughters : Ngr. Putu, the only
Sagung Adi. Sagung Made. son, by a mother of
Sagung Oka. Sagung Raka, &c. low extraction.
THE ISLAND OF BALI.
83
TABLE OF THE FAMILY OF KALERAN DEN PASSAR.
(i) Ngntrah Kaleran.
(2) Ngr. Me
Married to
dePanfch. Anak Agung Rahi.
Ngr. Putu In Kaleran Kawan.
Gusti alit Pam'ch. Gusti Ngr. K'tut.
In Kaleran Kangiman. In J ro TgaL
Mayun (devata di |
I
Made).
From whom is de-
scended the family
From whom is de-
scended the family in
The present
prince's name
of Kaleran Kawan
Kaleran Kangiman.
is Gusti Ngru-
and Kutta. The
present prince's
The present prince's
name is also Gusti Alit
rah T'gas.
name is also Anak
Pam'chuttan.
Agung Rahi.
^
All these are Punggawas of Den Passar.
1
1
(3) Ngr. Made Pan? chuttandevatadiKshatriya.
Mother of Pam'chuttan, wife of Pam'chuttan,
Ngr. Kassiman, now
named Ngr. G'de,
Ngr. Jambe,
mother of Ksha-
and of Tabanan extraction.
ruler of Badong.
triya, wife of
Mother, a Gusti
Pam'chuttan.
woman ; wife, Sa-
I
gung Raka.
Anak Agung
Alit G'de.
I 1
Ngr. G'de Putra + Ngr. G'de Oka. + (4)
Made Ngrurah. K'tut Ngr.
Ngr. K'tut,
probable suc-
cessor in Kassi-
man.
Other brothers of Kassiman are :
Ngr. G'de Dangin.
An. Ag. Alit Griha.
Devata Anak + Devata Banyar.
Agung Dauh,
wife of Pam'.,
became Bela.
An. Ag. Rahi
(in Belaluan).
I
Ngrurah Kajanan.
An. Ag. Raka pupuan
Ugung.
An. Ag. Put. Merangi
and An. Ag. Raka.
An. Ag. Alit
Raka.
All these are by mothers of low caste ; the family of Belaluan
has again raised itself in rank by means of^noble marriages. The
others, after one more intermixture with ^udran blood, will sink
into the position of ordinary Gustis.
FURTHER REMARKS ON THE CASTES.
The Dewas, Gustis, and Idas are much too numerous in Bali
to maintain their dignity; they, and especially the first two classes,
are too proud to gain their livelihood by work, and prefer unjust
1 84 AN A CCO UNT OF
privileges. They rob the people without limit ; they are the cause
of the plundering of ships, and of the extortions to which foreign
traders are often exposed. The high-caste princes are seldom
guilty of such misdeeds, but they wink at the robberies of their
relations, and it is difficult to obtain justice from the princes
against these pests of the land. The Gustis in the countries
ruled by Wesya princes, and the Devas in those where Kshatriyas
reign, have properly all the official posts about the prince, but,
although they derive much honour from these, they get little pay.
FEUDAL SYSTEM.
As Raffles has observed, the condition of Bali very much
resembles that of Europe in the Middle Ages ; there is a feudal
system throughout the land. The Deva Agung must be regarded
as the supreme feudal lord ; in ancient times he was so in reality
(see "Usana Java," above). How this is modified now, we have
already seen. He still calls the other seven princes, and also the
prince of Lombok, his Punggawas* which in Bali conveys the
idea of vassal. The rest of the princes (Kshatriyas and Wesyas)
have subdivided their land among the members of their families,
and so we find in Badong the princes of Kaleran Kawan, Kaleran
Kanginan, Ngrurali T'gas, the prince of Galogor, and Kassiman's
brother with their children as Punggawas of Den Passar (formerly
most of them were under Pam'chuttan, but they have preferred to
acknowledge as their lords the two warlike princes of Den Passar,
Kassiman's father and brother ; unless Kassiman had interfered,
even the sovereignty over Badong itself would have been trans-
ferred from Pam'chuttan to Den Passar). Now they are all under
Kassiman, but only in so far as Den Passar also is under his
sovereignty. The real Punggawas of Kassiman, in his small
original territory, are few in number, and, for the most part,
merely the descendants of his brothers, who were entirely without
means.
Under these princes, who are Punggawas of the highest princes
and most closely related to the royal families,! stand the rest of
the Gustis, also as Punggawas; these also have an undefined
authority over the men whom they rule, and have even the power
to punish with death ; the prince above them and the prince of
the whole kingdom do not trouble themselves about their subjects
further than their feudal duties are concerned. The Diaksas \
* Punggaiva, "bull" and "excellent" in the Sanskrit.
"f We do not find many exceptions to this rule ; in Tabanan the prince is
descended from Marga ; in Klongkong, e.g., Ngrurah Pinatih, from Arya
Bleteng. In Gyanyar from Sukawati.
I write Diaksa instead of Yaksa (as the judges are usually called in Bali
and Java), on the authority of a manuscript, where the writing with the
second d reveals the origin of the word. Diaksa^ which is also found as
THE ISLAND OF BALL 185
have merely to pronounce judgment between the Siuiras and
sometimes between Gustis of equal rank ; in all cases concerning
the distinctions of caste and feudal duties the princes and
Punggawas are judges ; in spiritual matters, however, the Padan-
das act in this capacity.
The primary feudal duty, as in the Middle Ages, is service in
war ; and further, the Punggawas and their subordinates have to
furnish assistance in all public works and festivals of the prince,
and the lower orders also have to carry out all the works of the
Punggawas. The people, under the guidance of the Punggawas,
have to build the princes' palaces and places of cremation, to
repair the roads, and besides this to contribute, mostly in kind,
towards the expenses of all offerings, family feasts, and cremations.
The direct taxes are very unimportant ; the common man pays a
small tax on garden land, and a little more on sawahs. The
princes, therefore, cannot be rich, unless they possess considerable
private means ; they are powerful, however, so long as their names
hold the Punggawas in subjection, and they can therefore cele-
brate their splendid feasts and cremations without cost to them-
selves, and sometimes even with advantage to their private
treasuries, their faithful vassals zealously contributing to these
ceremonies. (One of the reasons why the great cremation in Den
Passar does not take place is, that the present prince is not in
very high esteem with the Punggawas, and that not enough is
contributed towards it. Besides this, Kassiman retains the
revenue.) To the revenues of the princes and the Punggawas
belong also the duties on commerce, the customs-duties, and the
bridge or road- tolls.* Trade especially produces a considerable
revenue for the prince, and has made the princes of Badong com-
paratively rich, above all Pam'chuttan and Kassiman. Karang-
Assem also makes a great deal by it. These imposts affect only
the lower orders, and the Chinese, Buginese, and Europeans.
The Gustis, Devas, and Idas who carry on trade pay no duty
upon it. The feudal lords, princes, and Punggawas still do
something for the people they give them water, and the making
of canals and the effective irrigation of the rice-fields are their
duty ; in return for this, however, they draw a small revenue from
the rice-growers.
Adiaksa in the MSS., is Skt. adhi and aksha, an inspector, protector (see
Wilson). Yaksa, on the contrary, is a sort of demon, allied to the Raksasas.
[See Roorda's "Javanese Dictionary," s.v. jaksha, where the etymology from
adhy aksha is correctly given.]
* I am only acquainted, however, with two bridges in Bali (excepting those
over small brooks) in any way worthy of the name one at Kutta, built by
Europeans with Balinese aid, and one, very dangerous for want of planks, over
a rocky chasm in Tabanan.
1 86 AN A CCO UNT OF
SUDRAS.
The fourth caste, the Siidras, have many duties and hardly any
rights, at any rate as regards the higher castes. Their subjection
goes so far that the prince or Punggawa can take out of their
houses whatever he likes ; when the prince goes from one place
to another, the victuals, fowls, ducks, geese, &c.,are usually taken
by the Parakans (followers of the prince) from the houses of the
Sudras in the dessas through which the route lies, and the persons
thus robbed may not even complain. The prince or Punggawa
can even take away the wives of a Sudra, but religious feeling is
opposed to this, and still more to the murder of a Sudra, who has
committed no fault, by a noble. Both acts are done, however,
although they are of rare occurrence, in Badong. The wanton
young Gustis and Devas think that they prove their valour and
noble birth by the abduction of women and the murder of inno-
cent beasts of burden, such as the Sudras are. In Badong, old
Kassiman suppressed such deeds, and the fear of punishment
after death also has a deterrent effect.* Nevertheless the position
of the Sudras is most miserable, and only rendered supportable
by their courage and industry, and by the belief that they are born
to it. An exception to the rest of the Sudras is formed by the
Parakans (the followers of the princes, c.) ; these lead as idle a
life as the princes and Punggawas to whom they belong, and
plunder the rest of the people. These and the nobles are the
chief cock-fighters and opium-smokers, for the inhabitants of the
dessas take little part in these dissipations. Another exception to
the lot of the ordinary Sudras is formed by the Mandesas, PraU-
kels, and others, who occupy official posts.
Mandesas are the dessa-cbiefs ; they have been degraded to
Sudras by the Deva Agung, having been Wesyas by birth. Under
them are the Kabayan, Nguhukin, and Talikup, Sudras by birth,
who carry out the orders of the Mandesas.
Gaduh are the Mandurs in the dessas ; they are Sudras by birth.
Dangka and Batu-Aji are under these.
Pasek are also Wesyas who have been degraded to Sudras,
and still retain a certain superiority of rank above the rest of the
populace.
We find all these names in the "Usana Bali " (see p. 262), where
their special religion is mentioned (the " Usana Bali," as we have
seen, is only of importance to the Sudras and their chiefs of the
same caste).
* Balinese superstition regards the fate of the first Dtt'a Mangis, the founder
of Gyanyar, and that of the wife of the last Deva Agung, the much-feared
princess of Karang-Assem, as examples of such punishments ; the former was
changed into a serpent (nt1^a) t and the latter into a frog (dongk&ng). Both
had murdered many victims.
THE ISLAND OS BALL 187
The names are not heard in ordinary life, with the exception of
Mandesa. For all the rest the collective name Prab kd (the
collective ^B'kels) is used. These have a certain number of
common Stidras under them at the prince's disposal, to serve in
war or on public works, and also to exercise handicrafts (as
smiths, &c.). They are responsible for the presence and the
work of their subordinates, just as the Mandesa is responsible for
his dessa. PraVkel or Pairfkel, also M'kel, has become a title
for every superior among the Sudras the owner is the M'kel of
the slave, the husband the M'kel of the wife. This exalted rank
is also accorded to Europeans, but they stand below the three
principal castes ; ,they have not been born twice (dvija).
The Balinese Sudras are partly of Javan and partly of Balinese
origin. The former celebrate the new year (Sugian) six days
before the Balinese new year (Galungan). The latter, however,
is taken as the beginning of the calendar by the whole people.
In addition to the above, degraded Brahmans, Kshatriyas, and
Wesyas also belong to the Sudras. The original Wesyas, those
who came from Majapahit, have all become Sudras, and this at
the same time that the descendants of Arya Damar were degraded
to Wesyas. The reason given for this does not agree with that
of the degradation of the Kshatriyas. All castes, indeed, are said
to have descended a step, excepting the original Brahmans. The
descendants of Batu Henggong, the Deva Agungs, who, on account
of their procreation by the Padanda Dang hyang Kapakisan, were
also regarded as Brahmans, sank back, through the curse or the
anger of the Padanda who produced the Naga from the well (see
as to the Nagabandha, under Cremations], to true Kshatriyas, and
the consequence of this was the degradation of the other Ksha-
triyas to Wesyas, and of the Wesyas to Sudras. This explanation
is open to great doubt. Most of the descendants t of the Aryas,
the brothers of Arya Damar, have been degraded to Sudras. They
were degraded, as we have seen, to Wesyas, but when they had
also lost their governorships and their authority, oppressed by the
descendants ot Patih Gaja Maja (the royal race of Mengui-Karang-
Assem), they gradually sank to the rank of Sudras. Many of
them are still called Gustis by the people, but they have to give
their services as vassals to the princes in whose territory they live,
and no longer enjoy any honour. The conquered are always
despised and degraded in Bali. Brahmans are also degraded to
Sudras if, on account of frequent intermixture with Sudra blood,
they are no longer considered capable of retaining their dignity.
An instance of a Padanda being thus degraded is actually men-
tioned. Brahmans do, indeed, become Sudras, but the people
hold the caste in such great respect that they are still regarded
and honoured as Brahmans. Brahmana chute are mentioned in
the "Usana Java;" the Balinese explain these to be lying and
1 88 AN A CCO UNT OF
thieving Brahmans, t who, on account of their conduct, have
sunk to the rank of Sudras. Chute is, indeed, in Sanskrit, chyuta,
" fallen."
Sangguhu is a subdivision of the Sudras, who are acquainted
with the Vedas, and perform the ceremony of domestic worship,
as well as the priests. If they have, indeed, the Vedas (they may
be only Mantras, formulas), it would seem that they were originally
Brahmans. Now, we find in the " Usana Java " a statement that the
Sangguhus are, in fact, descended from Brahmana-Brahmani, and
thus are pure Brahmans, and that they were degraded on account
of the worship of the Dalem mur, that is, the god of death (perhaps
as indicating an exclusive Kdla-worship, which no longer exists
in Bali). We have therefore had in Bali departures from the
worship of the gods, not to say sects. The present Brahmans,
who have suppressed the sect referred to, now tell us, to conceal
the existence of any other worship but their own, that the Sang-
guhus are descended from a parakan (follower) of a learned
Padanda, who was hidden under the Bale whenever the Padanda
performed his domestic worship, and so came to understand the
Vedas. When he was discovered, the Padanda set him at liberty,
that he might perform the Veda-worship for himself and his
descendants. From him are descended the Sangguh or Sang-
guliu, which word is explained by means of kira : having the appear-
ance of being Brahmans. There is some number of them in Bali.
Byagaha are also mentioned in the " Usana Java " as a distinct
class, and these appear to be of the same rank as the Sangguhu.
The name, however, is almost unknown, even to the Panditas.
Thus in the three classes of Sudras last mentioned, we find also
no mixture of caste ; but they all three point to the former exist-
ence of sects, or at least to some departures from the orthodox
religion.
It would be a fortunate thing for the Sudras to be freed from
the oppression of the princes and Punggawas. The number of
noble idlers increases ^very day, and presses more and more
heavily upon the poor Sudra populace. One means of escaping
from the oppression of a prince is to flee to another State, but
even this avails them little, for the prince of the State in which
the fugitive seeks refuge immediately imposes upon him the very
burdens from which he has fled. Besides this, a fugitive who
arrives without his family is sold by the prince, in order to make
some profit out of him, and to prevent his escaping again to
another country. The best asylum is generally afforded by the
Padandas ; the prince has no right to fugitives who seek refuge
with them, and there are several kind-hearted really pious men
among these Padandas who consider it shameful to sell a person
seeking an asylum, or even to require much work from him.
Many Padandas, however, are also genuine natives, and avail
THE ISLAND OF BAH. 189
themselves of these cases to enrich themselves. t The chief curb
upon the despotism of the noble castes over the Sudras is, in fact,
the fear lest they should run away, and the power and revenues
of the nobles should thereby be diminished. The Balinese,
however, are unusually attached to their country and their dessa ;
ordinary extortion does not drive them to flight ; it is generally
only some glaring cruelty or the fear of being sold out of Bali that
induces them to escape. Still, the fact that there are always many
fugitives in Bali shows the injustice and cruelty of most of the
princes, and the degeneration of the system of caste. In Badong
there are very many fugitives from Gyanyar, Mengui, Klonkong,
and Karang-Assem ; on the other hand, scarcely any fugitives
from Badong are to be found in the other states. The reason of
this is, that the government in this state is comparatively mild and
humane, and this is attributable to Kassiman's higher intelligence
and his contact with Europeans. For the sake of humanity it is
desirable that intercourse between Europeans / and the Balinese
should increase, and that the position of the Sudras should thus
be improved, at any rate a little, if they could not be entirely freed
from their oppressors. The character of the Balinese as a people,
irrespectively of their castes, has been very accurately perceived
by Raffles : they are a manly nation, both in body and mind, far
superior to the Javanese (although outwardly less civilized), and
endowed with many virtues which in Java have disappeared
beneath the mire of immorality and fickleness. The Balinese
(excepting those who carry on trade and are corrupted by undue
gains) are faithful and honest and can work hard ; it is only for
their prince that they work reluctantly, for they receive from him
neither payment nor food ; and, finally, they are braver than all
other natives, but against the European they are as incapable ot
resistance as all the descendants of the black Adam, and the
defiant language employed to Europeans comes from a timid
heart. The European is as a tiger to all natives, and they know
him by instinct.
CASTE IN JAVA.
The existence of caste in Java has hitherto been denied, but
what we now know of Bali appears to have placed it beyond
doubt. That all the institutions of the Balinese are of Javan
origin is affirmed by the manuscripts and the oral tradition of the
people. The ** Usana Java" enumerates Kshatriyas of Koripan
(where Panji lived), of GaglangQi Singhasari, of Kediri or Daha,
and of Janggala. In all the great kingdoms of Java, therefore,
Kshatriyas existed. The descent of the Kshatriyas of Daha is
traced in the Brahmandapurana to the ^Muni Pulaha. In the
same work the Brahmans are divided into &iva-Brahmans, Buddha-
190 AN A CCO UNT OF
Brahuians, and Bujangga-Brahmans (thus the Brahmans in Java
were exactly like those now in Bali) ; these are sons of Brahmana
Haji, that is to say, the various forms of worship are derived from
Haji Saka, the founder of the Indo- Javanese calendar. Bujangga
seems in ancient times to have been a distinct sect. We have
spoken of the present meaning of this word ; it seems formerly to
have meant a particular kind of worship, in contradistinction to
Siva and Buddha.* The Wesyas are also mentioned in the
" Usana Java 51 as existing in Java ; to these belong the Patihs,
D'mangs, and Tumenggungs. The names Wesya and Gusti (see
"Tijdsch. v. N. I." 7, 2, 185) still exist in Java, although their
meaning has become obscure. The Sudras are expressly men-
tioned, and their name also is still heard in Java.
In the " Usana Java " (which really only relates to Java) the
Bralinians spring from Brahma's Sivadara, the opening in the
head which in children does not close for some years, the
Kshatriyas from his breast, the Wesyas from his abdomen, and the
Sudras from his knee.
The nobles of the courts of Daha and Majapahit, according to
express testimony, were Kshatriyas and Wesyas. To the Kshatriyas
belong all those who bear the title of Arya, K'bo, or Mahisa and
Rangga. The titles of the Wesyas have just been given.
The appearance of caste and the existence of the Veclas in
Bali afford the strongest evidence for the existence of both in
ancient Java, because all that we find in Bali was derived from
Java alone, not a single fact discovered in Bali proving, or even
indicating, direct communication between this island and India,
and because the Siva Brahmans, who, after having established
themselves for a short time at Majapahit, crossed over to Bali,
came, according to the Balinese, not direct from India, but from
the interior of Java (Kediri). The real origin of all the Balinese
institutions is also shown by the fact that we find nothing to
remind us directly of India, and no ancient writing, Sanskrit,
Kavi, or Javanese : we may well ask how could emigrants from
India have so neglected and forgotten their own writing and
language, and have merely preserved the modern Javanese writing
and the Kavi tongue ?
* The word Bujangga means a serpent ; in India the serpent-worship has
been adopted in the Brahmanical doctrine ; we find it especially in Cas'mira ;
serpent-worship is still found in Java and Bali (Vasuki), and the Bujanggas
appear to have originally been serpent-worshippers, who afterwards joined the
Sivaitic sect.
THE ISLAND OF BALL 191
THE CALCULATION OF TIME IN BALI.
The calculation of time is of two kinds : the Indian and the
Balinese. All that we know of it is contained in a work called
Wriga Garga* This work is composed of Indian and Polynesian
elements, and, like the rest of the literature, must be of Javanese
origin, as in that island alone do we find an almost identical
calendar.
We have an Indian division of time, according to the lunar
months. These months, however, are transformed by interpola-
tion, into the solar year, and^ the solar years are calculated from
the time of Salivahana or Saka^ as is also the practice in the
greater part of Hindustan and of the Dekkan. This calendar
begins on the i4th of March, 78 B.C. This calculation, however,
is but little used by the Balinese, and their scanty astronomical
knowledge renders it extremely difficult, although, for the sake of
agriculture and a few feasts, it must be preserved. The inter-
polation is irregular ; in India two months are introduced in five
years, but in Bali, on the one hand, the month of Kdrtika may
comprise two months, and, on the other, the month of Asada
may last till the constellation of the Pleiades (Krittika) is visible
at sunset. The only constellations used by the Balinese in such
calculations are the Pleiades and Orion. The latter is called
Waluku, the plough, and also by the Indian name langgala (in
Malay tangala). They understand by this, however, only the
three centre stars of Orion. To supplement this defective means
of calculating the year, the Balinese observe certain natural
phenomena occurring regularly every year at the same time, such
as the blossoming of certain flowers and plants and the appearance
of wings on the white ant, and also the phenomena of the sea.
A man of rank informed me that the interpolation of a month
had only happened three times in his life. This would give an
interval of about twelve years between each interpolation ; it is
more probable that an interpolation is made every ten years,
hence the name tejtggek for such a division. At present the
Balinese lunar months are rather more than thirty days in advance
of the Indian, as fixed by Wilson. The year begins with the
month of Kasanga ; the Indian name for this is Chetra, and
commences, according to Wilson, in March. This, then, proves
that the Balinese years do indeed start from the exact date of
Saka (i4th of March), and Raffles' and Crawfurd's conjectures as
* Garga, according to Wilson, is one of the ten Munis or saints ; one Garga
is also the author of astronomical works in India ; see Bentley, " Hindu
Astronomy," p. 54 ; his book is named Sanhita ; he lived 550 years B.C. [See,
however, Kern, in the Introduction to his edition of " Varaka-Mihira's Brihat-
sanhita," p. 31 ff.] The Balinese Calendar, as it is, cannot be derived from
him, because it contains elements never known in India.
192 AN ACCOUNT OF
to the difference of the calendar in Java and Bali can be solved
by assuming that the Balinese have retained the ancient and true
calendar, whilst the Javanese, through Muhammadan influence,
have forgotten how to calculate the solar year, and have, therefore,
got seven or eight years further from Saka than they should be.
All the months but two (Jyeshta and Asada) have both Sanskrit
and Balinese names.
The Sanskrit names are :
Sravana.
2. Bhadra or Badrawada (Sanskrit
Bhadrapada).
3. As'uji or As'uje (derived from
As'vayuj).
[ 4. Karttika.
5. Margas'ira or Margas'irshya.
6. Pos'ya or Paus'a.
7. Maga.
8. P'alguna.
9. Mad'umasa or Chetra.
10. Wes'aka.
11. Jyeshta.
12. Asadha.
The Balinese names for the first ten are the ordinal numbers
from i to 10 :
1. Kasa.
2. Karo.
3. Katiga.
5. Kalima.
6. Kanam.
7. Kapita.
8. Kahulu.
9. Kasanga.
10. Kadas'a.
4. Kapat.
We have no Balinese names for Jyeshtha and Ashadha (vulgo
Sada\ and this leads us to suspect that the original Balinese-
Javanese year, like that of the ancient Romans, had only 10
months. Now we find in the purely Balinese calendar, of which
we shall speak shortly, divisions of 35 days or 5 weeks, which, it is
true, do not now possess a special name, but no doubt originally
corresponded with the 10 months ; 10 times 35 gives about the
duration of the y^ear of lunar months, 354 days. The addition of
Jyeshtha and Ashadha, then, occurred at a later period, when
Indian influence had led to the division of the year into 12
months, each of 29 or 30 days.
In the year 1847 the first day of the month Kasanga or Chetta
was on the i6th of February ; in 1848 on the 5th of February ; in
1849 it will fall on the 24th of January. In 1844, on the con-
trary, it was on the 24th of March, almost on the day on which,
according to Wilson, the Indian month -Chetra. should begin.
Although in Bali, from ignorance of the calculations, the method
of fixing the solar year is obscure, and is known but imperfectly
to a few learned priests (who themselves cannot properly explain
the reasons for it), yet we learn from the tables called Pengalihan
lindan (the searching of the moon) how the time of the Balinese
calculation is brought into accordance with that of the lunar
months : 64 lunar months, of which 30 are of 29 and 34 of 30
days, give us 1890 days, which agree with 9 Balinese years of
210 days.
According to the Indian calculation, there should be one or
two intercalary months next year (1849), because then the
THE ISLAND OF BALI.
193
difference between the solar and the lunar year will be from the
24th of January to the 2oth of March, and will thus amount to
55 days; nothing, however, has yet been said to me of any such
intention. In India, as we have already pointed out, there are 2
intercalary months every 5 years called Malimlucha ; this name is
unknown in Bali, nor have I heard any name which could cor-
respond to it.
The six Indian seasons do not exist in Bali. The lunar months
are chiefly important for fixing certain feasts. For all other dates
use is made solely of
THE BALINESE CALENDAR.
This is formed by a combination of the Polynesian week of
5 days (pahing, puan, wage, kaliwon, mams) with the Indian week
of 7 days (Rediti* Soma, Anggara, Budd'a, Vrihaspti, Sukra,
Saneschara) ; this combination gives us 35 days, which form the
basis of the Balinese calendar ; we write them thus :
Budda kaliwon,
Wrihaspati manis,
Sukra pahing,
Sanes'chara puan,
Rediti wage,
Soma kaliwon,
Anggara manis,
Budda pahing,
Wrihaspati puan,
Sukra wage,
Sanes'chara kaliwon,
Rediti manis,
Soma pahing,
Anggara puan,
Budda wage,
Wrihaspati Kaliwon,
Sukra manis,
Sanes'chara pahing,
Rediti puan,
Soma wage,
Anggara kaliwon,
Budda manis,
Wrihaspati pahing,
Sukra puan,
Sanes'chara wage,
Rediti kaliwon,
Soma manis,
Anggara pahing,
Budda puan,
Wrihaspati wage,
Sukra kaliwon,
Sanes'chara manis,
Rediti pahing,
Soma puan,
Anggara wage.
We find these combinations of 35 days six times in each
Balinese year of 210 days, but they are not called months. On
the other hand, each of the 30 weeks has its own name. We find
these names in Raffles, vol. i. p. 476. The order of sequence is
the same in Bali, but some names must be corrected. Each
division begins with Budda kaliwon, the day on which Galungan,
the Balinese new year, falls, and not on Rediti (or Diti) Pahing,
as Raffles asserts. The week in which the year begins is Dun-
ghulan ; Sinta, however, is always named as the first week in Bali,
as in Java ; the names are :
* The name Rediti can only be explained by Aditya, "sun." The substi-
tution of ri for d may well have arisen by the uncommon initial a having been
taken for the sign r2, which is very well known. Vrihaspati is found in good
MSS. for Respati, which is the ordinary pronunciation. [Van der Tuuk,
"Notes on Kawi," p. 9. On the Batah Calendar, which in many points
agrees with the Balinese and Javanese arrangements, see J. B. Neumann, in
" Tijdschrift van het Nederlandsch Aardrijkskundig Genootschap," II., vol. iii.
p. 528 ff.]
SECOND SERIES. VOL. II. O
AN ACCOUNT OF
1. Sinta
2. Landep
3. Wukir
4. Kurantil
5. Tolu
6. Gumreg
7. Wariga
8. Warigadian, or Wariganing
Wariga
9. Julung Wangi
10. Julung Sungsang
11. Dunghulan (in Raffles Ga-
lungan)
12. Kuningan
13. Langkir
14. Madang Siha
15. Julung Pujut
1 6. Pahang
17. Kurw'lut
1 8. Marakih
19. Tambir
20. Madangkungang
21. Mahatal
22. Huje
23. Menahil
24. Prang Bakat
25. Bala Muki
26. Hugu
27. Wayang
28. Kulawu
29. Dukut
30. Watu gunung.
Raffles, vol. i. p. 376 sqq., gives the fable respecting the origin
of these names of the weeks. I have not yet heard anything of it
in Bali. In Bali this division, like the rest of the calendar, is said
to be derived from India. Its introduction was no doubt made by
the priests at an early period, in order to add to the sanctity of the
religious feasts and institutions, which are regulated according to
this calendar.*
Of the 35 days the first, eleventh, and twenty-first (Budda kali-
won, Saneschara kaliwon and Anggara kaliwoii) are sacred. Here
again, therefore, the decimal system prevails. In the first five
weeks, the feast of Galungan, the new year, falls on Budda kali-
won, the feast of Kuningan (in Crawfurd Galunan and Kuninan)
on Saneschara kaliwon, and finally the feast in the principal
temple of Uluwatu in Badong on Anggara kaliwon.
Some Balinese only call this year of 210 days a half-year. It is
indeed, however, their civil year, and they are only disposed to
regard it as a half-year to make it agree with the Muhammadan,
Chinese, and European year.
Each of these 35 days has a constellation, which indicates its
good or evil qualities, and is of special importance for nativities ;
these constellations are not the same, however, as our own, or
even as the Indian, for, with the exception of the Waluku (Orion)
and the Krittika (the Pleiades), the signs are somewhat arbitrary.
ASTROLOGICAL CALENDAR.
These constellations, which decide the good or evil fortune of
the day, are called lintangan (lintang = bintang, "star"). They
are supposed to pass into the body of people born on that day.
Rediti manis has the Ancha-ancha, a human figure standing on
* Sapta Rest (the seven ancient saints, sages) are said to have prepared it.
The pe ngalihan bulan, also, in which the lunar calendar is brought into agree-
ment with that of thirty weeks, is attributed to the same persons.
THE ISLAND OF BALI. 195
its head and hands. People bom on that day have weak bodies
and weak voices, but are impertinent to their parents.
Rediti patting, the Go/a, elephant. This signifies sorrow caused
by other people.
Rediti puan. Patrem, the creese. This signifies a predis-
position to suicide.
Rediti wage Waluku, the plough (Orion). Signifying hap-
piness late in life.
Rediti kaliwon. Gowang, a body without a head. A strong
propensity for stealing and robbery.
Soma manis. N*yu, the gooseberry-bush. Happiness and
riches, which, however, will not last long.
Soma pahing. Dupa, incense. Early death of the married man
or woman.
Soma puan. Ulanjar, a divorced woman. One who is prone
to take upon himself the faults of others, and gets into dilemmas.
Soma wage. Lembu, a white bull. Happiness and riches.
Soma kaliwon. Padati Sunya, the empty cart, deserted by its
driver. Easily robbed.
Anggara manis. Kuda, the horse. Disposed to bad tricks.
Anggara pahing. -Juju, the crab. Good sense and speedy
acquirement of wealth.
Anggara puan. Asu ajak, the wild dog. Excellence and
boldness in war, esteem among the great.
Anggara wage. Jong sarat, the overladen boat. Probable
misfortune at sea.
Anggara kaliwon. Cheleng, the hog. Success in breeding
swine.
Bud da manis. Ngerang-erang. weeping, lamenting. Much
sorrow all one's life.
Budda pahing. Gajamina, half elephant, half fish. A good
omen.
Budda puan. Lumbung, the rice-barn. Wealth of goods and
money quickly obtained ; fortunate in travel.
Budda wage. Krittika (the Pleiades). In marrying, will
obtain many slaves. Spends much and makes many presents, is
beloved by the princes.
Budda kaliwon. Titiwa, the carrying away of dead bodies, or
the depositing of the dead in the place of burial. Loss of
children.
Wrihaspati manis. Sangal tikel, the broken axe. All under-
takings unsuccessful.
Wrihaspati pahing. Salah ukur, discontent and constant strife
with men.
Wrihaspati puan. Bade, the bier (on which corpses are burnt).
Many long illnesses.
Wrihaspati wage. Rumba, the vessel with holy water.
O 2
196 AN ACCOUNT OF
Wrihaspati kaliwon. Naga, the serpent. A very bad dis-
position.
Sukra manis. Banyak angrim, the brooding goose. Loss of
property by theft.
Sukra pahing. Bubu bosor, the open bow-net. Quick change,
coming and going, rich and poor.
Sukra puan. Prawu p'gat, the broken boat. The husband
will be deserted by his wife, the wife by her husband.
Sukra wage. Mengrabut untang or glutan, the shifting of one's
faults on to another's shoulders.
Sukra kaliwon. Udang or Makara, the sea-lobster. Poverty
in youth.
Saneschara manis. D'pat, a head. (This and the Gowang, see
Rediti kaliwon, remind us of the head and the body of the demon
Rahu, which pursue the moon.) Many members of the family
will die.
Saneschara pahing. Ru, the dart. Boldness and skill in war,
esteemed by princes (Sanskrit, ru, "war, battle").
Saneschara puan. Sengenge ( Jav. Srengenge), the sun. Free-
dom from sickness.
Saneschara wage. Puhuh tarung, a species of quail (Dutch
uecht-kw artel, the " fighting-quail "). Much fighting.
Saneschara kaliwon. -Jampana, the bier. Much misfortune.
Repeated illness.
There is not much astronomical science in all this, and many
of the ideas are really ridiculous, yet the Balinese attach value to
them, and those who are born on a day with a good constellation
often boast of it.
In addition to these astrological meanings of the 35 days, there
are for each of the seven days of the week : (i) a god, who pre-
sides over it ; (2) a human figure, indicating the character of the
person; (3) a tree ; (4) a bird; (5) a buta (demon); (6) a satwa
(beast). The nature of these is supposed to indicate what the
character of the person born on that day will be.
The seven gods are given, following the order of the days of the
week, beginning with Sunday, as follows : Indra, Uma, Brahma,
Vishnu, Guru, Sri, Varna; or, according to another account, Indra,
Pritiwi, Wishmt, Brahma, Guru, Uma, Durga. The seven Butas
are : Hulu asu (dog's head), Hulu Kbo (buffalo's head), Hulu kuda
(horse's head), Hulu lembu (cow's head), Hulu singha (lion's head),
Hulu gaja (elephant's head), Hulu gagak (crow's head). From
these the man obtains his passions, and from the beasts his lower
qualities.
According to the astrological notions of the Balinese, the day is
divided into five parts, each of which has a separate name,
although they occur in different order on different days. Raffles
also mentions a division of the day into five parts in Java ; but in
THE ISLAND OF BALL
197
Bali we have different names and rules for this division. The
principal part is called Mrita (Amrita) ; he who is born at this
time of day is certain to have good fortune, The five parts of the
days undergo twelve changes, and to know their order on a
particular day, the number of the day of the Indian week is added
to that of the day of the Polynesian week, and the result is one of
the twelve combinations of the five parts of the day (and also of
the night). The rest of the divisions (besides Mrita) are sunya
(empty, poor), kala (passionate ; after the god Kala), pati (must
die), linyok (will become bad and thievish). The day is further
divided, in a civil aspect, into eight hours, dadauhan, calculated
from sunrise to sunset (their names are dauh pisan, stroke one,
dauh ro, dauh tiga or telu t &c.). The night is also divided into
eight parts in the same manner. To find the hour a sort of water-
clock (clepsydra) is used, consisting of a clapper with a little hole
in its bottom which rests on the water. As soon as the clapper is
filled it is emptied by an attendant, and the number of strokes are
given at the same time upon a drum. Contrivances of this kind
are to be found in the principal palaces e.g., in Den Passar (in
Badong), Mengui, &c.
The lunar month is divided into the white and the black half,
sukla-paksha and krishya-paksha (literally, the white and black
wing), as is the case in India. The days of the white half are
called tanggal) and are reckoned from new to full moon ; those of
the black half are called panluang. To describe a date, the
Balinese give the day of the week (of the Indian week of seven
days as well as of the Polynesian of five), the name of the week
(according to the Balinese division of 30 weeks), the name and
the half of the lunar month (white or^lack), the day of that half,
and finally the year, calculated from Saka. Instead of the year of
Saka, they also give simply the year of the century, the century
itself being understood. The century is divided into 10 divisions,
each of 10 years ; each such division is called tenggek, each single
year rah. Thus we have, for instance, on the 26th of June, 1847,
Saneschara (Saturday) kaliwon (according to the Polynesian week)
ivara Landap (in the week of Landap) masa kasa (in the month
of Kasa) sukla paksha (in the white half; also simply tanggal, with
a waxing moon) ping 13 (on the thirteenth day) rah 9, tenggek 6
(thus 69). Adding to this the century (1700), we get the year of
Saka 1769, to which 78 must be added to arrive at the year of our
Lord (1847).
Besides the Wriga garga, which, as we have said, is entirely of
Indian origin, there are two other works upon the calendar, which,
however, are no longer used ; their titles are Sundari* trus and
* The name Sundari is Sanskrit, and means " the fair one." Trus and
bungka are Polynesian additions ; the former seems to indicate that it contains
a continuous calendar.
198 AN A CCO UNT OP
Sundari bungka. As I have not yet been able to get a sight of
them, I cannot say much about them. The Pandita in Taman
said that these works are more recent than the Wriga Garga, and
were composed in Java, whereas the Wriga Garga was composed
in India (Kling). They were used formerly, but the Balinese
priests have given the preference to the older Wriga Garga. This
also is an instance of their adherence to all that is ancient and, in
their opinion, of Indian origin. The Wriga Garga appears to date
from the time of the kingdom of Daha (Kediri), whence our
priests are descended, and which kingdom, as we have seen, they
often confound with India and call Baratawarsa.
We have enumerated in the "Tijdsch. van N. I.," Jaargang
VIII. iv. 211, still further divisions of time, namely, a dvivara,
trivara, chaturvara, shadvara, astavara, sangavara, and dasavara,
besides the saptavara %x\& panchavara ; all these are less prominent
in ordinary life. The trivara serves to indicate the market days,
as the Panchavara does in Java ; there is a market every third day
(Dvara).
The shadvara is often found in dates added to the saptavara
and panchavara. The names of the shadvara are Polynesian :
Tunggleh, Haryang, Wurukung, Paniron, Was, Mahnhi. Thus
the first day of the week Sinta is Tungghh Faking Rediti, the
second Haryang Puan Soma, the third Wurukung Wage anggara,
the fourth Paniron Kaliwon Budda, the fifth Was Mam's Wrehas-
pati, the sixth Mahulu Pahing Sukra, the seventh Tungglch Puan
Saneschara. The remaining divisions are chiefly used for astro-
logical definitions. I have not yet discovered clearly how the
astavara are brought into accordance with the tutelar deities attri-
buted to each of the seven days of the week (see above). The
names of the gods of the Astavara are : Sri, Indra, Guru, Yama,
Rudra, Brahma, Kala, Uma ; the tutelar gods of the seven week-
days are : Indra, Uma, Brahma, Wishnu, Guru, Sri, Yama, or,
according to the other account, Indra, Pritiwi, Wishnu, Brahma,.
Guru, Uma, Durga.
The zodiac (rasi] is also used for astrological purposes. We
have given the names in the " Tijds. v. N. I.," Jaargang VIII.,
iv. 211. In good MSS. we find them written still better according
to the Sanskrit. The signs of the zodiac, which I found drawn in
a manuscript, are the Indian ones ; there is no aquarius, but only
the water-pitcher (kumba), and instead of the ram there is a shrimp
(udang = makara). Mrechika is Skrt. Vrischika, the scorpion.
Rakata should be Karkata. What is most remarkable is, how-
ever, the absence of the tula (scales) in the manuscript referred to.
This could not be an accidental omission, for the claws of the
scorpion stretch over the place where the tul& ought to have been,,
and the scorpion thus took up the room of two signs. Now it is
THE ISLAND OF BALI. 199
well known that the Greeks in ancient times had only eleven signs
of the zodiac, and that it was precisely the scales which were
absent, and also that it is supposed that the scales originated out
of the claws of the scorpion ; although it is not certain whether a
sign was lost in the course of transmission from the East (or from
Egypt), or whether there were only eleven originally. The dis-
covery of a zodiac with eleven signs in India now renders this
question still more intricate. This zodiac cannot have come from
the West, for the entirely Indian character of our zodiac excludes
this theory ; we must therefore suppose that in India also the
zodiac once had but eleven signs, and that it was not till later
times that the scales were added both in India and in Greece. In
Bali all twelve signs are given by name, and thus here also a
twelfth has been added ; but the fact that we found in Bali the
drawing referred to proves that at the time of the first intercourse
between India and Java the zodiac still had but eleven signs in the
former country. It is not known whether a similar zodiac has
been found in India, but it would be of the highest importance to
obtain further information from there on this point, and if possible
to determine the age of such representations, which would furnish
a date of the utmost importance both for the history of astronomy
and for that of civilization in Java.
The eclipses of the sun and moon are explained in Bali, as in
India, as the devouring of these bodies by a demon (Rahii) ; the
eclipses of the sun are called graha and those of the moon ra/iu,
which in India, however, is no distinction, as the former means the
act of devouring and the latter the devouring demon. To help the
moon on these occasions, the Balinese make a terrible noise with
their rice-blocks and other instruments, as they do on the eve of
Galungan and of the fast-days (nyept), when the evil spirits are
driven away by noise. The Panditas know that Europeans predict
the eclipses of the sun and moon, and questioned me about it ;
they themselves, however, are ignorant of the method of calcula-
tion. They also keep this knowledge secret from the people, as
the following instance will show. A European, Mr. M., lived some
time in Tabanan, and was very intimate with the young prince.
On the approach of an eclipse of the moon, he predicted its occur-
rence to the prince, who was very pleased thereat, but was com-
pelled by his priests to banish the European from his country, for
since the European knew more than the priests, it necessarily fol-
lowed that he was possessed of an evil spirit from which he obtained
his knowledge. Had the European first told it to the priests, they
would not have driven him from the country, but would have dis-
played their knowledge to the people.
As I do not possess a MS. of the Wriga Garga, I cannot give
here any further information as to the astronomical and astrological
200 NOTICES ON ZOOLOGICAL SUBJECTS.
science of the Balinese, although it is most worthy of attention.
The greater portion of the people are still utterly ignorant of the
calendar. Even among the priests this science is only retained
mechanically ; but what knowledge they have they use especially
to make the people dependent upon them. Every important
undertaking requires the help of the priest as the mediator with
the deity, and as astronomer and astrologer.
XI.
NOTICES ON ZOOLOGICAL SUBJECTS.
By Messrs. DIARD and DUVAUCEL,
Naturalists employed under the authority of the Lieut. Governor of
Bencoolen. 1
[Translated from "Malayan Miscellanies" (Bencoolen, 1821), vol. i. No. 9.]
NOTICE ON THE PANGOLING ANT-EATER OF THE
EAST INDIES.
THE forms and proportions of certain animals appear to har-
monize so little with their habits, that one may, so to speak,
be pardoned for considering them as the accidental sports of organ-
ization, escaped from the hands of Nature and betraying, by the
heteroclitic contrasts of their structure and their course of life, the
imperfection of their original outline.
When one sees, for example, such large and strong quadrupeds
as the tamandua, sharing a mode of living with the birds which
appears unworthy of the most miserable quadruped, can we help
being shocked at the part which has been theirs, and looking upon
an animal more than four feet in length, feeding solely upon ants,
otherwise than as a monster in creation, or at least, as a creature
ridiculously degraded by the insectivorous regimen to which it has
been subjected ?
In vain would one search in the necessity with which, for example,
the tamandua in burrowing and digging out the earth, often at
great depths, to get at its prey, for a sufficient reason for excusing
1 [Concerning these gentlemen, see the "Memoir of the Life and Public
Services of Sir T. S. Raffles" (London, 1830), pp. 372 and 703 ff. ; and more
especially vol. xiii. of the " Transactions of the Linnean Society," pp. 239, 240.]
NOTICES ON ZOOLOGICAL SUBJECTS. 201
its large body; one would be forced to put aside this explanation,
since other very small species of the same class execute absolutely
the same manoeuvres, and do not in the least appear to need an
enormous force to declare war against a colony of ants. One
must therefore return to the belief that Nature has not always
realized, even in the most perfect relationships, that this animal
must make an exception to the harmonic rules which we admire
in the greatest part of its productions.
When we say that Nature in creating certain animals did not
quite conceive of them in the highest sense, we are told that
absolute analogy was not reckoned among their organic and
instinctive faculties. Certainly, if it pleased Nature sometimes to
omit placing things in harmony in some of her works, it can never
have been at the expense of the creatures she has created. Whether
we recognize or not some anomalous quadruped, with a mouth, with-
out the sign of teeth, and only provided with a tongue like a bird's,
protractile and almost filiform, it nevertheless still belongs to the
class of the most wonderful animals, and to those most worthy of
the researches and investigations of naturalists.
There are five or six species of ant-eaters to be found in America,
one of which, as we have said before, has an enormous body, con-
sidering its insectivorous habits; but in the Eastern hemisphere
only two species have been found as yet, one in Africa and the
other in Asia : anatomically speaking, these species are exactly the
same as those in America, but differing completely in the scale-
like armour which covers the whole of their body. These animals,
which have been transported into Europe several times, have been
seen and described by naturalists who call them generically
Manis, and who distinguish them by the special names of
brachyura and macrura. The macrura is the phatagin of Button ,
the brachyura is his Pangoling, which he had all reason to
call it, as the word Pangoling really means in Malay an animal
that rolls itself up, and is one of the true names of the Manis in
this country. To complete this descriptive word, the Malays
generally say Pangoling sisik, the scaled animal that rolls itself
up ; nevertheless, it is well to observe, that along the west coast
of Sumatra, the Manis is not called Pangoling, but Pangilling; 1
the Malays want to express by this the faculty which this animal
has for climbing ; this is what we cannot decide.
Although the Pangolings are not very rare animals, nevertheless,
as they are extremely valuable to most of the Indians, an account
of the wonderful medicinal properties attributed to their scales and
to their nails, it is very difficult to procure them, and as yet we
have only one in our collection.
The specimen we found at Pulo Pinang seems to be full-grown ;
its total length is three feet and a half, and its tail is only twenty
1 [Properly panggoling, panggiling.]
202 NOTICES ON ZOOLOGICAL SUBJECTS.
inches long, while its body is very long, and its legs are very short ;
its little head, which is of the same narrowness as its neck, has the
appearance of being sharpened into a pointed cone ; its tail, on
the contrary, is very thick and strong, rounded above and flat
underneath, sharp at the sides, and diminishing in size from the
top to its point.
This animal has a very pointed snout, little eyes, a very small
mouth cleft underneath, and its round ears resemble great pads
more than real conchas. With the exception of the upper part of
the nose, the side and inferior parts of the head, underneath the
neck, and the stomach, and the inner side of the front legs, the
Pangoling is entirely covered, even to the roots of its nails, with
scales strong and sharp, and of different shapes and sizes accord-
ing to the parts they are destined to protect. For instance, those
at the top of the head, and those which cover the limbs, get
smaller as they come to the nose and fingers ; it is the same with
those of the tail, they are very narrow at the point and very large
at the top. The largest scales are those covering the flanks and
rump, they are more than an inch in diameter. Their surfaces
are all marked with wrinkles and divergent flutings, and most of
them have their ends divided into three lappets. Those which
entirely cover the hind legs, and those which fringe the sides of
the tail, are alone cut in the shape of a pointed angle. These
angular scales are also divided into different classes. The first
kind have a rather sharp keel, the second kind are folded into
grooves angularly, according to their length, so as to be able to fit
in above and below. The sharp sides of the tail have a still more
wonderful shape ; from underneath each scale several single long
hairs are seen to appear, while there is not a single one to be seen
on the parts we have indicated as being naked.
The Pangolings are not such slow animals as you might expect
at first ; they run fast enough, and climb quite easily, by means of
their strong and sharp nails, and by helping themselves with their
tail, using it as a hand or as a buttress. These animals always go
along ferreting about and looking under dead leaves and under old
trunks of trees for larvae and insects, which is their staple food.
But the most remarkable faculty the Pangoling possesses is that
of rolling itself up into a ball, when it is threatened with danger.
To do this it stops all of a sudden, doubles itself up, puts its head
between its front legs and finishes by covering itself with a com-
plete scaly armour by putting its long and broad tail over its feet
and above its head.
These animals have five fingers on each foot, armed with strong
and pointed nails ; but of these five fingers, pulled out to their
utmost length, only the three middle ones are strong and long
enough to be of service to the Pangoling in climbing and in dig-
ging, the outer ones are too short to be of any use to them.
NOTICES ON ZOOLOGICAL SUBJECT*. 203
NOTICE ON TWO SHREW-MICE OF INDIA.
These little quadrupeds have been distinguished under the
name of field-mice, shrew-mice, Sorex musaraneus, because,
owing to their common form, they are so very closely related
with rats, strictly speaking Mits, but which in other ways differ
completely in their organic characters and even in certain exterior
characters, remarkable enough, even if they only existed for the
purpose of preventing them from being confounded with the rats.
The shrew-mice, essentially destined for an insectivorous life, have
partly an organization quite fitted to this kind of life : their teeth
have been armed with sharp points, and their jaws furnished on
all their circumference with a scarcely interrupted series of canine
sharp incisors. None the less are their exterior characters to be
compared to their nocturnal and subterraneous habits. Their
snout, which is formed into a movable and pointed trunk, serves
them as an instrument for digging into the earth to look for food;
and their eyes, reduced to two bright specks scarcely visible,
suffice to guide them in their excavations underground, and at
the same time by their smallness protecting them against things
which would have got into them had they been larger ; but what
serves to distinguish the shrew-mice more than anything else, is
the strong musky smell which generally all their species give out ;
it is a strong smell which comes from a sort of pomade secreted
by a particular small matter of granular follicles, which is found
on all the true shrew-mice on the lower parts of their body.
These follicles, covered with little strong short hairs convergent
together, are dotted about on a longitudinal line along nearly the
whole of the length of the flanks of all the well-known shrew-mice.
On the contrary, in the other two specimens which we have to
notice as we live in India, these little glands are concentrated into
one, and are round and can be seen on both sides of their body
a little behind the shoulder ; it is principally to note this slight
irregularity in the form of an organ which seems to have some
internal connection with the organization of these little animals,
that we will give here the description of the two varieties
which form part of our collection. Although they are one
of the most remarkable species of their kind, because of their
body, which is sometimes as long as six inches and more, with-
out counting the tail, which is more than three inches long, and
although it is extremely common in Bengal and in the Indian
Archipelago, where it is vulgarly known under the name of musky
rat, it has nevertheless as yet only been imperfectly described.
The only writer who mentioned it particularly is Buffon, who
indicates it, more than describes it, as a low animal brought from
Pondicherry by Mr. Sonnerat : but as his description is incomplete
and as he omitted, according to his habit, to give this species a
204 NOTICES ON ZOOLOGICAL SUBJECTS.
Latin name, the result was that it has been confounded with some
other kind by the nomenclators ; at any rate it has been impossible
for us to recognize them in any of their catalogues, if we venture
to give it the name of Sorex Indicus, without fear of giving the
word a double meaning.
This shrew-mouse, as we have said before, is a little more than
six inches long, its tail being about half the size, while its size
and form are like the common rat ; its fur is of a pretty light grey
underneath, and slightly browny above ; it may yet be distinguished
by the pale pink tint of its naked parts and feet, its tail, the sides
of its mouth and ears.
Its quadrangular tail is covered with tiny scaly compartments,
which are very fine and furnished with spare hairs, some short and
some long. Its snout, straight and flexible like a little trunk,
divided at its extremity into two little tubercles, reaches more
than half an inch further than the inferior snout.
Lastly, its eyes are extremely small, and its ears round and
short, bare, and pressed against the head, are joined in the inside
by two large valves, which can almost completely shut the auditory
meatus.
The generic characters in this shrew-mouse, as we have already
said, consist in a tiny moschiverous gland situated behind each
shoulder, and in a digestive organ, just like that of other shrew-
mice : thus they have six incisors above and four below, the ends
of which are very long ; it has one canine and four molar teeth in
the upper jaw, and only one canine and three molar teeth in the
lower one.
With the shrew-mouse as well as with several kinds of rodents
and marsupialia the vulva and the anus open into one common
duct.
Our second species of shrew-mouse only differs by its body
from the other species we have just described ; three inches and
a half from the extremity of its snout to the beginning of its tail,
is its greatest length, and otherwise it is exactly the same animal,
with the short and quadrangular tail, the long and flexible snout,
bare ears with large interior valves, tiny little eyes, the feet with
five toes, and the grey colour a little darker on its back ; the teeth,
the intestinal duct, and the moschiverous gland are exactly the
same in every respect to the preceding species, and the vulva and
anus open into a common drain too.
NOTICE ON THE VlVERRA MUNGOS.
Owing to a kind of inborn pride which most of us possess, of
only thinking of ourselves as the object of all the wonders of
creation, and above all if they can come in for our own use, the
happy instinct which renders certain animals natural enemies of
NOTICES ON ZOOLOGICAL SUBJECTS. 205
some hurtful species, has always been considered as a means
employed by Nature for protecting mankind from too great a
multiplication of some destructive beings who have escaped from
its bountiful keeping. This idea, above all with those people
naturally inclined to mysticism, ought particularly to lead to
exaggerate the instinctive inclinations of these little quadrupeds
so celebrated under the name of ichneumon for their combats
with the most dangerous reptiles. That is why the Egyptians
(who worship these animals) believe that they are incessantly
hunting for crocodile's eggs, which they break just for the pleasure
of doing an abhorred creature an injury ; and not able to attack
these great amphibians openly, they try to attack them when
asleep with wide-open mouth, then they throw themselves with
fury at their throats and kill them by eating out their tongue and
their guts : but one fact which has contributed more than any-
thing to increase the marvellous in the history of the ichneumons
is that of their deadly combat with the Cobra di Capello. It is
scarcely to be believed that such feeble quadrupeds could stand
up before such mighty adversaries, without supposing that Nature
taught them the means of neutralizing the terrible effects of the
most terrible of poisons.
To verify this phenomenon, nothing better can be found than to
imagine that the ichneumons knew a certain Dyctom the power
of which rendered the bite of the worst vipers harmless. If only
this precious plant could be discovered, it would be pointed out
as a sure specific against every kind of animal, and the botanists
would consecrate its astounding properties by giving it the name
of Ophioriza Mtmgos. A circumstance easy enough to explain,
very probably, gave rise to this fable.
When the Cobra di Capello, attacked by the ichneumon, keep
up the defence too long and vigorously, it happens sometimes
that these last named, worn out with fatigue, retire from the battle-
field a minute to take breath, and then return to the combat
with renewed ardour. Thus this momentary retreat can easily be
interpreted. But how can we be content with an explanation
natural enough in a case as extraordinary as this? The most
marvellous ought to seem the most likely. Besides, people were
convinced that the ichneumons only disappeared for a few minutes
to be able to look for a remedy for their wounds. As to what is
true in all these stories, manufactured by report, is that they
possess most wonderful courage and voracity, and that they are
very fond of all kinds of eggs, and that they seem to prefer
reptiles to any other prey ; that they attack everything they have
strength enough to put to death ; and that if they are not even
afraid of fighting the most venomous, it is because their poison
has no great effect upon them, which would not be anything so
extraordinary, as we know of several poisons which have no effect
206 NOTICES ON ZOOLOGICAL SUBJECTS.
upon certain animals, or, rather, because they are agile enough to
avoid the deadly fangs of their enemy; and we must own that
this seems the most likely. If the ichneumon did not fear the
poison of the cobras, why would it seize them in a way to avoid
their poisoned fangs ? and why, from fear of a single bite, should
it prolong the attack for hours, often uselessly an attack which
would be extremely hurtful to it if it were not kept back from
ending it by the fear of a single bite ?
We have ourselves no confirmed observation on this subject ;
but the assertions of several trustworthy persons, who have assured
us of their having seen ichneumons killed by the cobras, seems to
us to tend greatly to the support of this last-named probability.
The ichneumons, as well as the other Viverras, are to be found
in all warm climates of the old continent. As their species are
not numerous (there are not more than four kinds), and as their
instinct has always attracted the attention of all travellers, it
follows that there are descriptions exact enough ; nevertheless,
we do not think it useless to give the ichneumon of India
(Viverra Mungos of Linneus) a place here. This species, which
is the most celebrated for its combats with the Cobra di Capello,
is not as large as the ichneumon of Egypt. Its body, from the
tip of the nose to the beginning of the tail, is never longer than
fifteen inches, and its slender tail, which ends in a point, is from
nine to ten inches long. Its head is of quite a different shape
than the other Viverras. The bones of the nose are very much
arched above the nostrils, the forehead seems to end in a snout
rather higher than large, and the round ears are pressed against
the head, and are very far back ; this gives its head a still more
particular aspect ; in short, this ichneumon is a very long animal
and low on its legs. Its body, covered with rather long and
rough fur, and shaded from four to five times with dark-brown
and white, has the appearance of being uniformly quilled with
the same colour. On the snout and at the extremities of the
four paws, this fur is very scarce; everywhere else it is lined
underneath with a woolly and russet-coloured down like the
otters.
This Viverra, like the others, has five toes to each foot, but
the nails are all longer, sharper, and hardly retractile; like the
others it has also six little incisors, three false molars and two
long canine ones in each jaw, one carnivorous one and three
tuberculous above, and below four false molars, one tuberculous
one, and a carnivorous one; then the anus of the Viverra
Mungos opens at the end of a large bag in which certain glands
exude a particular humour. Although these little animals are
instinctively very cruel, and are not satisfied except with slaughter,
they can be tamed nevertheless easily enough, and be allowed to
run about freely in the house. But domestication does not in
NOTICES ON ZOOLOGICAL SUBJECTS. 207
any way affect their sanguinary instinct, for, however well fed they
may be, they kill anything that happens to resist them just as
cruelly, and are always giving chase to the rats, fowls, birds, and
snakes.
NOTICE ON THE PORCUPINES OF INDIA.
If there are animals who ought to be positively well known, it
would be the porcupines, the numerous specimens of which are
very common, and they are otherwise very curious for their sharp
quills which clothe their body instead of fur ; nevertheless, when
we come to consult the descriptions the nomenclators have given
of them, we find them so very vague and incomplete, that we
cannot be positively sure of either their number, identity, or
of the different varieties indicated. So, to keep to the specimens
of the old continent, the common porcupines of the southern
parts of Europe, Africa and India, which are certainly different,
at any rate those of the last-named country, we count as one
animal, while certain authors have divided them into two distinct
species from the only porcupine with the long penicillegerous
tail of the Eastern continent and isles ; let us show them that we
know more. We have said that the common porcupine of India
was different to that of Italy ; we will add further, that we believe
there are two particular varieties in Asia quite similar in form, it
is true, to the Hystrix cristata, but which differs in the way in
which the black and white bands run on their quills.
During our stay in Bengal we were able to procure several
specimens of this species of porcupine, which is commonly found in
this country ; but being ourselves led away by the general opinion,
we did not give it enough attention to notice that it differed from
the European species, and it was only after our first journey to
Pulo Pinang that we verified this difference, having accidentally
compared the quills of the short-tailed porcupine which you find
in this island as well as in Sumatra, and probably throughout the
Sunda archipelago, with some of those we brought from Calcutta,
and we were struck with the different placing of the colours on
each kind, and saw then that they belonged to a different species of
animal, not only they, but neither of these species were, as people
have believed till now, a long-haired porcupine ; in fact, this latter
kind has all its quills white at the extremities.
The quills of the Bengal porcupine, white at first, have a little
above their roots a little black ring, then a white one. and then
they are entirely black to their extremities, while those of the
Sunda islands are quite white, with the exception of a single black
ring situated a little above the middle part ; this is the only real
difference we can indicate as yet between these three species ;
but it is probable that comparing each one would lead us to find
out other kinds.
2o8 NOTICES ON ZOOLOGICAL SUBJECTS.
The total length of the largest we have seen does not exceed
twenty-eight inches : its head is nearly five inches, and its tail
hardly three inches long ; when it bristles up and lifts up its back,
it stands about from fourteen to fifteen inches high ; otherwise
this species has just the same shape and proportions as the
common European porcupine. Its ears are round and bare ; its
snout thick, shaped like a trunk ; it has five toes on the hind and
four on the fore-legs ; one nailed tubercle serves as a thumb ; its
long quills only cover the posterior half of the body ; the chest,
the upper and lower parts of the shoulders, are covered with
another kind of quill, much shorter, flat like the blade of a sword,
and quite black, with the exception of some which are white at
the ends, and forming under the throat a kind of half-collar of
this same colour. Those on the nape of the neck are white at the
tips, but they are differently shaped ; they are thick round hairs,
a little longer than the rest, nevertheless without being sufficient
to form a mane ; the thick stiff hairs which cover the feet and legs
are more sturdy, shorter, and of a very deep black : the tail is
short and straightens itself when the animal stands up. At the
beginning of the tail are quills like those on the back, and those
at the tip resemble very delicate and elastic hairs, entirely white,
swollen at the ends into round tubes and ending naturally in sharp
points. Although they are mostly open and hollow, when the
porcupine bristles itself up these tubes rub together and produce
a kind of audible dull noise : not counting the different quills we
have spoken of, the porcupines have still here and there, and
chiefly on the lower parts of their body, other very long and
slender quills, with a single black ring on them, like the others.
To terminate the history of this porcupine we must add that at
Bencoolen it is called Landah, and that we have provisionally
specified it under the name of Hystrix torquatus ; lastly, we must
mention that, in Bengal, the one we described at the commence-
ment of this article is called Lazaroo.
We must now speak of the long brush-tailed porcupine, and to
prove (what is in fact the general opinion) that the Hystrix
macrura is nothing else than the fasciculata of Linneus. In fact,
the only particular character which Seba and Shreber gave to their
macrura, was the twofold enlargement of the thongs of the elastic
hair at the end of the tail, which several descriptions of the
porcupine of Queda mention ; but this character is of no value
whatever ; for on the same animal these enlarged hairs are always
to be found. Besides this the long-tailed porcupine is well enough
known ; it has even been seen in Europe several times, neverthe-
less we believe it to be still possible to add some interesting facts
to his description.
It seems that this species is never longer than from seventeen
to eighteen inches, without counting the tail, which is from eight
DESCRIPTIONS OF MALA VAN PLANTS. 209
to nine inches long ; it stands lower on its legs, its body and its
head are longer, and its proportions are generally far less massive
than those of the porquatus ; its quills, the longest of which is
never more than an inch and a half, are of the same shape as those
covering the shoulders of the last named, that is to say, they are
flat like the blade of a sword and very much fluted at the top ;
those on the back are of a greyish colour, with a large brown
speckle in the middle ; those on the lower parts of the body,
on the contrary, are only white at the roots, and are quite black
otherwise : imperceptibly those on the legs take the form of coarse
round hairs ; besides the Malacca porcupine has, like the others, on
the lower parts of the body, several long black hairs, dotted about
here and there among its ordinary quills : their tail is covered at
the top with needles like those on the back, and is not, as some
say, scaly nearly everywhere ; on the contrary, it is well covered
with very fine and delicate spikes which appear to be the remains
of the old ones, which surround the ends in the form of a tuft :
otherwise these are, as we have already said, either singly or doubly
in thongs, and these thongs are naturally pointed at the tip, and
are never truncated, except by accident. This singular species
always carries its tail lifted up like a trumpet, and makes the tuft
at the end tremble, like the others.
We have counted on each jaw, like on the preceding ones,
except the two great ordinary incisors, four cylinder- like teeth
concentrically striated at their crown, and it has, like the others,
five toes on the hind and four on the fore-feet with a tubercle
taking the place of a thumb.
XII.
DESCRIPTIONS OF MALAYAN PLANTS.
By WILLIAM JACK.
["Malayan Miscellanies" (Bencoolen, 1820-22), vol. i. Nos. I and 5 ;
vol. ii. No. 7.]
ZINGIBER GRACILE. W. J.
Monandria Monogynia.
N. O. Scitamimce.
Foliis glabris, scapis erectis, spicis cylindricis gracilibus coloratis,
bracteis ovatis acutis, corollae labio trilobo, lobo medio bifido.
Native of Pulo Pinang.
Stem erect, somewhat recurved, round and smooth. Leaves
alternate, subsessile on their sheaths, broad lanceolate, 6 or 7
SECOND SERIES. VOL. II. P
2io DESCRIPTIONS OF MALA VAN PLANTS.
inches long, acuminate, very entire, very smooth, shining above.
Sheaths smooth, with a long scariose ligula, often lacerated on the
edge. Scapes erect, a foot high, invested with alternate sheaths.
Spikes cylindrical, oblong, imbricated with bright red ovate acute
bracts, shorter than the flowers. An inner bract or involucre
surrounds the base of each flower. Calyx shorter by one-half
than the corolla, membranaceous, curved, cleft on one side.
Corolla yellowish white ; exterior limb 3-parted, longer than the
inner one ; lacinise acuminate, the upper one longer and incum-
bent ; interior limb unilabiate, lip 3-lobed, middle lobe bifid, with
reflexed margins. Anther terminating in an incurved horn. Ova-
rium 3-celled, many-seeded. Style filiform, longer than the horn,
of the anther, embraced at the base by two linear corpuscules.
AMOMUM BIFLORUM. W. J.
Monandria Monogynia.
N. O. Scita??iinea.
Foliis lato lanceolatis glaberrimis, caule ancipite, spicis bifloris..
Native of Pulo Pinang.
A slender delicate species. Stem erect, somewhat recurved,
3 feet in height, compressed, double edged. Leaves alternate,
bifarious, short petioled upon their sheaths, broad lanceolate,
acuminate, narrow at the base, entire, very smooth, the middle
nerve somewhat pubescent. Sheaths striated, slightly tomentose,
with a short, round ciliate ligula. The base of the leaf-bearing
stem is swelled into a tuber, which throws out horizontal shoots
of some feet in length, of the thickness of a quill, and invested
with membranous sheaths. These shoots, which run under-
ground, send up from their joints a number of biflorous peduncles,
or scapes, which are enveloped in bracteal sheaths. Flowers,
generally two, the one appearing after the other. At the base of
each flower is a single lanceolate, acute reddish bract ; besides
this there is a tubular bract, or involucrum, surrounding the base
of the germen, membranaceous, half as long as the calyx, and
deeply cleft on one side. Calyx superior, tubular, 2 or 3 cleft.
Corolla white, tubular; upper part of the tube villous within;
exterior limb membranaceous, 3-parted, segments nearly equal ;
interior limb unilabiate, lip broader above, rounded, thickened,
and yellow in the middle. Filament of the stamen broad, incum-
bent. Anther short, thick, 2-lobed, crowned with an erect 3-lobed
crest. Style filiform ; stigma infundibuliform. Nectaries two,
linear, at the base of the style. Ovarium 3-celled, many -seeded.
DESCRIPTIONS OF MALA VAN PLANTS. 2 1 1
PSYCHOTRIA MALAYANA.-W. J.
Pentandria Monogynia.
N. O. Rubiacea.
Foliis lato lanceolatis, stipulis indivisis, paniculis terminalibus
corymbosis, corollae fauce villosa.
Byumbada. Malay.
Native of Pulo Pinang.
A shrub with round smooth branches. Leaves petiolate, opposite,
broad lanceolate, 10 inches in length, acuminate, decurrent upon
the petiole, entire, very smooth. Petioles short, thick, round,
surrounded at the base by a prominent ring, from which a thick
rib diverges on each side and unites with a similar one from the
base of the opposite leaf to form the nerve of the large inter-
petiolar ovate acute stipule. Panicles corymbose, terminal.
Flowers numerous. Bracts broad, membranaceous, embracing.
Calyx superior, erect, quinquefid. Corolla white, with greenish
limb, infundibuliform, longer than the calyx, mouth closed with
dense white hairs, limb 5-parted, somewhat reflexed, laciniae ovate.
Stamina 5, erect, inserted on the tube, filaments very short, anthers
linear. Style filiform, stigmata two, thick and linear. Capsule
inferior, 2-celled, 2-seeded.
RONDELETIA CORYMBOSA. W. J.
Pentandria Monogynia.
Tetrandra, pedunculis plerumque terminalibus dichotome corym-
bosis, floribus unilateralibus, foliis obovato lanceolatis.
Native of Pulo Pinang.
Stem erect, shrubby, from 4 to 6 feet in height, with somewhat
compressed villous branches. Leaves opposite, petiolate, obovate
lanceolate, acute, attenuated to the petiole, entire, punctate above
with callous dots, villous below. Petioles short, thickened at the
base. Stipules interpetiolar, long, erect, tongue-shaped, obtuse,
villous, with a thick middle rib formed by the union of one from
each axil. Peduncles terminal, and from the upper axils, sup-
porting dichotomous corymbs, composed of unilateral spikes.
Flowers erect, sessile, disposed alternately in a double series.
Calyx superior, 4-cleft, with short acute laciniae. Corolla white,
tinged with red, funnel-shaped, much longer than the calyx, faux
naked, limb erect, 4-parted, laciniae sub-rotund. Stamina 4,
inserted into the faux, filaments very short, anthers linear. Style
filiform, exsert. Stigma bifid. Capsule crowned with the calyx,
2-celled, many-seeded, with central placentae.
P 2
2 1 2 DESCRIPTIONS OF MALA VAN PLANTS.
PHYTEUMA BEGONIFOLIUM.
Pentandria Monogynia.
N. O. Campanulacecz.
Foliis semicordatis inequilateralibus serratis, spiels unilateralibus
axillaribus, revolutis.
Phyteuma begonifolia. Roxb. " Hort. Beng." p. 85.
Pulo Pinang.
A small herbaceous plant. Stem procumbent, i or 2 feet in
length, thick, villous chiefly at the summit, with fasciculate hairs.
Leaves alternate, petiolate, semicordate, inequilateral, turning to
one side, 8 inches long, acute, with gross subspinescent serratures,
villous beneath, adult leaves smooth above, nerves generally
dichotomous. Petioles thick, round, furrowed above. Stipules
none. Peduncles axillary or supra-axillary. Flowers unilateral,
erect, arranged in two rows on a recurved spike, nearly sessile,
crowded. Bracts cuneiform, obtuse. Calyx semi-superior, ovate,
villous, 5-lobed, lobes obtuse. Corolla white, campanulate, per-
sistent, limb recurved, 5-lobed, lobes obtuse ; after florescence
the corolla becomes green and enlarges. Stamina 5, erect, short,
inserted on the calyx and opposite to its divisions. Anthers
linear, acute. Ovarium surrounded by the calyx and connected
with it by five longitudinal septa or processes, from which the
stamina spring, 3 or 4-celled, many-seeded, placentae from the
inner angles of the cells. Style short, thick. Stigma large, thick,
3-lobed. Capsule 3 or 4-celled, containing numerous seeds
arranged on convex placentae.
The septa which unite the calyx and ovary appear continuous
with the filaments of the stamina. The young parts of the plant
are densely villous, but the hairs are easily rubbed away. In
drying, the plant assumes a bright yellow colour. It appears
extremely doubtful whether this plant be truly referrible to Phy-
teuma ; it does not, however, agree well with any other genus of
the family of Campanulaceae, and it will deserve consideration
whether it ought not to constitute a new genus in that order.
CURCULIGO SUMATRANA ROXB.
Hexandria Monogynia.
Foliis lato-lanceolatis plicatis glabris, spicis densis brevibus,
tubo perianthii bacca longiore.
Involucrum. Rumph. " Amb." vi. p. 114, t. 53.
Kalapa puyu. Malay.
Sumatra and Pulo Pinang.
Root composed of fibres proceeding from a tuber. Leaves
radical, petiolate, ovate-lanceolate, acuminate, attenuated to the
DESCRIPTIONS OF MALA VAN PLANTS. 2 1 3
base, plicato-nervose, very entire, smooth. Petioles erect, chan-
nelled above, keeled beneath, sheathing at the base. Spikes
radical from among the sheaths of the petioles, erect, dense,
much shorter than the petioles. Flowers erect, sessile, adpressed
to the rachis, each furnished with an ovate acuminate membra-
naceous spathe. Calyx none. Corolla yellow, superior, limb
spreading 6-parted, laciniae lanceolate acute, tube impervious,
being a thick, solid column on the summit of the germen.
Stamina 6, erect, opposite to the laciniae of the corolla. Anthers
linear. Style short. Ovarium 3-celled, many-seeded. Capsule
baccate, ovate, 3-sided, containing from 8 to 10 ovate black seeds
which are imbedded in pulp.
I found at Singapore another species, agreeing in most respects
with this, but having hirsute leaves.
LORANTHUS COCCINEUS. W. J.
Floribus spicatis tetrandris, spicis axillaribus erectis foliis sub-
ovatis glabris.
Found at Singapore.
Branches long, vimineous. Leaves alternate petiolate, oblong-
ovate, subcordate at the base, attenuated towards the apex, which
is obtuse, entire, smooth. Petioles short. Spikes axillary, solitary,
or in pairs, erect, longer than the leaves ; flowers sessile, closely
pressed to the rachis before expansion. A single small, ovate
ferruginous bract is situated at the base of each flower. Calyx
superior, nearly entire, scarcely toothed. Corolla coccineous,
4-petaled, erect, tubular, limb spreading, petals nearly linear,
broader at the base. Stamina 4, red, erect, inserted into the
middle of the petals and equalling them in length; anthers
oblong adnate, red. Style red, erect, scarcely longer than the
stamina. Stigma obtusely capitate. Berry ovate, elongated
above, i -seeded. Seed contained in a hard shell, 4-sided, its
apex immersed in gluten, into which the radicle shoots. Embryo
inverse, the radicle produced beyond the albumen.
This species is nearly allied to the L. pentapetala of Roxburgh,
agreeing with it in habit and inflorescence.
LORANTHUS FERRUGINEUS. ROXB.
Ferrugineo villosa, foliis ellipticis obtusis supra glabris, pedun-
culis fasciculatis axillaribus 2-6 floris, floribus tetrandris extus
ferrugineo villosis.
Roxb. " Hort. Beng." p. 87.
Sumatra, &c.
2 1 4 DESCRIPTIONS OF MALA VAN PLANTS.
A parasitic shrub which attaches itself firmly to the branches of
trees by means of long runners and numerous circular bands. The
branches are long and hanging, and when young densely covered
with reddish ferruginous wool. Leaves opposite, short petioled,
coriaceous, elliptic, obtuse, entire, smooth and green above, ferru-
ginous and densely villous beneath. Stipules none. Peduncles
fascicled, from i to 4 in each axil, 2-6 flowered. A small scale-
like bract embraces the base of the ovary. Calyx (if any) an
entire margin crowning the ovarium. Corolla covered externally,
as well as the peduncles and ovary, with ferruginous tomentum,
green and smooth within, tubular, divisible into four petals, which
commonly adhere at their base but separate at the limb, which is
generally more deeply cloven on one side. Stamina 4, inserted
into the tube, and nearly as long as the limb. Filaments flat,
deep purple. Style as long as the corolla. Stigma sub-rotund.
Berry ovate, ferruginous, i -seeded.
NEPHELIUM LAPPACEUM.
Marsd. " Hist. Sumatra," pi. iv.
Rambutan. Malay.
Frequent throughout the Malay countries and islands.
A tree. Leaves alternate, pinnate, leaflets generally from 5 to
7, ovate, acute at both ends, very entire, smooth. Panicles
terminal, erect. Flowers numerous, small, white, male and her-
maphrodite. Calyx from 4 to 6-parted, spreading. Corolla none.
Stamina from 5 to 8, spreading, longer than the calyx, inserted
into a disc below the germen. Anthers sub-rotund. Ovarium
2-seeded, abortive in the male flowers. Style i. Stigmata 2,
revolute. Fruit geminate, one commonly abortive, the rudiment
of which remains at the base of the perfect one, which is sub-
rotund, covered with a coriaceous rind and echinate with long
soft spines, i -seeded, the seed covered with a white acid pulp.
The fruit is much esteemed, and has an agreeable subacid
flavour. The parts of the flower vary much in number; six is
perhaps the most frequent number of the stamina. There is but
one style, not two as commonly described. The affinities of this
tree seem to have been little understood. It belongs without
doubt to the family of the Sapindi, and is closely related to
Scytalia, as justly conjectured by the author of the botanical
articles in Rees' " Cyclopaedia."
SAPINDUS RUBIGINOSUS. ROXB.
Octandria Monogynia.
Arborescens inermis, paniculis terminalibus, calicibus 5 phyllis,
corollis 4-petalis, baccis tribus connatis oblongis.
Kulit layu. Malay.
DESCRIPTIONS OF MALA VAN PLANTS. 2 1 5
Pulo Pinang.
Arborescent. Leaves alternate, abruptly pinnate leaflets nearly
opposite, subsessile, ovate-lanceolate, obtuse, with a small mucro
or point, very entire, nearly smooth, with a few scattered hairs,
chiefly on the under surface. Petioles tomentose. Panicles ter-
minal, erect, composed of numerous simple racemes. Pedicles
short, generally in pairs. Bracts subulate. Calyx 5-leaved,
leaflets sub-rotund, concave, the two outer ones smaller. Corolla
white, 4-petalled, somewhat longer than the calyx, petals ovate,
obtuse, appendiculate at the base, appendices furnished with two
transverse lines of white hairs. Stamina 8, of which the 5 upper
and longer are incumbent over the remaining 3. Filaments
villous. Anthers oblong, yellow. Style i, short, persistent.
Stigma capitate, 4-sided, villous. Germina 3, i-seeded. Berries
3, connate at the base, purple, oblong, i -seeded.
MELIA EXCELSA. W. J.
Decandria Monogynia.
Foliis pinnatis, foliolis integerrimis, paniculis coarctatis axil-
laribus foliis paullo longioribus.
Pulo Pinang.
A lofty tree, with straight trunk and light grey bark. Branches
rough with the vestiges of the fallen leaves, foliose at their
summits. Leaves crowded, disposed in a spiral manner, pinnate
with an odd one which is often wanting, leaflets sub-opposite,
oblong-lanceolate, inequilateral, obtusely acuminate, very entire,
smooth, shining above. Petioles round, smooth, thickened, and
somewhat scaly at the base. Panicles axillary, ascending, rather
longer than the leaves, not diffuse. Flowers pedicellate, pedicles
bracteolate. Calyx very small, 5-parted. Corolla white, 5-petalled,
spreading, petals linear. Staminiferous tube erect, gibbous at the
base, 10 dentate, 10 furrowed, as if consisting of 10 united fila-
ments. Anthers 10, oblong, yellow, within the mouth of the tube.
Style as long as the tube. Stigma capitate.
MICROCOS TOMENTOSA. SMITH, in Rees' " Cyclopaedia."
Polyandria Monogynia.
N. O. Ttliacea.
Foliis trinerviis subtus villosis.
Grewia Paniculata. Roxb. " Hort. Beng." p. 93.
Native of Pulo Pinang.
A moderate-sized tree with rough bark, the branchlets villous
and ferruginous. Leaves alternate, short petioled, elliptic oblong,
broader above, with a short acumen, 3-nerved, dentate, serrate
2 1 6 DESCRIPTIONS OF MALA VAN PLANTS.
towards the apex, scarcely pilose above, densely villous beneath,
the hairs divaricate and often stellate. Stipules linear, generally
bifid. Panicles terminal. Flowers for the most part in threes,
involucred with deciduous trifid and linear bracts. Calyx 5-leaved,
spreading, leaflets oblong concave. Corolla yellow, less than the
calyx, petals ovate, unguiculate and without nectaries. Stamina
numerous, inserted below the germen. Germen stipitate. Drupe
containing a nut marked externally with five lines, 3-celled, 3-
seeded.
This agrees perfectly with the excellent description given by
Sir J. E. Smith in Rees' " Cyclopaedia," from a specimen preserved
in the herbarium of the younger Linnaeus, unaccompanied with
any notice concerning its native country, and also deficient iu
fruit. Its affinity to the original species of Microcos is fully
proved on actual examination of the fruit, and this exact agree-
ment affords a further confirmation of the propriety of separating
Microcos from Grewia. The terminal inflorescence and involucral
bractese form a peculiar and distinctive character. In this species
the flowers are generally three together, and are surrounded by
three trifid bractese, within which are found three other smaller
and linear ones.
MICROCOS GLABRA. W. J.
Foliis trinerviis serratis glabris.
Found on the Island of Carnicobar.
It nearly resembles the M. Tomentosa, differing chiefly in
having smooth leaves. In inflorescence and fruit it is entirely
similar. The young branches are tomentose. There are fre-
quently flowers in the uppermost axils.
MIMOSA JIRINGA.
Arbor inermis, foliis conjugate pinnatis, foliolis 3-jugis glaber-
rimis, paniculis fasciculatis axillaribus, capitulis paucifloris legu-
minibus maximis articulato-contortis nigris.
Mimosa Djiringa. Roxb. " Hort. Beng." p. 93.
Bua Jering. Malay.
Pulo Pinang, Malacca, &c.
A lofty tree, unarmed, with grey bark and round smooth
branches. Leaves alternate, conjugato-pinnate, leaflets 3-paired,
on short thick pedicles, ovate lanceolate, obtusely acuminate, very
entire, very smooth, the upper pairs larger. Petioles round, some-
what keeled above. An indistinct gland above the base of the
common petiole. Capitula few flowered, panicled ; these panicles
are fasciculate, axillary, or in the axils of fallen leaves. Flowers
white. Calyx 5 -toothed. Corolla twice as long as the calyx,
DESCRIPTIONS OF MALA VAN PLANTS. 2 1 7
5-cleft. Stamina numerous, monadelphous, long, fertile. Style
as long as the stamina. Legumes solitary, very large, almost
black, about a foot in length, spirally contorted, articulate, 2-
valved, articulations subrotund, i -seeded, convex and prominent
on both sides. Seeds large, subrotund, double convex.
This species belongs to the genus Inga of Willdenow.
CLERODENDRUM MOLLE. W. J.
Didynamia Angiospermia.
Caule erecto tetragono, foliis cordatis acuminatis integerrimis
tomentosis, panicula terminal], tubo corollse calycae vix longiore,
calyce fructus ampliato caraoso albo.
Frequent in Sumatra, Pulo Pinang, &c.
A shrub from 3 to 6 feet in height, erect, little branched ; stem
4-sided, villous. Leaves opposite, petiolate, cordate, acuminate,
very entire, softly tomentose. Panicle terminal, oppositely tricho-
tomous, erect, with leaf-like bracts. Calyx 5-parted, tomentose,
laciniae ovate, acute, erect, with reflexed margins. Corolla tomen-
tose without, tube as long as the calyx, limb 5-parted, spreading,
secund, laciniae nearly equal, crisped at the margin. Stamina
exsert, horizontally deflexed to each side. Style erect, as long as
the stamina. Stigma bifid. Calyx of the fruit flat, enlarged,
fleshy and white. Berry from i to 4-seeded, according to the
number that abort.
This species approaches nearest to the C. infortunatum, but is
abundantly distinguished by the softness of the leaves, which are
larger and more deeply cordate, by the comparative shortness of
the tube of the corolla, and by the white calyx of the fruit.
Besides this species, I have met with another in various parts
of these islands, and particularly at Acheen, which has been
figured in Andrews' " Repository " under the name of Cleroden-
drum pyramidale. It is a large, showy plant. A still more
beautiful species, and perhaps the most elegant of the whole
genus, is the C. nutans, so named by my friend Dr. Wallich,
Superintendent of the Botanic Garden at Calcutta, who received
it from the North-eastern frontier of Bengal. I found it not
uncommon at Pulo Pinang, and this is not the only instance in
which I have had occasion to observe a coincidence between the
plants of these distant countries. This species is characterized as
follows :
C. nutans, Wall Foliis lanceolatis acuminatis glabris, pani-
culis longissimis terminalibus nutantibus, pedunculis remotis divari-
catis paucifloris.
These panicles or racemes hang gracefully from the extremity
of the branches, the flowers are white, not numerous, the peduncles,
2 1 8 DESCRIPTIONS OF MALA YAN PLANTS.
or primary divisions of the panicle being remote, opposite, divari-
cate, short, and seldom bearing more than three flowers. It is
called Unting unting by the Malays.
GMELINA VILLOSA. ROXB.
Spinosa, foliis rhomboideis subtus villosis, racemis terminalibus,
bracteis magnis acuminatis, drupis sphericis dispermis.
Radix deiparae. Rumph. "Amb." ii. p. 124, t. 39.
Kayo Briang.
Native of Sumatra, &c.
Arborescent. Leaves opposite, broad ovate, sometimes obscurely
3-lobed, rather obtuse, entire, smooth above, villous beneath, as
well as the petioles and branchlets. Racemes terminal. Bracts
large, ovate, acuminate. Calyx obliquely 4-toothed, marked
externally with 6 green scutellae or pustules. Corolla yellow,
ventricose. Anthers 2-lobed. Ovary 4-sporous. Drupe with a
2-seeded nut.
VITEX ARBOREA, ROXB. " Hort. Beng." p. 46.
Didynamia Angiospermia.
Arborea, foliis ternatis, foliolis ovato lanceolatis integerrimis
subtomentosis, paniculis terminalibus, bracteis calyce longioribus.
LSban. Malay.
Sumatra, &c.
A tree, with somewhat four-sided branches. Leaves opposite,
petiolate, ternate, sometimes quinate, leaflets ovate lanceolate,
acuminate, very entire, rigid, covered with a very short tomentum.
Petioles long, thickened at the base, pulverulent. Panicles ter-
minal ; flowers subsessile. Bracts opposite, ovate lanceolate, acute,
tomentose, longer than the calyces. Calyx 5-dentate, tomentose,
persistent. Corolla coerulescent, or nearly white, longer than the
calyx, contracted and almost closed at the mouth, limb bilabiate,
upper lip 2-lobed, lobes diverging, lower lip larger, 3-lobed, the
lateral lobes reflexed, the middle one larger, subrotund, concave,
tomentose at the base, and of a deeper blue than the rest.
Stamina 4, didynamous, ascending, longer than the corolla. Style
longer than the stamina. Stigma bifid. Berry black, juicy, con-
taining a 4-celled, 4-seeded nut.
The wood of this tree is very hard, and is employed by the
inhabitants of Sumatra in the construction of houses, also for
paddles. They consider an infusion of the bark as a useful
application in cases of ophthalmia.
DESCRIPTIONS OF MALA VAN PLANTS 2 1 9
SPHENODESME. W. J.
Didynamia Angiospermia.
Vitices Juss.
Calyx tubulosus 5-dentatus. Corolla 5-loba subirregularis.
Stamina 4-5 exserta. Ovarium 4 loculare, 4 sporum. Bacca
monosperma.
Flores fasciculati, involucrati.
SPHENODESME PENTANDRA. W. J.
Foliis oblongo ovatis glabris, involucris 5-6 phyllis, fasciculis
67 floris, floribus pentandris.
Rosccea pentandra. Roxb. " Cat. Hort. Beng." p. 46.
Native of Pulo Pinang.
A climbing shrub, with 4-sided, somewhat pilose branches.
Leaves opposite, petiolate, oblong ovate, subcordate at the base,
acuminate (sometimes with a retuse acumen), very entire, very
smooth. Petioles short, pilose. Fascicles 6 or 7 flowered,
peduncled, disposed in panicles at the extremity of the branches
and in the upper axils. Involucres consisting of 5 or 6 oblong,
obtuse, membranaceous, reticulated leaflets, which are longer
than the sessile flowers. Calyx campanulate, 5-plicate, 5-dentate.
Corolla infundibuliform, faux villous, limb 5-lobed, nearly regular.
Stamina 5, long, exsert. Style filiform, bifid. Ovary very hairy,
3 to 4-celled ; cells i-seeded.
There is always one leaflet less in the involucrum than the
number of flowers in the fascicle, the central flower having no
fulcrum. This species was sent to Dr. Roxburgh from Sylhet,
and by him called Roscoea ; that name, however, being pre-
occupied, a new one has become necessary. I have therefore
given it that of Sphenodesme (fasciculus alatus).
STERCULIA COCCINEA. ROXB.
Monadelphia Decandria.
Foliis oblongo lanceolatis obtuse acuminatis glabris, racemis
axillaribus et lateralibus nutantibus, laciniis calycinis linearibus
patentibus, folliculis coccineis.
Native of Pulo Pinang.
A large smooth shrub. Leaves at the summits of the branches,
alternate, petiolate, 8-10 inches long, oblong lanceolate, obtusely
acuminate, abrupt at the base, entire, smooth on both sides.
Petioles thickened at both ends. Racemes lateral from among
the leaves at the end of the branches, drooping ; flowers alternate,
pedicellate; pedicles articulate. Tube of the calyx somewhat
220 DESCRIPTIONS OF MALAYAN PLANTS.
ventricose, limb 5-parted, laciniae linear with revolute margins,
twice as long as the tube, spreading. Corolla none. Stamina 10,
sessile on the stipes of the germen. Ovarium stipitate on a
column of the length of the tube, subrotund, 5-lobed, crowned
with a declinate style. Stigmata 5, linear, revolute. Fruit com-
posed of five nearly equal crimson follicles, each of which contains
two or three seeds, which are enveloped in a black pulpy arillus.
Dr. Roxburgh's S. coccinea is a native of Sylhet, and is said to
have panicled flowers and 4-8 seeded follicles. My plant agrees,
however, so well in every other respect that I cannot consider it
to be really distinct, as those differences may be merely the effect
of a less favourable situation.
STERCULIA ANGUSTIFOLIA. ROXB.
Foliis lanceolatis superne latioribus acuminatis subtus villosis,
racemis extra axillaribus nutantibus, laciniis calycinis linearibus
apice connexis.
Unting Unting Besar. Malay.
Native of Pulo Pinang.
A tree. Branches covered with ferruginous wool. Leaves at
the summits of the branches, alternate, petiolate, lanceolate,
broader above, acuminate, narrowing to the base and there
rounded, entire, smooth (in adult leaves) above, covered beneath
with stellate hairs. Petioles thickened at both ends, ferruginously
villous, as well as the nerve of the leaf. Stipules linear, acute,
shorter than the petiole, deciduous. Racemes (panicles ?) near
the extremity of the branches lateral or extra axillary, branched,
lax, ferruginous. Bracts linear lanceolate acute. Calyx deeply
5-parted, tomentose, lacinise long, linear, acute, connected at
their points and gaping at the sides, greenish yellow, with a red
spot at the base. Corolla none. Stamina 10, on a curved
column. Ovarium stipitate, tomentose, 5-lobed. Style declinate.
Stigma 5-lobed.
A great proportion of the flowers are male, and I have not seen
the perfect fruit.
Dr. Roxburgh's plant was a native of Chittagong.
CALLA HUMILIS. W. J.
Monacia Monandria.
Acaulis, foliis ellipticis supra glabris, pedunculis 4-5 ex-axillis
foliorum petiolis brevioribus.
Kladi Ayer. Malay.
Pulo Pinang, &c.
A small stemless plant, growing under the shade of forests,
5 or 6 inches in height. Root a leaf-bearing tuber, which sends
DESCRIPTIONS OF MALA VAN PLANTS. 2 2 1
out numerous long villous fibres. Stem none, except the above-
mentioned tuber, which is everywhere invested by the sheaths of
the petioles. Leaves erect, petiolate, elliptic, ovate, rather obtuse,
with a subulate acumen, slightly cordate at the base, entire, with a
pellucid, crisped margin, smooth and green above, somewhat
hoary beneath, with villous papillae. Petioles shorter than the
leaves, channeled above, sheathing, and dilated into a waved
margin at the base. The bases of the sheaths are often perforated
by the fibres of the root. Peduncles 4-5 axillary, i -flowered,
snorter than the petioles, furnished with membranous sheaths at
the base. Spathes of an obscure red colour, oblong, convolute,
acuminate, as long as the spadix. Spadix cylindrical, entirely
covered with florets, male above and female below for about a
quarter of the length. Anthers numerous, subrotund, yellow,
sessile. Germina ovate. Styles very short. Stigmata obtuse,
peltate. A few anthers are intermingled with the pistilla. Capsules
membranaceous, globose, somewhat 4-lobed (2-celled?) generally
8-seeded. Seeds somewhat kidney-shaped, arranged round the
axis.
CALLA ANGUSTIFOLIA. W. J.
Acaulis, foliis lanceolatis utrinque acutis glabris, pedunculis
4-5 ex-axillis foliorum petiolis brevioribus.
Pulo Pinang.
A small plant of the same size and nearly related to the pre-
ceding. Leaves radical, petiolate, lanceolate, acute at both ends,
entire, smooth. Petioles sheathing at the base. Peduncles 4-5
axillary, i-flowered. Flowers, &c., exactly as in the preceding.
These two are so closely allied that it is doubtful whether they
might not be considered varieties.
CALLA NITIDA. -W. J.
Foliis ovato-lanceolatis acuminatis, scapis compressis foliis bre-
vioribus, baccis monospermis.
Found at Pulo Pinang.
This is a large subcaulescent species ; the leaves are from a
foot to a foot and a half in length, ovate lanceolate, acuminate,
very entire, very smooth, with numerous parallel nerves pro-
ceeding from a middle rib. Petioles sheathing nearly their whole
length. Scapes compressed, smooth, shorter than the leaves.
Spadix invested by the spathe, covered with florets, male above,
female beneath. Berries oblong, large, i -seeded.
FLACOURTIA INERMIS. ROXB.
Arborescens inermis, floribus hermaphroditis fasciculatis axil-
laribus, foliis ovatis serratis glabris.
2 2 2 DESCRIPTIONS OF MALA VAN PLANTS.
Koorkup. 1 Malay.
Sumatra and Pulo Pinang.
A tree of moderate size. Leaves alternate, short petioled, ovate,
obtusely acuminate, with large blunt serratures, very smooth, lucid,
from 6 to 8 inches in length. Peduncles fasciculate in the
axils, many flowered. Flowers hermaphrodite. Calyx 4-leaved,
spreading, somewhat tomentose, leaflets subrotund, sharpish.
Corolla none. Nectary composed of numerous small, subrotund,
orange- coloured glands, situated at the base of the calyx and
surrounding the stamina. Stamina numerous (20-30), hypogy-
nous, longer than the calyx, filaments white, anthers yellow, sub-
rotund. Ovary superior ovate, crowned with 4-5 short, thick,
diverging styles ; stigmata capitate, 2-lobed. Berry reddish purple,
with a juicy, acid flesh, in which are imbedded from 8 to 10
pyrense, according to the number of the styles.
The fruit of this, though rather too acid to be eaten in its raw
state, is much esteemed in tarts and pies.
ROTTLERA ALBA. ROXB.
Foliis rhomboideo-ovatis, subtus incanis, paniculis terminalibus
laxis, fructibus stellate pilosis spinis mollibus echinatis.
Baleh angin. Malay.
Sumatra and Pulo Pinang.
A tree of moderate size. Branches roundish, furfuraceous, with
appressed, stellate hairs. Leaves alternate, petiolate, rhomboidal
ovate, often approaching to 3-lobed, long acuminate, rounded
and bi-glandular at the base, where the petiole is inserted within
the margin, remotely denticulate towards the apex, smooth and
green above, hoary and tomentose beneath. The young leaves
have stellate, deciduous hairs on the upper surface. Petioles
long. Stipules none. Panicles terminal, or from the bifurcations
of the branches, peduncled, lax, and drooping. Flowers small,
numerous, short pedicled. Bracts small, and together with the
peduncles and calyx sprinkled with furfuraceous tomentum.
Male. Calyx 3-phyllous, leaflets ovate acute. Stamina numerous
in the centre of the flower. Anthers subrotund.
Female. Calyx 4 sometimes 5-parted, erect, lacinise acute.
Styles 3, diverging, hirsute above. Stigmata simple. Fruit tri-
coccous, beset with soft flexible spines, and covered with stellate
hairs, 3-seeded. Seeds subrotund, attached to the superior and
internal angle of the cells.
1 [ ? Rukam.]
DESCRIPTIONS OF MALA VAN PLANTS. 223
DIDYMOCARPUS. WALLICH.
Calyx 5-fidus. Corolla infundibuliformis, labio superiore brevi,
inferiore trilobo. Stamina 5 nunc 4, quorum 2 vel 4 fertilia.
Capsula siliquseformis, pseudoquadrilocularis, bivalvis; dissepi-
menti contrarii lobi valvulis paralleli iisdemque aemuli (ideoque
fructum bicapsularem mentientes), margine involute seminiferi.
Semina minuta nuda, pendula ?
Herbse villosae, resinoso-glanduliferse, aromaticae.
Genus Bignoniaceis, Brown ; admissa Incarvillea, adsociandum,
huicque proximum, Wallich.
I am indebted for the above character of this hitherto unpub-
lished genus to my esteemed friend Dr. Wallich, who has ascer-
tained five species, natives of Nepaul; the four following have
been since discovered in the Malay Islands.
DIDYMOCARPUS CRINITA. W. J.
Erecta, pilosa, foliis longis spatulatis acutis serratis subtus
rubris, pedunculis 2-5 axillaribus unifloris basi cum petiolo
coeuntibus, staminibus duobus fertilibus.
Timmu. Malay.
Native of the forests of Pulo Pinang.
Root long and tapering. Stem short, erect, thick, rough
beneath with the vestiges of fallen leaves. The whole plant is
covered with hairs. Leaves alternate, crowded, subsessile, long,
spatulate, 9 or 10 inches in length, acute, obtuse at the base,
serrated, rugose, hairy, brownish green above, purplish red
beneath, middle nerve strong and thick, forming a short petiole
at the base. Stipules none. Peduncles 2 to 5 in each axil,
i-flowered, round, 2 inches long, uniting at the base into a
short thick unilateral rachis, densely pilose, and adhering beneath
to the petiole. Bracts linear, 2, alternate on each peduncle.
Calyx 5-parted, hairy, reddish, laciniae erect, linear, acute, the
upper one smaller. Corolla white, tinged with purple externally,
much longer than the calyx, infundibuliform ; tube somewhat
gibbous at the base, incurved, expanding above, limb bilabiate,
upper lip 2-lobed, lower 3-lobed, larger, internally streaked with
yellow, all the segments roundish, obtuse, not very unequal.
Stamina inserted within the tube, 2 fertile, with the rudiments of
2 abortive ones, the former scarcely so long as the corolla,
conniving at their summits. Anthers composed of 2 divaricate
transverse lobes. Ovarium linear, surrounded at the base with a
white tubular entire nectarial ring or cup, and produced into a
tomentose style of the same length as the stamina. Stigma
2 24 DESCRIPTIONS OF MALA YAN PLANTS.
obtuse, truncate. Capsule long, linear, silique-shaped, cylindrical,
acute, somewhat tomentose, an inch long, 2-valved, 2-celled,
dissepiments contrary, with 2 lobes which are parallel to the
valves, revolute and seed-bearing at their margins, and part the
cells in such a manner as to give the appearance of a 4-celled
siliqua. Seeds numerous, naked, small, and subrotund.
Obs. The deep red colour of the lower surface of the leaves
and the crested disposition of the flowers in their axils render this
a very remarkable species. The aestivation is imbricate, the two
lateral lobes of the lower lip being the outermost. The genus is
nearly related to Incarvillea, but differs in having simple naked
seeds.
DIDYMOCARPUS REPTANS. W. J.
Prostrata, reptans, foliis petiolatis ellipticis crenulatis, pedun-
culis 1-3 axillaribus unifloris, staminibus duobus fertilibus.
Timmu kechil. Malay.
Found in the forests of Pulo Pinang with the preceding.
Stem prostrate, round, villous, striking root at every joint, often
a foot in length. Leaves lying flat, opposite petiolate, oblong-
oval or elliptic, rather obtuse, sometimes slightly cordate at the
base, slightly crenate, covered with white hairs, green above,
paler and sometimes reddish beneath. Petioles villous. Pe-
duncles 1-3 axillary, i-flowered, erect, as long as the leaves,
pilose, furnished with 2 bracts near the summit. Calyx 5-parted,
with erect acute lacinise, the uppermost smaller. Corolla white,
infundibuliform, bilabiate, similar to that of D. crinita but smaller,
as well as the whole plant. Stamina 2 fertile conniving above,
2 sterile. Anthers approximate, reniform, 2-celled. Nectary
surrounding the base of the ovarium, obsoletely 5-toothed at the
margin. Style equal to the stamina. Stigma simple. Capsule
long, straight, silique-shaped, pseudo-quadrilocular as in the
genus. Seeds numerous, naked.
DIDYMOCARPUS CORNICULATA. W. J.
Erecta, foliis alternis obovatis acuminatis serratis, floribus
diandris fasciculatis secundis super pedunculum axillarem elon-
gatum.
Found at Tapanooly in Sumatra.
The stem is nearly erect, from i to 2 feet in height, herbaceous
or somewhat shrubby, villous. Leaves alternate, petiolate,
obovate, acuminate, narrowing to the base, serrated, pilose above,
villous below. Peduncles axillary, solitary, elongated, bearing
DESCRIPTIONS OF MALA VAN PLANTS. 225
several dense fascicles of flowers all turned to one side, depressed
or bent at an angle to the peduncle, and spreading in a kind of
half circle, somewhat in the manner of the Lotus corniculatus.
Flowers many, white ; pedicles articulate below the calyx, covered
as well as the calyx with glandular hairs. Bracts linear, acute.
Calyx 5-parted, segments linear. Corolla white, much longer than
the calyx, infundibuliform, wide at the faux, limb somewhat
oblique, bilabiate, the lower lip longer, 3-lobed. Stamina 2,
connected above by their anthers, whose lobes are transverse.
Style as long as the stamina. Stigma capitate. Capsule silique-
shaped, 2-celled, cells bipartite (as if 4-locular), 2-valved, gener-
ally bursting at one side, many-seeded. Seeds naked.
The disposition of the flowers and fruit is peculiar, the capsules
spreading horizontally like radii in a sort of semicircle of which
the peduncle is the axis.
DIDYMOCARPUS FRUTESCENS. W. J.
Caule suffrutescente erecto, foliis oppositis longe petiolatis
ovato-lanceolatis acuminatis supra glabris subtus canescentibus,
floribus axillaribus fasciculatis didynamis.
Native of Pulo Pinang.
Stem generally simple, suffrutescent, densely covered with
ferruginous appressed scales or chaffy hairs. Leaves opposite,
long petioled, ovate-lanceolate, acuminate, attenuated to the base,
slightly serrated, 8 or 10 inches long, smooth above, hoary and
tomentose beneath, with appressed hairs. Petioles 3 inches long,
furrowed above, thickened at the base, villous. Stipules none.
Peduncles axillary, fascicled, 1-3 flowered, shorter than the
petioles, purplish. Bracts lanceolate acute. Calyx tomentose
with glandular hairs, tubular, 5-parted, laciniae linear, spreading
above. Corolla white, tomentose without like the calyx, much
longer than it, infundibuliform, incurved ; all the laciniae subrotund
obtuse. Stamina 4, didynamous, arcuate, approximate at their
summits, each pair connected by their anthers. The filaments of
the upper pair are thickened below their middle. Anthers white,
adnate to the filaments, consisting of two lobes nearly parallel.
Style of the length of the stamina. Stigma truncate. Capsule
long, linear, silique-shaped, 2-valved, 2-celled, cells 2 -parted by
the septiform lobes of the dissepiments, which are revolute and
seminiferous at their margins. Seeds numerous, naked.
SONERILA ERECTA. W. J.
Triandria Monogynia.
Erecta, ramosa, foliis lanceolatis serratis, racemis terminalibus
paucifloris, floribus sessilibus.
SECOND SERIES. VOL. II. Q
226 DESCRIPTIONS OF MALAYAN PLANTS.
Summow. Malay.
Native of the forests of Pulo Pinang.
Root fibrous. Stem erect, from 6 inches to a foot in height,
oppositely branched, round, tinged with red, fringed with 2
opposite longitudinal lines of hairs (like that of the Veronica
chamoedrys). Leaves opposite, petiolate, ovate-lanceolate, acute
at both ends, serrated, villous with erect hairs, 3-nerved, green
above, reddish beneath. Petioles nearly smooth. Stipules none.
Peduncles terminal, springing from the centre of a 4-leaved
verticil which terminates the branch, and of which two opposite
leaves are smaller. The spike is unilateral, about 4-flowered,
recurved, smooth, each flower sessile on the upper sides of the
clavate peduncle, which is there thickened and as it were scooped
out to receive it, and is attenuated downwards to the point of
insertion into the branch. Bracts none or very minute. Calyx
smooth, trifid, laciniae acute. Corolla of a light flesh colour,
composed of 3 lanceolate-ovate acuminate spreading petals.
Stamina 3, alternating with the petals, erect, scarcely so long as
the corolla. Anthers 2-celled, acute, cordate at the base. Style
erect, equal to the stamina. Stigma obtuse. Ovarium long,
linear, inferior. Capsule oblong, obtusely 3-angled, 3-celled,
3-valved, many-seeded, the dissepiments opposite to the valves.
Seeds attached to a central columnar 3-sided placenta.
Obs. This plant differs considerably in habit from the other
species of Sonerila in having an erect slender brachiate stem, and
small lanceolate leaves, not oblique at the base as in most of the
genus.
The uppermost leaves are quatern, forming a kind of involucre
to the slender peduncle which springs from their centre.
SONERILA MOLUCCANA. ROXB.
Subcaulescens, villosa, foliis oblique cordatis integris oppositis
altero minore, pedunculis axillaribus, racemis unilateralibus.
Roxb. "Fl. Ind." vol. i. p. 122.
Pouh. Malay.
A native of the moist shady forests of Pulo Pinang.
A small herbaceous plant whose root is fibrous, and whose stem
does not exceed a few inches in length. Every part is thickly
covered with red hair. The leaves are petiolate, opposite, one
much smaller and rounder than the other, unequally cordate,
acute, very entire, of a deep green on the upper surface, red
beneath, with quintuple nerves. Petioles round, and hairy.
Stipules none. Peduncles generally from the axils of the smaller
leaves, erect, bearing from r to 3 unilateral somewhat recurved
racemes, and furnished about the middle with 2 small opposite
bracteolar leaflets. The racemes are at first revolute, but unroll
DESCRIPTIONS OF MALAYAN PLANTS. 227
themselves as the flowers open. The flowers are unilateral
arranged in 2 rows upon short pedicles, and each supported by
a linear ciliate bract. Calyx superior, covered like the rest of the
plant with red hairs, 3-parted, laciniae lanceolate, acute. Corolla
white, composed of 3 petals inserted between the divisions of the
calyx, ovate, acute, with a few red hairs along the middle of the
under surface. Stamina 3, alternating with the petals ; filaments
linear, ascending; anthers linear, bending towards the style,
yellow, 2-celled. Style declinate in an opposite direction to the
stamina. Stamina simple. Capsule ovate, crowned by the calyx,
hairy, 3-celled, 3-valved, many-seeded, the dissepiments opposite
to the valves, the placentae peltate, pedicellate, affixed to the axis
of the capsule.
RHOPALA ATTENUATA. \V. J.
Tetrandria Monogynia.
Proteacea. Juss and Br.
Foliis alternis ovatis acuminatis, racemis axillaribus foliis
longioribus, pedicellis geminatis calycibusque glabris.
Native of Pulo Pinang.
Arborescent, with round smooth branches. Leaves alternate,
petiolate, ovate, acuminate, attenuated to the base and decurrent
on the petiole, 10 or n inches long, entire, sometimes with i or
2 toothlets near the point, very smooth. Petioles short, thickened
at the base. Stipules none. Spikes rather longer than the leaves,
axillary, cylindrical, flowers geminate, short pedicled. Perianth
4-leaved, leaflets linear, dilated and staminiferous at the summit,
revolute. Stamina 4, inserted near the apex of the perianth ;
filaments scarcely any ; anthers linear, 2-celled. Style filiform, as
long as the corolla. Stigma clavate. Ovarium i-celled, con-
taining 2 erect ovula.
RHOPALA MOLUCCANA. BR.
Foliis alternis obovatis obtusiusculis integerrimis, racemis
plerumque lateralibus, pedicellis bifidis calycibusque glabris.
Found in a garden at Pulo Pinang.
Arborescent with grey bark. Leaves alternate, petiolate, 6 or
7 inches long, obovate or cuneately ovate obtuse, very entire, very
smooth, yellowish green. Petioles an inch long, flattened above,
thickened at the base. Spikes lateral, generally below the leaves.
Flowers geminate on a bifid pedicle. Bracts very small. Perianth
4-leaved, leaflets revolute, dilated and stamen bearing at the
summit. Stamina 4 ; anthers linear, nearly sessile. Style filiform.
Stigma clavate. Ovarium i-celled, 2-sporous.
Obs. In the preceding the leaves are acuminate and the
Q 2
228 DESCRIPTIONS OF MALAYAN PLANTS.
flowers in pairs, each with its proper pedicle ; in this the leaves-
are rounded and obtuse at the apex, and the flowers are geminate
on a common pedicle.
IXORA PENDULA. W. J.
N. O. Rubiacecs.
Foliis elliptico-lanceolatis glaberrimis, corymbis longe peduncu-
latis pendulis.
Bunga yarum. Malay.
Native of Pulo Pinang, &c.
A shrub with smooth compresssd branches. Leaves opposite,
short petioled, n or 12 inches long, elliptically lanceolate, rather
obtuse, very entire, very smooth, shining above. Petioles little
more than half an inch in length. Stipules interpetiolar, broad at
the base, ending in a subulate point. Corymbs terminal, long
peduncled, hanging, trichotomous, many-flowered. Flowers red.
Bracts 2, small at the base of the calyx. Calyx small, 4-parted,
slightly tomentose. Corolla red, tube long and slender, limb'
4-parted, lobes ovate-lanceolate, rather acute. Stamina spreading.
Style filiform. Stigma clavate.
Obs. This is a beautiful species, at once distinguishable by its
long pendulous corymbs. Bunga Yarum is the generic Malay
name of the Ixorae.
EPITHINIA. W. J.
Tetrandria Monogynia.
N. O. Rubiacecz.
Calyx cylindricus superus, quadridentatus persistens. Corolla
tubulosa, limbo patente quadripartite, fauce villosa. Stamina
exserta. Stylus exsertus. Stigma bifidurn. Bacca sulcata,
dipyrena, nucibus oblongis dispermis, semine uno super alterum.
EPITHINIA MALAYANA. W. J.
Found in mangrove swamps on the island of Singapore.
A moderate-sized shrub with brown bark and smooth branches.
Leaves opposite, petiolate, obovate, obtuse, rounded at the
summit, attenuated at the base into the petiole, very entire, very
smooth, almost without veins, shining above, paler beneath.
Stipules none. Peduncles axillary, dichotomous, many-flowered,,
i-flowered in the bifurcations. Calyx cylindrical, persistent,
almost entire or obsoletely 4-dentate. Corolla white, tube longer
than the calyx, limb spreading, 4-parted, lobes ovate, rather acute,
faux closed with white hairs. Stamina 4, exsert, spreading,,
inserted alternately with the lobes of the corolla, filaments short,,
DESCRIPTIONS OF MALA VAN PLANTS. 2 29
anthers linear, acute, dark coloured. Ovary oblong, compressed,
2-celled, cells 2 -seeded, the one placed over the other. Style
exsert. Stigma birid, with thick linear lobes. Fruit inferior,
oblong, marked with 8 deep longitudinal furrows, crowned with
the calyx, containing 2 long narrow oblong nuts, each with 2
seeds, the one placed above the other. One of them sometimes
proves abortive.
Obs. I have not been able to refer this to any known tetran-
drous genus ; it seems to come nearest to Malanea of Aublet, but
differs in several essential characters. The position of the seeds
is peculiar.
MORINDA TETRANDRA. W. J.
N. O. Rubiacea.
Tetrandra, pedunculis umbellatis terminalibus, corollis quadri-
fidis intus hirsutis, foliis lanceolatis.
Pada vara. Rheed, " Mai." vii. p. 51, t. 27.
Mangkudu kSchil. Malay.
Native of the Malay Islands.
A small diffuse shrub, with long slender branches, nodose at
the bifurcations. Leaves opposite, short petioled, lanceolate,
acuminate, very entire, very smooth, the nerves reddish below,
and furnished with ciliated glands in the axils. Stipules inter-
petiolar, truncate. Peduncles from 5 to 10, umbellate, terminal.
Flowers aggregate on a common receptacle. Calyx an entire
margin crowning the ovary. Corolla infundibuliform, 4-parted,
the lacinise densely covered within with long white hairs. Stamina
4, shorter than the corolla, and alternating with its divisions ;
filaments very short ; anthers oblong. Ovary inferior, 2-celled,
4-seeded. Stigma bifid. Fruit subglobose, yellow, composed of
coadunate berries, angular by their mutual compression, crowned
with the vestige of the calyx, 4-seeded ; seeds osseous.
Obs. Rheed describes his Pada vara to be 14 feet in height ;
this is the only particular in which it differs from my plant. In
every other respect they agree exactly.
MORINDA POLYSPERMA. W. J.
Tetrandra, pedunculis axillaribus et terminalibus, corollis quad-
rifidis intus hirsutis, foliis ovatis acuminatis, baccis bilocularibus
polyspermis !
Found on the island of Singapore.
A shrub, with short subdichotomous flexuose branches. Leaves
opposite, petiolate, ovate, acuminate, obtuse at the base, very,
entire, very smooth, coriaceous, flat, about 3 inches long. Stipules
short, interpetiolar. Peduncles axillary * and terminal; axillary
2 3 o DESCRIPTIONS OF MALAYAN PLANTS.
ones opposite, terminal ones from i to 4 in a kind of umbel.
Capitula few-flowered. Calyx an entire margin. Corolla infun-
dibuliform, 4-parted, densely covered within with white hairs.
Stamina-4 shorter than the corolla ; filaments short ; anthers linear.
Style erect. Stigma bifid. Berries coadunate, 2-celled, many
seeded ! Seeds numerous, angular.
Obs. The flowers of this species are perfectly similar to those
of the preceding, but the fruit presents a singular anomaly in
being polyspermous. Both differ so much from the other species,
of Morinda that I think they might properly constitute a new and
distinct genus.
EUTHEMIS. W. J.
Pentandria Monog)<nia.
Calyx inferus 5-phyllus. Corolla 5-petala. Stamina quinque,
hypogyna, antheris oblongis acuminatis apice poro dehiscentibus.
Stylus filiformis, staminibus equalis. Bacca 5-sperma, serninibus
circa axim dispositis, oblongis, intus angulatis, arillo fibroso
inclusis, albuminosis, embryone inverso cylindrico longitudine
fere seminis, radicula superiore.
Frutices, foliis alternis pulcherrime striatis nervis parallelis,
racemis terminalibus, demum peracta floratione lateralibus et
oppositifoliis.
EUTHEMIS LEUCOCARPA. W. J.
Foliis lanceolatis pulchre spinuloso serratis, racemis basi ramo-
sis, baccis niveis globosis.
Palawan bCruk. Malay.
Native of the forests of Singapore.
A shrub of uncommon elegance and beauty, erect, 4 or 5 feet
in height ; branchlets round, smooth, sometimes slightly angled.
Leaves alternate, petiolate, lanceolate, acute, decurrent on the
petiole, spinuloso-serrate, very smooth and shining, beautifully
striated with fine parallel transverse nerves. Petioles margined,,
flat and channelled above, dilated at the base into a thick rounded
prominent rim, which half embraces the stem. Stipules lanceo-
late, acuminate, ciliate, very deciduous. Racemes erect, with i
or 2 branches near the base, at first terminal, afterwards lateral
nd oppositifolious, by the shooting up of the stem from the base
f the peduncle. Flowers predicellate, generally in pairs. Bracts
vate, acute. Calyx inferior, 5 -leaved, spreading, leaflets ovate,
obtuse, ciliate, the 2 inner ones rather smaller. Corolla white,
sometimes tinged with purple, 5-petaled, petals twice as long as
the calyx, rerlexed, ovate-oblong, obtuse. Stamina 5, inserted
below the ovarium ; alternating with these are sometimes found
5 short abortive filaments. Filaments very short. Anthers longer,.
DESCRIPTIONS OF MALA VAN PLANTS. 23 1
erect, conniving round the style, oblong, prolonged into acumina,
which are sometimes a little contorted, and which open at their
summits by a pore, the cells are adnate below to the sides of the
filament. Ovary oblong, acute. Style filiform, erect, equal to
the stamina. Stigma simple. Berry snow-white, globular, ob-
scurely angled, crowned with the persistent style, which is obliquely
deflexed ; of a spongy or farinose substance, containing in the
centre 5 seeds, which are disposed round the axis, and enclosed
in arilli composed of tough longitudinal fibres. Seeds (pyrenje ?)
oblong, somewhat reniform, hard. Albumen conform to the seed.
Embryo inverse, cylindrical, nearly as long as the seed. Cotyle-
dons semicylindric, obtuse. Radicle superior, longer than the
cotyledons.
The branches are terminated by long corniculate buds, in which
the gemmation is involute.
EUTHEMIS MINOR. W. J.
Foliis angusto-lanceolatis leviter serrulatis, racemis simplicibusi
baccis rubris-angulatis acuminatis.
Found at Singapore along with the preceding.
This is a smaller shrub than the former, branched, and smooth.
Leaves alternate, petiolate, linear-lanceolate, rather obtuse,
with a mucro, attenuated to the petiole, slightly serrulate, very
smooth, shining, finely striated with transverse veins. Petioles
short, thickened at the base, channelled above. Stipules linear,
siliate. Racemes simple, erect, at first terminal, becoming after-
wards lateral. Flowers alternate, pedicellate, often in pairs.
There is a single leaflike bract and several smaller ones at the
base of the pedicles, less deciduous than in the preceding.
Calyx 5-leaved, leaflets ovate, ciliate. Corolla white, spreading,
5-petaled, petals lanceolate, acute. Stamina 5, erect, conniving,
hypogynous ; filaments very short ; anthers yellow, oblong, broader
at the base, 2-celled, cells adnate to the sides of the filament,
prolonged above into an acumen opening at the top by a pore.
Ovary oblong, acute. Style a little longer than the stamina.
Stigma simple. Berry red, 5-angled, acuminate, composed of a
whitish farinaceous pulp, and containing 5 seeds, each enveloped
in a tough, fibrous arillus, and in structure the same as the
preceding.
CELASTRUS (?) B1VALIS. W. J.
Pentandria Monogynia.
Foliis lanceolatis acuminatis integerrimis, pedunculis lateralibus
paucifloris, corollis nullis, capsulis bivalvibus monospermis.
A shrub with smooth branches. Leaves opposite, petiolate,
lanceolate, acuminate, acute at the base, very entire, very smooth.
2 3 2 DESCRIPTIONS OF MALA VAN PLANTS.
Stipules none. Peduncles lateral, divaricately dichotomous, few
flowered (5-10 flowered). Bracts small. Calyx 5-parted, bibrac-
teate at the base, laciniae roundish, imbricated. Corolla none.
Stamina 5, erect, united beneath into a 5-toothed ring or urceolus ;
filaments flat ; anthers oblong. Style erect, as long as the stamina.
Stigma truncate. Capsule ovate, green, smooth, crowned with
the style, 2-valved, i-celled, i-seeded; valves opening from the
base, and falling off from the seed, which is more persistent, and
remains on the peduncle. Seed ovate, contained in a beautiful
crimson arillus, which is delicately veined. Albumen cartilaginous,
conform to the seed. Embryo erect, central, as long as the
albumen. Cotyledons flat, foliaceous, ovate, obtuse. Radicle
inferior, obverse to the umbilicus, round, much shorter than the
cotyledons.
STYPHELIA.
LEUCOPOGON MALAYANUM. W. J.
Pentandria Monogynia.
N. O. Epacridea. Br.
Spicis axillaribus multifloris erectis brevibus, drupis globosis
5-locularibus, foliis lanceolatis mucronatis subenerviis subtus
glaucescentibus.
M&itada. Malay.
Found abundantly at Singapore.
A small branchy shrub with hard dry leaves, exhibiting the
peculiar character of this family. Leaves alternate, sessile,
lanceolate, acute, mucronate, very entire, very smooth, shining
and convexed above, somewhat glaucous below, and, when ex-
amined by the microscope, appearing to be covered with numerous
very minute white dots, firm, with scarcely perceptible longitu-
dinal nerves. Spikes axillary, erect, much shorter than the
leaves ; peduncles somewhat tomentose. Calyx supported at the
base by two oval acute concave bracts, 5 -leaved, oblong, acute,
leaflets lanceolate, glaucescent, ciliate. Corolla infundibuliform,
a little longer than the calyx, qumquefid, puberulent, segments
lanceolate, bearded above beyond the base. Stamina 5, short,
alternate with the laciniae; filaments subulate; anthers subpen-
dulous, marked on each side with a longitudinal furrow, simple,
and bursting longitudinally in the manner so accurately described
by Mr. R. Brown, " Prodr. Fl. N. Roll." p. 535. Pollen globose.
Ovary surrounded at the base by 5 distinct erect obtuse scales,
5 -celled, each cell containing a single oblong ovulum. Style
erect, villous. Stigma subglobose. Drupe baccate, subglobose,
5-celled, cells i-seeded.
Obs. The discovery of this species is remarkable, as forming an
DESCRIPTIONS OF MALA VAN PLANTS. 233
exception to the general geographical distribution of the Epac-
rideae, a family almost exclusively confined to Australasia or at
least to the southern hemisphere. Singapore, situated at the
extremity of the Malay peninsula, and forming as it were the con-
necting link between continental or Western India and the islands
of the great Eastern Archipelago, partakes of this character in
its Flora, which exhibits many remarkable points of coincidence
with the Floras of both regions. I have had occasion to observe
resemblances between its productions and those of the northern
frontier of Bengal on the one hand, and of the Moluccas on the
other, while the present connects it with the still more distant
range of New Holland.
RAUWOLFIA SUMATRANA. W. J.
Pentandria Monogynia.
N. O. Apocynea.
Foliis ternis quaternisve elliptico-oblongis superne latioribus
glabris, floribus terminalibus umbellatis, corollce fauce villis
clausa.
Tarn pal badak or SSmbu badak. Malay.
Frequent in the neighbourhood of Bencoolen.
It grows to a small tree, having somewhat the habit and foliage
of the Mangga laut, or Cerbera manghas. The whole plant is
lactescent. Leaves verticillate, generally in threes, sometimes in
fours, short petioled, about 6 inches long, elliptic oblong, broader
above and terminating in a short point, very entire, very smooth,
rather firm, and having nearly transverse nerves. Peduncles 3 or
4, umbellate, terminal, long, round and smooth, bearing compound
umbels of small white flowers. Calyx small, 5-lobed. Corolla
white, tube longer than the calyx, limb spreading, 5-parted, lobes
subrotund, faux closed with white hairs which appear to form 5
tufts. Stamina 5 incluse ; filaments very short ; anthers yellow,
sagitate, acute, conniving over the stigma. Ovary furrowed on
both sides, 2-celled, tetrasporous, surrounded by an obscurely
5-lobed nectarial ring. Styles 2, united together. Stigma pel-
tate, capitate, glutinous, papillous. Berry globose, smooth,
containing 2 nuts, which are compressed, rugose, gibbous below
and tapering towards the top, subunilocular with an imperfect
dissepiment ; generally i -seeded. Seed compressed.
Obs. This species appears to have considerable resemblance to
Rauwolfia nitida, but is sufficiently distinguished by its inflor-
escence. The wood of this tree is very light, and employed by
the Sumatrans for the scabbards of their swords and krises.
234 DESCRIPTIONS OF MALA VAN PLANTS.
TACCA CRISTATA. W. J.
Foliis indivisis lato-lanceolatis, involucro diphyllo, umbella
secunda cernua superne intra involucrum foliolis duobus involucro
duplo longioribus stipata.
Native of Singapore and Pulo Pinang.
Root thick and tuberous, sending out a number of fibres.
Leaves nearly 2 feet long, numerous, radical, erect, petiolate,
ovate-lanceolate, acute, entire, smooth. Petioles sheathing at the
base. Scape erect, round, nearly as long as the leaves, striated,
smooth. Flowers peduncled, all drooping to one side; peduncles
subumbellate, arranged transversely in 2 parallel rows, and uniting
into a kind of crest, from which proceed 10 long pendulous
filaments. Involucre 2-leaved, leaflets ovate, acute, broad at the
base, nervose, purplish, twice as long as the peduncles, the upper
one erect, the lower reflexed and bent down by the drooping
flowers. From within the upper leaflet of the involucre spring 2
erect folioles, which are twice as long as the involucre, obovate,
attenuated below into straight, flat, deep purple petiolar ungues,
acute at the apex, pale coloured, with purplish nerves. Perianth
superior, of a dark purple colour, campanulate and somewhat
ventricose, rather contracted and 3-cornered at the mouth, where
it is also striated, limb 6-parted, somewhat reflex, lacinise hyaline,
oblong, broad, obtuse, the 3 interior ones larger. Corolla none.
Stamina 6, in the bottom of the perianth and opposite to the
lacinise. Filaments broad at the base, arching upwards into a
vaulted cucullus, within which the anthers are concealed. Anthers
adnate, 2-lobed. Style thick, shorter than the stamina, with 6
prominent angles. Stigma flat, umbilicate, orbicular, 6-rayed,
three alternate sinuses deeper. Berry ovate, 6-angled, i -celled,
seeds numerous, attached to 3 parietal receptacles.
* Obs. This approaches to T. integrifolia, "Curt. Mag." t. 1488,
but is a much larger plant, and is abundantly distinguished by the
2-leaved involucre, the long erect leaflets within it, and the flowers
drooping to one side.
VERATRUM (?) MALAY AN UM. W. J.
Foliis radicalibus lanceolatis, scapis erectis verticillato-pani-
culatis, baccis trilocularibus.
Native of Pulo Pinang.
An erect herbaceous plant. Leaves radical, 3 or 4 feet in
length, petiolate, lanceolate, acuminate, attenuated into a petiole
at the base, very entire, tomentose, striated with parallel nerves,
which run nearly longitudinally but diverge from a central one.
Petioles canaliculate, obtusely carinate, sheathing at the base.
DESCRIPTIONS OF MALA VAN PLANTS. 235
Scape erect, round, tomentose, verticillately panicled. Peduncles
alternately semiverticillate, divaricate and spreading. Flowers
sessile, on hermaphrodite or male plants fascicled, on female
solitary. Beneath each semiverticil is a large floral leaf, which is
ovate, acute, and contracted at the base into a flat, straight,
petiole-like unguis, which embraces the stem. Perianth 6-parted,
the 3 inner laciniae petaliform, white, spreading. Stamina 6 ; fila-
ments flat, dilated at the base. Styles 3, short. Stigmata 3.
In the female the calyx embraces a globular berry which is
3-celled, each cell i-seeded.
Obs. The true place of this plant is somewhat ambiguous, and
I am doubtful whether it can be admited as a genuine species of
Veratrum. It does not, however, agree exactly with any other
genus of the same family ; in habit it is somewhat like Alisma.
MEMECYLON CCERULEUM. W. J.
Octandria Monogynia.
Foliis cordatis amplexicaulibus, pedunculis axillaribus brevibus,
pedicel lis appositis divaricatis brevibus, fructibus ovatis.
Kulit nipis. Malay.
Native of Pulo Pinang.
A handsome shrub, of 10 or 12 feet in height, with round
smooth branches. Leaves opposite, subsessile, about 5 inches in
length, cordate, amplexicaul, oblong, acute, very entire, margin
reflexed, coriaceous, very smooth, deep green and shining above,
lateral nerves inconspicuous, uniting at their extremities into a line
which runs parallel to the margin. Stipules none. Peduncles
axillary, solitary, short, few-flowered; pedicels short and thick,
opposite, somewhat verticillate, divaricate, forming a kind of
corymbiform head. Flowers blue. Bracts opposite, short, acute.
Calyx superior, coloured, smooth, nearly entire, becoming by age
more distinctly 4-toothed. Corolla deep blue, 4-petaled, spread-
ing, petals broad, ovate, acute. Stamina 8, erect, shorter than the
corolla. Filaments short. Anthers blue, attached by their
middle, horizontal, shaped somewhat like the head of an axe, with
a knob behind ; cells parallel on the anterior edge. Before ex-
pansion the anthers are bent downwards (somewhat in the
manner of the Melastomae), and the surface of the germen and
bottom of the calyx are marked with their impressions, of which
the 4 inner are the deepest; the ridges between them form &
sharp prominent rays, and there are 8 other less conspicuous lines
formed by the faces of the bilocular anthers. Ovarium ovate, i-
celled, containing from 6 to 8 erect ovula. Style filiform, a little
longer than the stamina. Stigma acute. Berry cortical, crowned
by the persistent calyx, ovate, a little oblique at the base, i -seeded,
236 DESCRIPTIONS OF MALAYAN PLANTS.
the rudiments of the abortive ovula surrounding the umbilicus.
Seed ovate, umbilicate at the base and a little oblique. Albumen
none. Embryo erect. Cotyledons membranaceous, contortu-
plicate. Radicle cylindrical, nearly as long as the seed, obverse
to the umbilicus.
Obs. The different species of Memecylon have not been well
defined by authors ; this appears to differ from M. cordatum
Lamarck, and M. grande Retz., or Nedum chetti (Rheed,
"Mai." ii. p. 21. t. 15) in having ovate not globose fruit, and in
the flowers not being umbelled. In the latter the flowers are
small, yellow and numerous, in this they are larger, blue, and
much fewer in number.
LAURUS PARTHENOXYLON. W. J,
Enneandria Monogynia.
Foliis venosis ovatis acutis petiolatis subtus glaucis, paniculis
brevibus paucifloris axillaribus et lateralibus, fructu globoso calyci
truncato insidente.
Kayu gadis. Malay.
Abundant in the forests of Sumatra.
This is a lofty timber tree. Bark brown and rough. Leaves
alternate, rather long petioled, ovate, acute, often acuminate, and
varying in breadth, about 3 inches long, entire with somewhat
revolute edges, smooth, glaucous beneath, nerves lateral and
irregularly alternate. Petioles round, an inch long. Peduncles
from the young shoots at the extremity of the branches, axillary
or lateral, terminated by a short few-flowered panicle, and
generally longer than the young leaves, from whose axils they
spring. Bracts none. Perianth funnel-shaped, 6-parted, yellowish.
Stamina 9, arranged in 2 rows, the outer 6 naked, the inner
3 furnished at the base with 2 yellow glands ; filaments flat ;
anthers adnate, the cells opening with a longitudinal valve or
operculum. Style as long as the stamina. Stigma obtuse, 4-
cornered. Drupe seated on the enlarged cup-shaped, persistent
truncated base of the perianth, globose, containing a i-seeded
nut. Embryo inverse. Cotyledons hemispherical. Radicle
superior, within the edge of the cotyledons.
Obs. This species has considerable affinity to L. cuptilaria.
The fruit has a strong balsamic smell, and yields an oil, which is
considered useful in rheumatic affections, and has the same
balsamic odour as the fruit itself. An infusion of the root is
drank in the same manner as sassafras, which it appears to re-
semble in its qualities. The wood is strong and durable when
not exposed to wet, and in that case considered equal to teak.
Kayu gadis signifies the virgin tree, whence the specific name.
DESCRIPTIONS OF MALAYAN PLANTS. 237
May this be the Oriental sassafras wood mentioned under the
article " Laurus " in Rees' " Cyclopaedia ? "
GOMPHIA SUMATRANA. W. J.
Decandria Monogygnia.
N. O. OchnacecK.
Foliis lanceolatis vel oblongo ovalibus acuminatis obtuse
denticulatis nitidis subquinque nerviis, stipulis intrapetiolaribus
deciduis, paniculis terminalibus.
Siburu. Malay.
Sumatra.
A large shrub or small tree. Leaves alternate, short petioled,
8 or 9 inches in length, from lanceolate to oblong oval, varying
considerably in breadth from 2 to 3 inches, acuminate, acute at
the base, obtusely denticulate, very smooth, shining, middle nerve
very strong, lateral veins numerous, transverse, somewhat reticu-
late, delicate, uniting near each margin into two nerves, which run
parallel to it almost the whole length, and give the leaf the appear-
ance of being 5-nerved. Petioles very short. Stipules intra-
petiolar, broad at the base, acuminate, deciduous. Panicles
terminal, not much branched ; pedicles slender, rarely solitary,
surrounded at their bases by small acute bracts. Calyx 5-leaved,
persistent, leaflets ovate, acute, smooth, lucid. Corolla yellow, 5-
petaled, scarcely longer than the calyx. Stamina 10 ; filaments
very short ; anthers long, linear, opening at the top by 2 pores.
Style as long as the stamina. Stigma acute. Ovaries 5, surround-
ing the base of the style, and elevated on a receptacle. This
receptacle enlarges as the fruit ripens. The number of abortive
ovaries is variable ; sometimes only i comes to perfection.
The berries are drupaceous, obliquely reniform, somewhat com-
pressed, i-seeded. Seed exalbuminous.
Obs. This appears to have so much resemblance to the
G. Malabarica, Decarid. Puah Chetti (Rheed, "Mai." v. p. 103. t.
52), that I have some hesitation in proposing it as a distinct
species. The points of difference are the following : The leaves of
this are much longer than those of the Malabar species, which are
described as almost veinless, while in this the transverse veins
unite into two very distinct marginal nerves, which it is difficult to
suppose could have escaped observation had they existed in the
other. The representation of the inflorescence in Rheed's figure
is unintelligible, and his description of it is not much clearer ; but
as far as it can be made out, it appears different from this.
Further examination of the Malabar plant will be necessary to
determine whether this is really distinct, and whether the differ-
ences above noticed exist in the plant itself, or are mere omissions
in the description.
2 38 DESCRIPTIONS OF MALA VAN PLANTS.
MURRAYA PANICULATA.
Decandria Monogynia.
IV. O. Auranticz.
Foliolis ovatis acuminatis, floribus terminalibus axillaribusque
subsolitariis, baccis oblongis scepins dispermis.
Chalcas paniculata. Lour. "Fl. Coch." p. 270.
Camunium. Rumph. "Amb." v. p. 26, t 17.
Kamuning. Malay.
This is an abundantly distinct species from M. exotica, though
unaccountably confounded with it by later authors. Loureiro
discriminates between them very well, and his description is, on
the whole, good. Rumphius's figure is bad, but preserves several
of the distinguishing characters, particularly in the inflorescence
and leaves, which, however, are not sufficiently acuminate. It
grows to the size of a small tree, and the wood is much employed
for the handles of krises, being capable of receiving a fine polish.
The leaflets are generally 5, ovate, terminating in a long acumen,
which is slightly emarginate at the point, shining and very entire,
the terminal one considerably the largest. In M. exotica, the
leaflets are more numerous and closer, obovate, blunt, and of a
much firmer, thicker substance. The flowers of M. paniculata are
fewer and larger than those of M. exotica, and are sometimes
terminal, generally i or 2 together from the axils of the upper
leaves. The ovarium is 2-celled; the berries are oblong,
reddish, and mostly contain 2 seeds, which are covered with
silky hairs. The berries of M. exotica are ovate and generally
i -seeded. The whole habit of the 2 plants is very distinct. The
specific name paniculata is objectionable, as the flowers are much
less panicled than in the other species.
The Camunium sinense (Rumph. v. t. 18, f. i), which is com-
monly met with in gardens in all the Malay islands, is quite a
distinct genus from the other 2 Camuniums, and has been
described by Loureiro, "Fl. Cochinch. ;) i. p. 173, under the
name of
AGLAIA ODORATA.
It has a 5 -parted inferior calyx, and 5-petaled corolla. The
stamina are 5 in number, and are inserted in the manner of the
Meliacese on the inside of an ovate nectarial tube, which is
contracted at the mouth, and conceals the anthers. The stigma
is large, sessile, simple as far as I have observed, not double as
stated by Loureiro. The ovary appears to be i -celled, and to
contain 2 pendulous ovula. It rarely ripens its fruit in these
islands, but, according to Loureiro, it bears a small red i-seeded
berry. The flowers are very small, yellow and fragrant, in small
axillary panicles.
DESCRIPTIONS OF MALA VAN PLANTS. 239
In the catalogue of the "Hortus Bengalensis," p. 18, this plant
is specified under the name of Camunium sinense, after Rumphius.
The Murraya paniculata above described is the true Kamuning of
the Malays, and the name C. sinense is only applied by Rumphius
in the manner of the older botanical authors, as one of comparison
and resemblance, for want of a better of native origin ; if, therefore,
the generic name Camunium is to be adopted at all, it ought to
be applied to the plant to which it really belongs, and cannot be
admitted for one of a different family not indigenous to the Malay
islands. On this account Loureiro's name is to be preferred.
RHIZOPHORA CARYOPHYLLOIDES. W. J.
Dodecandria Monogynia.
Fruticosa, foliis ovato-lanceolatis utrinque acutis, pedunculis
axillaribus trifloris, rarius dichotome quinquefloris, floribus 8-fidis,
radicula subcylindrica acutiuscula.
Mangium caryophylloides. Rumph. " Amb." iii. p. 119, t. 78.
Found at Singapore and Pulo Pinang.
This is a much smaller shrub than the common mangrove, and
does not divide its roots so much. It is generally found in
shallow sandy salt marshes, rising with a tolerably erect stem and
branched nearer to the base than the common species. Leaves
opposite, petiolate, about 4 inches long, oval or ovate-lanceolate,
acute at both ends, sometimes slightly inequilateral, very entire,
very smooth, coriaceous ; the lower surface appearing under the
lens dotted with minute white points. Petioles round, furrowed
above, smooth. Stipules long, enveloping the corniculate buds in
the manner of the Ficus, very deciduous. Peduncles axillary,
solitary, 3-flowered, shorter than the petioles ; sometimes they
are dichotomously 5-flowered, having a flower in the bifurcation.
Calyx semi-inferior, surrounding the ovary, ovate, limb 8 -parted,
spreading, laciniss linear, acute, thick, rather incurved at their
points. Corolla white, 8-petaled, petals nearly erect, alternate
with the laciniae of the calyx, conduplicate, enclosing the stamina
by pairs, bifid, furnished with a few threads or filaments at the
point, ciliated on the margin. Stamina double, the number of
the petals inserted on the calyx in a double series, the inner ones
shorter, erect, not so long as the petals, enfolded by them until
the period of complete expansion, when they burst from their
recesses with an elastic force, and disperse their pollen. Anthers
linear, acute, 2 -celled. Ovarium contained within the calyx,
2-celled, tetrasporous ; ovula subrotund, affixed near the top of
the cells. Style filiform, as long as the stamina. Stigma bifid
with acute laciniae. Fruit contained in the persistent calyx, i-
seeded, the other 3 ovula proving abortive. The. seed is at first
24 o DESCRIPTIONS OF MALAYAN PLANTS.
ovate or roundish, with albumen conform, the embryo inverse, in
the upper part of the seed. As the fruit advances, the radicle is
elongated and becomes at length nearly cylindric, obsoletely
angled, and rather acute at the point. I have generally found
3 cotyledons, rarely 4.
Obs. Rumphius's figure is by no means a good representation
of the plant, but his description of it is correct. It comes nearest
to the R. cylindrica, Kari Kandel (Rheed, " Mai." vi. p. 59, t. 33),
which differs from this in having the radicle very obtuse and
more exactly cylindrical, and the peduncles generally i or 2-
flowered. According to Rumphius this species is rather rare, and
is called Mangi Mangi Chenke, or Clove Mangrove, whence his
appellation Caryophylloides, which I have thought proper to
retain, as the resemblance holds good in some particulars.
ACROTREMA. W. J.
Dodecandria Trigynia.
Calyx pentaphyllus. Corolla pentapetala patens. Stamina
quindecim, erecta, filamentis brevibus, antheris longis linceribus
apice biporis. Ovaria tria, distincta, 2-spora, ovulis angulo,
interiori affixis. Styli tres. Stigmata simplicia. Capsule uni-
loculares.
Herba acaulis, pilosa, pedunculis racemoso-multifloris.
Genus Saxifrageis afrme, numero partium inusitato distinctum.
ACROTREMA COSTATUM. W. J.
Found on hills, and among rocks at Pulo Pinang.
Root tapering, sending out a few fibres. Stem scarcely any.
Leaves alternate, spreading, short-petioled, 6-inches long, oblong-
obovate, obtuse, sagittate at the base, dentato-serrate, somewhat
ciliate, pilose, furnished with a short tomentum and also with
more remote longer appressed hairs ; the nerves are very hairy,
parallel, and terminate in the denticulse of the margin. Petioles
short, sheathing ; their margins dilated into membranaceous
auricles which might be considered as adnate stipules. Peduncles
or scapes central, erect, from 3 to 6 inches high, pilose, recurved
at the summit, 8 or lo-flowered. Flowers yellow, pedicellate,
racemose. Calyx 5-leaved, pilose, leaflets ovate acute. Corolla
yellow, spreading, 5-petaled, petals broader above, lanceolate.
Stamina 15, erect, hypogynous ; filaments very short. Anthers
very long, linear, 2-celled, opening by 2 pores at the top.
Ovaries 3, distinct, superior, i-celled, 2-seeded, each bearing i
style of the height of the stamina. Ovula attached to the inner
angles. Stigmata simple. Capsules 3.
DESCRIPTIONS OF MALA YAN PLANTS. 241
Obs. I am at a loss to determine the exact affinities of this
plant ; it has the habit of the Saxifragece, but the number of both
the male and female parts of fructification is greater by one-third,
and the ovaries are distinct.
LAGERSTRCEMIA FLORIBUNDA. W. J.
Icosandria Monogynia.
Foliis suboppositis ovato-oblongis glabris, paniculis terminalibus
ramosissimis multifloris ferrugineo vellosis, staminibus inequalibus
calycibus turbinatis sulcatis.
Found at Pulo Pinang.
A tree. Leaves subopposite, short-petioled, rather recurved,
7 or 8 inches long, ovate-oblong, somewhat acute, entire, smooth,
with strong prominent nerves and reticulate veins. Panicle
terminal, much branched, spreading, many-flowered. Peduncles,
pedicels and calyces ferruginous, densely villous with stellate hair.
The flowers are smaller than those of L. Reginae, but much more
numerous and in much larger panicles, pale rose colour on their
first expansion, and passing through various gradations of intensity
until at length they become nearly purple. Calyx covered with
ferruginous wool, turbinate, regularly marked with many deep
longitudinal furrows or ribs, giving it a fluted appearance, limb
spreading, 6-parted. Before expansion the calyx is obconical and
nearly flat at the top. Corolla 6-petaled, spreading, petals
inserted by short ungues alternately with the segments of the
calyx, ovate, not much undulated. Stamina red, numerous,
inserted on the calyx, 6 of them longer, thicker and more con-
spicuous than the rest. Ovary thickly covered with white hair, 6-
celled, many-seeded. Style erect. Stigma clavate.
Obs. This beautiful and splendid species may be readily dis-
tinguished from the L. Reginse by the greater size of the panicles,
and their ferruginous colour. The flower-buds in that species
represent in some degree a double cone, in this a single inverted
cone, being flat and even depressed at top. The L. hirsuta
" Lam." is also quite distinct from this, having hirsute leaves.
TERNSTRCEMIA RUBIGINOSA. W. J.
Polyandria Monogynia.
Foliis ovatis spinuloso serratis subtus incanis floribus lateralibus
et axillaribus fasciculatis, monadelphis pedunculis calycibusque
glanduloso-pilosis, fructu triloculari.
S'ingo ingo. Malay.
Sumatra.
A tree. Branches cinereous, young parts covered with acute
SECOND SERIES. VOL. II. R
242 DESCRIPTIONS OF MALA VAN PLANTS.
scales. Leaves alternate, petiolate, ovate, acuminate, spinuloso-
serrate, smooth above, hoary and white beneath, the nerves
furnished with ferruginous paleaceous scales. Flowers in fascicles,
lateral and axillary. Peduncles and calyces covered with glandular
hairs. Bracts small about the middle of the peduncles. Calyx
5-parted. Corolla white, campanulate rotate, 5-parted, divided
about half-way down. Stamina numerous ; filaments short, united
at the base into a ring which is inserted on the bottom of the
corolla ; anthers oblong, recurved, affixed by the middle, 2-celled,
opening at the top by two oblique pores. Ovary ovate, acute,
covered with glandular hairs, 3-celled, polyspermous, placentae
central. Style trifid, divided to the base. Stigmata simple.
TERNSTRCEMIA PENTAPETALA. W. J.
Foliis obovata-lanceolatis, spirmloso denticulatis glabris, floribus
lateralibus fasciculatis, pedunculis glabris fructu triloculari.
Native of Pulo Pinang.
A shrub with grey bark and leafy at the summit. Leaves alter-
nate, petiolate, i o to 12 inches long, obovato-lanceolate, acuminate
spinuloso-denticulate, smooth ; the nerves are furnished with a few
appressed, innocuous scale-like spines. Petioles about an inch in
length, covered, as well as the summits of the branches and buds,
with small ferruginous scales. Flowers in fascicles below the
leaves from the axils of the fallen ones of the preceding year;
they are pedicellate and white. Calyx coloured, 5-leaved, the two
outer leaflets smaller. Corolla white, 5-petaled, petals subrotund,
a little longer than the calyx. Stamina numerous, distinct, inserted
on the base of the petals ; filaments short ; anthers oblong, yel-
lowish white, didymous, truncate at the top and there opening by
two pores. Ovarium ovate, 3-celled, many seeded, placentae from
the inner angles of the cells. Style deeply trifid (Styles 3?).
Stigmata 3.
I have not seen the ripe fruit of this, but have been informed
that it produces a white berry.
EL^OCARPUS NITIDA. W. J.
Polyandria Monogynia.
Foliis ovato-lanceolatis, serratis, racemis axillaribus foliis
brevioribus staminibus quindecim, nuce quinqueloculari, loculis
plerumque quatuor abortivis.
Bua manik. Malay.
Native of Pulo Pinang.
A tree of moderate size, with grey bark and round smooth
branches. Leaves alternate petiolate, 3 or 4 inches long, ovate-
DESCRIPTIONS OF MALA VAN / 7 . / . 7.S. 243
lanceolate, acuminate, obtusely serrate, attenuated to the base, very
smooth. Stipules none. Racemes simple, axillary, secund, shorter
than the leaves. Flowers white, short-pedicelled. Calyx deeply
5 -parted, lacinise linear, acute. Corolla 5-petaled, fimbriated at
the summit. Nectary of 5 yellow retuse glands surrounding the
ovary. Stamina 15, erect; 10 are inserted by pairs between the
glands of the nectary, the remaining 5 between those glands and
the ovary. Anthers linear, bilamellate at the summit. Style as
long at the calyx. Stigma simple. Drupe globose, containing a
5-celled nut, which is rugose and marked with 5 obtuse longitu-
dinal ridges ; in general only i cell is fertile and contains a single
seed. Seed furnished with albumen; embryo inverse with flat
cotyledons and superior radicle.
Obs. This may perhaps be one of the smaller varieties of
Ganitrus mentioned by Rumphius ; it differs from K. ganitrus of
Roxburgh, who quotes Rumphius, III. t. 10, in the number of the
stamina, the position of the racemes, and the number of fertile
cells in the nut. Compare Adenoda sylvestris, Loureiro, " FI.
Cochinch." which agrees in the number of the stamina. I sus-
pect Gaertner must have fallen into an error in representing the
embryo erect in his Ganitrus ; in this it is certainly inverse.
MONOCERA. W. J.
Elaocarpi species.
Calyx pentaphyllus. Corolla pentapetala, petalis apice laciniatis
-saepe sericeis. Stamina plura, antheris apice dehiscentibus, uni-
cornibus valvula latera majore. Ovarium basi glandulis cinctum,
biloculare, polysporum. Drupa nuce 1-2 sperma.
This genus, whose characters appear to be sufficiently distinct,
will include, besides the following new species, several hitherto
referred to Elaeocarpus viz., E. Monocera cavanilles, the separation
of which has already been suggested, and of which the specific
name may be appropriately adopted for the genus E. rugosus, E.
aristatus, and E. bilocularis of Roxburgh, probably also E. grandi-
flora and E. reticulata, Sir J. E. Smith in Rees' " Cyclopaedia."
The E. dentata, Dicera dentata, Forst., may also belong to this,
if, as remarked by Sir J. E. Smith, Rees' " Cyclopaedia " in loco y
the anthers have only one of their valves awned, not both equal,
as originally stated by Forster. His capsule may perhaps be only
the ovary, which will then agree with the present genus.
MONOCERA PETIOLATA. W. J.
Foliis longe petiolatus ovato-lanceolatis integris labris, racemis
axillaribus foliis brevioribus, petilis medio intas incrassatis villosis.
Native of Pulo Pinang.
R 2
244 DESCRIPTIONS OF MALA VAN PLANTS.
A lofty tree. Leaves petiolate, alternate or scattered, 8 or 9>
inches long exclusive of the petiole, ovate lanceolate, generally
obtusely acuminate, entire, very smooth, deep green and shining
above, with lucid nerves and veins which are destitute of glands.
Petioles 4 inches long, smooth, thickened at the base and summit.
Racemes axillary, as long as the petioles ; flowers pedicellate,
turning one way. Calyx white, 5-leaved, leaflets lanceolate acu-
minate. Corolla white, 5-petaled, as long as the calyx, petals
ovate lanceolate, fringed at the point, sericeous without, thickened
along the middle, and covered with white hairs within, margins
inflexed. Ten thick 'sub rotund yellow glands surround the
stamina. Stamina numerous (23-30), inserted within the glands,
erect, shorter than the petals; filaments short; anthers longer,
linear, bivalved at the apex, the outer valve elongated, the inner
short and acute. Style filiform, longer than the stamina. Stigma
acute. Ovary ovate. 2-celled, many seeded. Drupe ovate, con-
taining a smooth, i -celled, i-2-seeded nut.
MONOCERA FERRUGINEA. W. J.
Foliis oblongo-ovatis acuminatis integris subtus cum pedunculis
ramulesque ferrugineo villosis, racemis axillaribus foliis brevioribus.
Found at Singapore.
A tree. Branchlets rusty and villous. Leaves irregularly alter-
nate, petiolate, oblong ovate, acuminate, 6 or 7 inches long, entire
with revolute edges, smooth above, ferruginously villous below,
nerves without glands. Petioles from 2 to z\ inches long, villous
and ferruginous, thickened under the leaf. Racemes axillary,
shorter than the leaves. Flowers pedicelled. Peduncles and
pedicels ferruginous. Drupe oval, of the form of an olive but
smaller, with a single rather smooth nut, which generally contains
but one perfect seed ; sometimes there is a second smaller, and
the vestiges of the partition and abortive ovula can almost always
be observed. Seed oblong, pointed above. Albumen conform ;
Embryo inverse, extending nearly the whole length of the albumen.
Cotyledons flat, oblong, with a distinct nerve along their middle.
Radicle superior clavato-cylindrical, much shorter than the
cotyledons.
Obs. I have not seen the flowers of this species, but its fruit
and general resemblance to the preceding leave no doubt as to
the genus, and its characters are sufficiently marked to distinguish
it from the others.
TETRACERA ARBORESCENS. W. J.
Polyandria Tetragynia.
Foliis obovatis integerrimisglabris,floribus paniculatis axillaribus
et terminalibus, calycibus pentaphyllis.
DESCRIPTIONS OF MALA VAN PLANT*. 245
Found near the shores of the Bay of Tapanooly in Sumatra.
Arborescent. Leaves alternate, petioled, about 3 inches lonj;,
oblong obovate, rounded at the apex and terminating in a short
point, very entire with reflex edges, smooth, shining above, coria-
ceous and firm, veins reticulate, nerves somewhat pilose or, me
under surface, petioles short. Panicles axillary and terminal,
many-flowered. Calyx 5-leaved, spreading, persistent, smooth,
Stamina numerous. Capsules generally 3, smooth and shining,
roundish ovate, opening on one side, containing a single seed
attached to the base of the capsule, and enveloped in a pale
yellowish laciniate arillus. The vestiges of two or three abortive
ovula are observable in the bottom of the capsule.
UVARIA HIRSUTA. W. J.
Polyandria Polygynia.
Tota hirsuta etiarn calyces fructusque pilis erectis, floribus sub-
solitariis, petalis patentibus subequalibus, foliis ovato oblongis basi
cordatis.
Pulo Pinang.
The whole plant is hirsute with long erect hairs. Branches
round. Leaves alternate, short-petioled, ovate oblong, acuminate,
cordate at the base, entire, simply pilose above, hirsute beneath
with stellate fasciculate hairs. Flowers lateral, almost solitary,
short-peduncled. Bracts lanceolate acute. Calyx hairy as well
as the peduncles and bracts, bursting irregularly, often into two
segments. Corolla of a deep red colour, 6-petaled, petals
spreading lanceolate acute. Stamina numerous, with long linear
anthers. Germina numerous; styles and stigmata the same.
Berries numerous, long pedicelled, oblong, hirsute with ferrugi-
nous hairs, many-seeded. Seeds arranged in a double longitudinal
series.
CAREYA MACROSTACHYA. W. J.
Monaddphia Polyandria.
Arbor, foliis petiolatis obovatis subserratis racemis lateralibus
nectantibus densissime multifloris, floribus sessilibus multi seriatis.
Pulo Pinang.
A tree with grey bark and smooth branches. Leaves alternate
or scattered, petiolate, obovate or oblong ovate, acuminate, some-
times obtuse with an acumen narrowing to the base, slightly
serrated, very smooth. Petioles roundish, thickened at the base.
Stipules none. Racemes or spikes lateral, hanging, thick, massive,
cylindrical, densely covered with flowers, which are sessile ana
.arranged in numerous spiral lines; the whole is 8 or 10 inches in
length. Bracts none. Calyx superior, purple, 4-parted, laciniai
246 DESCRIPTIONS OF MALA VAN PLANTS.
rounded, smooth, somewhat ciliated on the margin. Corolla
purplish red, longer than the calyx, 4-petaled, petals ovate, obtuse
inserted into the base of the calyx. Stamina white, very nu-
merous, longer than the corolla, united at the base into a thick
ring. Anthers yellow, didymous, the lobes bursting on opposite
sides, so as to give the whole the appearance of a double 4-celled
anther. Nectary surrounding the style within the stamina, hypo-
crateriform, red and striated within, yellow and entire on the
margin. Ovarium inferior, 4-celled, many-seeded ; about 4 seeds
in each cell attached to its upper and inner angle. Style red, as
long as the stamina. Stigma simple. Fruit a berry or pome.
Obs. -The inflorescence of this tree is very remarkable, and
quite different from the other species of Careya.
CLERODENDRUM DIVARICATUM. W. J.
Didy namia A ngio sperm ia .
Foliis qtovato-lanceolatis acuminatis glabris, paniculis termi-
nalibus ere|ps elongatis,pedicellis fructus reflexis, calyce subintegro
fructifero, vix aucto.
Found at Laye, on the West Coast of Sumatra.
Stem shrubby, erect, about 2 feet in height, smooth, with oppo-
site branches, which are thickened at the joint. Leaves opposite,
short petioled, obovate lanceolate, acuminate, entire, sometimes
denticulate, smooth. Panicle erect, terminal, long, composed of
opposite divaricate ramifications which are subdichotomous and
many-flowered. Pedicels of the fruit reflexed. Bracts large ovate,
acuminate, foliaceous. Calyx cup-shaped, nearly entire. Corolla
tubular, limb 5-parted, secund, the lower segment longer and of a
blue colour. Stamina long, exsert. Style i. Berry deep purple,
resting on the calyx, which is scarcely at all enlarged, 4-lobed,
4-seeded, from i to 3 seeds occasionally proving abortive.
HEDYCHIUM SUMATRANUM. W. J.
JV. O. Scitaminac.
Spica imbricata nutante, corollse labio bifido, laciniis oblongis-
divergentibus.
Gandasuli utan. Malay.
From Salumah, on the West Coast of Sumatra.
Stem erect. Leaves alternate, short petioled on their sheaths^
lanceolate, very entire, very smooth, parallel veined ; above a foot
in length. Sheaths smooth, prolonged into a very long ligula.
Spike terminal, nodding, short, dense, strobiliform. Bracts
lanceolate, as long as the calyx ; within this the ovary is
embraced by a tubular bract about half the length of the other.
DESCRIPTIONS OF MALA YAN PLANTS. 247
Flowers numerous. Calyx superior, tubular, oblique at the
mouth. Corolla long, outer limb 3-parted, with long narrow
segments; two segments of the interior limb much shorter and
broader ; the third segment or lip, which is united to the filament,
bifid, the divisions narrow and diverging. Filament very long,
embracing the style. Anther recurved, naked. Style length of
the stamen. Stigma thick. Ovary pilose, 3-celled, several
seeded. Nectarial bodies oblong.
Obs. This is a handsome species, and though its flowers are
not so large and showy as those of the H. coronarium, this is in
some degree compensated by the greater number which expand
at one time. It is the first wild species I have met with in the
Eastern islands.
ALPINIA ELATIOR. W. J.
N. O. Scitaminea.
Scapis radicalibus elatis, spicis ovatis, corolla? labio integro
basi mutico, foliis basi subcordatis glabris.
Bunga kenchong. Malay.
Found on Pulo Nias, also at Ayer Bangy on the West Coast of
Sumatra.
The stems are from 5 to 8 feet high, round, somewhat com-
pressed, smooth, striated. Leaves alternate, bifarious, petiolate
on their sheaths, ovate oblong, broad, subcordate at the base,
acuminate, very smooth on both sides, polished above, striated
with fine parallel nerves ; from i to 2 feet long. Ligula of
the sheath rounded. Scapes, rising at a little distance from the
stems, 2 or 3 feet high, erect, round, smooth, invested by
sheaths which are rounded at their points and mucronate below
the apex. Spikes short, thick, ovate, compact, densely covered
with flowers. The lower bracts are of a fine rosy colour, large
and spreading, so as to form a kind of involucre to the head ; the
upper bracts are shorter, imbricated, oblong or tongue-shaped,
rosy with white ciliate edges, each supporting a single flower,
The involucel or inner bract which embraces the ovary is tubular
and irregularly bifid, being cloven more deeply on one side than
the other. Calyx reddish, deeply cloven on one side, by which
the three regular segments become secund. Corolla, outer limb
three parted, segments nearly equal, erect, the upper one rather
the largest ; inner limb unilabiate, longer than the outer, lip
ascending, involving the anther, deep purplish red with yellow
edge, rhomboid ovate, entire, somewhat crisped at the point,
without spurs or sterile filaments at the base. Stamen shorter
than the lip ; anther naked. Style as long as the anther. Stigma
thick, triangular, anteriorly concave. Ovary sericeously pilose,
3-celled, many-seeded
248 DESCRIPTIONS OF MALA VAN PLANTS.
Obs. This is a very remarkable species, easily distinguished
from the other Alpinise with radical inflorescence by the great
height of the scapes, and the fine rosy colour of the lower
bracts.
ALPINIA CAPITELLATA. W. J.
Foliis longe petiolatis supra glabris, racemo terminali composito,
capitulis florum bracteis involucratis.
In the interior of Bencoolen.
Stems 4 or 5 feet high. Leaves alternate, bifarious, long
petioled on their sheaths, broad lanceolate, fine pointed, entire,
parallel veined, smooth above, slightly tomentose beneath.
Sheaths villous near the top, terminating above the petioles in
a long ciliate ligula. Raceme terminal, compound, inclining, red.
Flowers in heads which are embraced by large round bracts.
Calyx tubular, 3-cornered, nearly entire. Corolla, outer limb
3-parted, the upper segment fornicate ; the inner limb unilabiate,
of one large coloured segment. Stamen i, anther 2-lobed,
naked. Ovary tomentose, 3-celled. Style slender. Stigma
concave.
Obs. The peculiar manner in which the involucral bracts
embrace the capitulate flowers and subdivisions of the panicle
forms a good distinctive character. The whole inflorescence is
stiff and rigid, and wants that copiousness and richness which
marks the greater part of this splendid genus.
GLOBBA CILIATA. W. J.
Foliis ovato-lanceolatis nervis supra pilosis, panicula terminali
erecta, anthera bicalcarata.
Puar arnas. Malay.
Stem slender, erect, from i to 2 feet high, somewhat com-
pressed, spotted towards the base with purple. Leaves alternate,
bifarious, subsessile on their sheaths, ovate lanceolate, rounded at
the base, acuminate, entire, the upper surface furnished with
erect hairs disposed in lines along the principal nerves, lower
surface smooth, dotted under the lens with minute papillae ; about
4 inches long. Sheaths striated, smooth, ciliate along the margins,
extending very little beyond the petioles, and there bifid. Panicle
terminal, nearly erect, with alternate, divaricate, somewhat rigid
branches, on which are disposed alternately several subsessile
yellow flowers. Bracts lanceolate. Calyx trifid. Corolla orange-
yellow, 2-bordered, the exterior 3-parted, of which the upper
segment is largest and concave ; the inner consisting of 2
smaller segments alternating with the outer ones. Lip elevated
on the lower part of the filament and reflexed, emarginate, with a
DESCRIPTIONS OF MALAYAN PLANTS. 249
purple spot in the centre. Filament long, tubular. Anther with
2 subulate recurved horns or spurs. Style simple. Nectarial
bodies long and linear. Ovary containing several ovula.
Obs. It is a small delicate species, growing in moist hollows
on the sides of the hills and among the forests in most parts of
Sumatra. The ciliary lines of hairs on the upper surface of the
leaves distinguish it from most of its congeners.
ARISTOLOCHIA HASTATA. W. J.
Gynandria Hexandria. N. O. Aristolochuc.
Foliis hastato-trilobis glabris, racemis axillaribus, perianthio
basi inflato, lamina erecta elliptica marginibus revolutis.
Found at Natal on the West Coast of Sumatra.
Suffrutescent. Branches long, spreading over the neighbouring
shrubs, but not twining, angulate, jointed, smooth. Leaves
alternate, petiolate, from 6 to 10 inches long, hastately 3-lobed,
middle lobe elongated and terminating in a blunt acumen, very
entire, very smooth, 5-nerved, and strongly veined. Petioles
2 inches long, thick, round, channeled above. Racemes axillary,
longer than the petioles. Flowers alternate, pedicellate, some-
what distichous; rachis flexuose. Perianth superior, purplish-
red, smooth without, inflated at the base into an ovate 6-angled
ventricle, from which rises an ascending infundibuliform curved
tube with revolute margin ; lamina erect, elliptic, revolute at the
sides, tomentose on the inner surface, as is also the inside of the
tube. Style short, thick. Stigma orbicular, peltate, divided on
the summit into 6 conical erect lobes. Anthers sessile, regularly
arranged in a circle below the stigma, 6 in number, each con-
sisting of 2 lobes which are 2-celled and deeply furrowed along
the middle. (As these are not arranged by pairs, might they not
with equal propriety be considered as 12 distinct 2-celled
anthers ? ) Ovary oblong, obtusely 6-angled, 6-celled, many-
seeded.
Obs. This is a large and very beautiful species of Aristolochia,
remarkable for the size and form of its flowers. The ventricle at
the base is large, and the narrow urn-like tube rises upwards with
a very graceful curve. In this species the anthers might properly
be considered as 12 in number, each 2-celled, as they are all
arranged at equal distances round the stigma, and it seems ques-
tionable whether the genus itself ought not to be referred to
Dodecandria in place of Hexandria. The arrangement of the
anthers by pairs in the other species does not appear to neces-
sitate the supposition of a deviation from the usual structure in
Ascribing to them 4 parallel cells in place of the more usual
number of 2, nor does the analogy of other cognate genera
250 DESCRIPTIONS OF MALAYAN PLANTS.
furnish anything opposed to the inference so strongly suggested
by the present species.
BEGONIA. LINN.
The island of Sumatra abounds with Begonise, a tribe of plants
which are chiefly found in moist shady situations at the foot of
hills and in the recesses of forests. Being succulent herbs they
are with difficulty preserved in herbaria, and the specimens are
frequently deficient in one or other of the parts of fructification.
Descriptions from the living plants in their native soil are there-
fore particularly desirable, and in this view the following account
of the species which have fallen under my observation will not be
uninteresting. They seem to differ from all those described by
Mr. Dryander in the first volume of the Linnean " Transactions,"
and no great additions have been since made to our knowledge
of the genus.
BEGONIA C^SPITOSA. W. J.
Subacaulis, foliis inequaliter cordatis angulatis acuminatis
glabris, pedunculis dichotome cymosis, capsulse alis equalibus
obtusangulis v. rotundatis.
At Bencoolen.
Nearly stemless. Leaves petiolate, oblique, cordate at the
base, with rounded slightly unequal lobes overlapping each other
a little, somewhat falcate, rounded and sublobate on one side,
straighter on the other, attenuated into a long acumen or point,
spinulose but scarcely serrated on the margin, smooth, shining
above, pale and punctato-papillose beneath ; nerves 5-9,
branched towards the margin. The leaves are of unequal size
and vary somewhat in shape, the old ones being much rounder
and more decidedly lobed than the younger ones, which have the
point so much incurved as to be nearly falcate on one side.
Petioles red, pilose. Peduncles often as long as the leaves,
smooth, bearing a dichotomous cyme of white flowers. Bracts
ovate, concave. Male. Perianth 4 -leaved, the inner pair
smaller. Stamina numerous, collected into a head. Female.
Perianth superior, 3-leaved, 2 exterior large, subrotund, applied
to each other as in the male flowers, and enclosing the third,
which is much smaller and oblong. Style trifid. Stigmata lunato
bifid, yellow and glanduloso-pilose. Capsule 3-winged, wings
nearly equal, obtuse angled or rounded.
BEGONIA ORBICULATA. W. J.
Subacaulis, foliis orbicu latis cordatis crenatis glabris, pedunculis
subdichotomis, capsulce alis subequalibus obtusangulis.
DESCRIPTIONS OF MALA \ 'AN PLANTS. 2 5 1
Interior of Bencoolen.
Nearly stemless. Leaves petiolate, subrotund, from 3 to 4
inches in diameter, slightly oblique, cordate at the base where
the lobes overlap each other, remotely crenatc, rounded at
the point, smooth, except on the nerves of the under surface,
beautifully and finely punctate above. Stipules scariose,
acute. Peduncles erect, subdichotomous, nearly as long as the
leaves i.e., about 6 or 8 inches in height. Flowers white.
Male. Corolla 4-petaled, the outer pair large, oblong ; the
inner small. Stamina numerous. Female. Capsule 3-celled,
many-seeded, 3-winged ; wings obtuse angled, nearly equal.
BEGONIA SUBLOBATA. W. J.
Repens, foliis cordatis subquinque-lobis vel angulatis dentato-
serratis margine reflexis glabris, capsulse alis equalibus obtusan-
gulis.
Found under moist rocks on Pulo Penang, West Coast of
Sumatra.
Repent, with a thick knotty root. Leaves alternate, petiolate,
cordate, sometimes unequally, large and broad, often 6 or 7 inches
long, angulate, sometimes with 5 acute lobes, sometimes nearly
ovate, acuminate, dentato-serrate, edges recurved, very smooth,
5-7-nerved, finely punctate, the dots appearing elevated on the
upper surface and depressed on the lower. Petioles 4-6 inches
long, nearly smooth, furnished immediately below their junction
with the leaf with a semiverticil of linear acute appendices or
scales. Stipules large, ovate, rather laciniate towards the apex,
i on each side the petiole. Peduncles axillary, erect, 6-8 inches
long, red, very smooth, terminated by a dichotomous divaricate
panicle of white flowers tinged with red. Bracts roundish.
Male. Perianth 4-leaved, leaflets rather thick and fleshy, the 2
outer ones much larger and subrotund, before expansion com-
pletely enclosing the inner 2, and having their edges mutually
applied to each other in such a manner that they form an acute
carina round the unexpanded flower. Stamina numerous, in a
roundish head ; filaments short, inserted on a central column
which rises from the base of the flower. Anthers oblong, cells
adnate to the sides of the filaments, bursting longitudinally.
Female. Capsules with 3 equal obtusely angled wings, 3-celled,
3-valved, valves septiferous in the middle, sutures corresponding
to the wings. Seeds numerous, attached to placentae, which
project from the inner angle of the cells.
Obs The serratures are hard and cartilaginous, and recurved
in such a manner along with the margin of the leaf that, when
only observed on the upper surface, their place is perceived by
2 5 2 DESCRIPTIONS OF MALA VAN PLAN7S.
an indentation. It seems to resemble the B. grandis, Dryand.,
which differs, however, in having oblique, doubly serrated leaves
-and purple flowers.
BEGONIA FASCICULATA. W. J.
Foliis inferioribus alternis, superioribus oppositis, oblongo-
ovatis basi semicordatis duplicato-serratis pilosis, perianthiis
jnasculis diphyllis, capsulse alis equalibus obtusangulis.
Found at Tappanuly on the West Coast of Sumatra.
Caulescent. Stem weak, jointed, thickened at the joints,
round, covered with red hairs. Leaves petiolate, the lower
ones alternate, the upper ones opposite, oblong ovate, inequi-
lateral, semicordate at the base, acuminate, irregularly serrate,
covered above with red erect subspinescent hairs, beneath with
softer and weaker hairs. Petioles densely pilose. Stipules linear,
acuminate, pilose. The flowers come in fascicles from the middle
of the petioles, and these flower-bearing leaves are always opposed
to another without flowers ; hence it is that the upper leaves are
opposite, while the lower are alternate. Fascicles composed of
male and female flowers ; pedicels slender, smooth, white. Bracts
several at the base of the fascicles, acute, pilose, red. Male.
Perianth diphyllous, white. Stamina numerous. Anthers yellow.
Female. Perianth superior, white, cup-shaped, 5-leaved ; petals
ovate, acute, with a few short red hairs on the outside. Style
deeply trifid ; lobes convolute, infundibuliform. Capsule 3-winged,
3-celled, wings equal, obtuse-angled.
BEGONIA PILOSA. W. J.
Foliis subsessilibus irregulariter serratis acuminatis pilosis subtus
xubris, bracteis ad basin pedicellorum subrotundis ciliatis, capsulse
alis subequalibus parallelo rotundatis.
Interior of Bencoolen.
Caulescent, pilose. Leaves alternate, scarcely petiolate, ovate,
inequilateral, acuminate, slightly and irregularly serrate, pilose witli
long red hairs, under surface of a bright red colour ; about three
inches long. Stipules large, lanceolate, pilose externally. Pedun-
cles oppositifolious, subdichotomous. Bracts at the base of the
pedicels, roundish, ciliate. Flowers white. Male. Corolla 4-
petaled, the inner pair smaller. Stamina numerous. Female.
Corolla 5-petaled ; the two outer petals larger. Capsule 3-winged ;
wings nearly equal, parallel and rounded.
DESCRIPTIONS OF MALAYAN PLANTS. 253
BEGONIA BRACTEATA. W. J.
Foliis duplicato-serratis acuminatis pilosis, pedunculo i-3-floro
bracteis numerosis appressis vestito, capsulis basi bibracteatis, alls
equalibus rotundatis.
Near the foot of Gunong Bunko, in the interior of Bencoolen.
Suberect, strong and branching, very villous, shaggy. Leaves
alternate, short petioled, ovate, semicordate at the base, acuminate,
duplicato-serrate, pilose, 3-4 inches long. Stipules large, pilose.
Peduncles oppositifolious, generally supported by a smaller leaf,
invested particularly towards the base with many pair of opposite
ovate acute pilose ciliate bracts, which are pressed flat against
each other ; the uppermost pair is distant from the rest, and sup-
ports from i to 3 pedicels. Flowers white. Male. Corolla
4-petaled; the outer two large, subrotund. Stamina numerous.
Female. Corolla 5-petaled ; petals nearly equal. Styles three.
Stigmata lunate, villous with yellow short glandular hairs. Capsule
embraced by 2 bracts at the base, 3-celled, 3-winged ; wings equal,
rounded.
BEGONIA RACEMOSA.W. J.
Foliis obovato-oblongis irregulariter dentatis acuminatis glabris,
racemis erectis masculis, flore femineo axillari, perianthiis masculis
diphyllis, capsulae alis equalibus parallelo-rottmdatis.
Interior of Bencoolen.
Layang-layang simpai. Malay.
Suberect, stem smooth, jointed. Leaves alternate, short pe-
tioled, obovate oblong, attenuated towards the base, which is
unequally cordate, acuminate, irregularly and unequally dentate,
smooth, 6-7 inches long. Stipules large, oblong. Racemes,
oppositifolious, long, erect, bearing numerous fasciculate male
flowers, and having a single female one in the axil. Male.
Corolla 2-petaled, petals very thick. Stamina numerous. Female.
Capsule with 3 equal parallel rounded wings, 3-celled.
BEGONIA GENICULATA. W. J.
Caule geniculato, foliis ovato-oblongis denticulatis acuminatis
glabris, pedunculis divaricate dichotomis, floribus superioribus
masculis dipetalis, inferioribus femineis, capsulae alis equalibus
obtus angulis.
Rumput udang-udang. Malay.
Sumatra.
Caulescent, stems smooth, compressed, channelled, jointed,
thickened at the articulations. Leaves alternate, petiolate, semi-
cordate at the base, obovate oblong, acuminate, denticulate, smooth.
254 DESCRIPTIONS OF MALA YAN PLANTS.
Peduncles oppositifolious, dichotomous, divaricate, many-flowered,
lower flowers female, upper male. There is often a female flower
from the axil. Male. Perianth 2 -petal ed, white. Stamina
numerous; anthers oblong, broader above. Female. Capsules
long, 3 -winged, wings obtuse-angled, equal, smooth.
Obs. The leaves of this plant are used by the natives for
cleaning andtaking out rust from the blades of crises. It has
considerable resemblance to the preceding species.
SONERILA HETEROPHYLLA. W. J.
Foliis oppositis altero minimo reniformi, altero oblongo acumi-
nato versus basin attenuate ibique semicordato, supra glabris,
pedunculis axillaribus brevissimis paucifloris.
. Found at Tappanuly, on the West Coast of Sumatra.
Stem creeping, round, covered with appressed scaly hairs.
Leaves opposite, almost sessile, one very minute and reniform,
the other about 3 inches long, oblong, broader above, acumi-
nate, narrowing to the base, semicordate, the outer lobe forming a
rounded auricle, obsoletely denticulate or nearly entire, a small
spinule on the denticulations ; 3-nerved, smooth above, whitish
beneath, with some hairs on the nerves. Petioles scarce any.
Flowers from the axils of the small leaves, sometimes nearly soli-
tary, sometimes 4 or 5 on a very short peduncle. Pedicels reddish,
seated on small tubercles, furnished with glandular hairs. Calyx
superior, trifid. Corolla 3-petaled. Stamina 3. Capsule turbi-
nate, 3-celled, many-seeded.
Obs. This species is remarkable by the extreme difference in
the size of the opposite leaves, one of which is so minute as almost
to escape observation. The same peculiarity exists in the Sonerila
Moluccana.
RHODODENDRON MALAYANUM. W. J.
Foliis oblongis glabris punctatis, floribus terminalibus, pedicellis
cernuis, corolla punctata basi gibba.
Observed on the summit of the Sugar Loaf Mountain, in the
interior of Bencoolen.
This is a large shrub or small tree much branched. Bark brown
and spotted. Leaves alternate or scattered, short petioled, lanceo-
late-linear, 2^-3 inches long, attenuated to both ends, somewhat
bluntish at the point, entire, smooth, thickly sprinkled beneath
with brown dots and green above with depressed points; the
middle nerve is strong, the lateral ones scarce any. Stipules none.
Flowers from a short terminal bud, which is at first closely invested
by numerous imbricated broad bracts, which successively fall off
DESCRIPTIONS OF MALA VAN PLANTS. 255
and at length leave the short thick peduncle annulated by their
cicatrices. It throws out near the point several nodding i -flowered
pedicels, which are doited in the same manner as the leaves.
Calyx very small, 5-toothed. Corolla crimson, tubular, expanding
into a 5-lobed limb, sprinkled with callous dots, tube gibbous at
the base and marked with 5 furrows. Stamina 10, leaning to one
side, inserted on the very base of the corolla and about as long as
its limb; filaments red; anthers yellow, opening at top by 2
.oblique pores. Style a little shorter than the stamina. Stigma a
round head marked with 5 indistinct rays. Ovary superior, oblong,
5-sided, covered with brown spots, 5-celled, polysporous.
Obs. I found this and the following species of Vaccinium on
the very summit of Gunong Bunko, a remarkable insulated moun-
tain in the interior of Bencoolen, commonly called by Europeans
the Sugar Loaf, in reference to its shape. Its elevation is not
estimated to exceed 3,000 feet, yet the character of its vegetation
is decidedly Alpine. This character is probably more marked
than it would be at a similar height on the side of a differently
shaped hill, owing to the steepness, which refuses space for large
trees ; and the consequent exposure and want of shelter on its
sharp conical peak.
VACCINIUM SUMATRANUM. W. J.
Racemis axillaribus foliis brevioribus, foliis elliptico-ovatis in-
tegerrimis coriaceis.
Found on the summit of Gunong Bunko, or the Sugar Loaf
Mountain, in the interior of Bencoolen.
A small tree, with reddish brown bark and smooth branches.
Leaves alternate, short petioled, elliptic ovate, acuminate, some-
times obtuse, entire, edges a little reflexed, very smooth, firm, stiff
and leathery, pale green beneath ; about 4 inches long. Stipules
none. Racemes axillary, shorter than the leaves, often from the
stem below them. Flowers white, pedicellate, alternate. Calyx
small, cup-shaped, slightly 4-toothed. Corolla oblong ovate, con-
tracted at the mouth; limb short, recurved, 4-parted. Stamina
8 incluse, inserted on the base of the corolla ; filaments dilated at
the base, pilose, tinged with red ; anthers 2-lobed, between which
are 2 short filaments or processes, each lobe prolonged upwards
into a membranaceous horn or awn, which is bifid at top and
opens by a pore. Ovary semi-inferior, 4-celled, polysporous, ovula
attached to the inner angles of the cells. Style columnar, a little
longer than the stamina, incluse. Stigma round, obtuse.
256 DESCRIPTIONS OF MALAYAN PLANTS.
HALORAGIS DISTICH A. W. J.
Foliis alternis distichis obliquis integris, floribus axillaribus-
subsolitariis, petalis tridentatis.
Kayo kanchil. Malay.
This species is not unfrequent in Sumatra, at Singapore, and
other parts of the Malay Archipelago.
A shrub with ferruginous pilose branches. Leaves alternate,
distichous, arranged in two series, one of large leaves and another
of very small ones which resemble stipulse, being regularly placed
a little below the insertion of the large ones, so as to lie over their
bases ; the large leaves are subsessile, rhomboid oblong, inequi-
lateral, acute, entire, nearly smooth above, pilose with short
appressed hairs beneath ; from an inch to an inch and a half long ;
the small leaves are similar in shape, but more acute, and little
more than a quarter of an inch long ; they are arranged on the
anterior side of the branch and are closely appressed to it, so as
to resemble stipules. Flowers axillary, generally solitary, subses-
sile. Calyx 4-leaved, persistent. Petals 4, shorter than the calyx,
trifid. Stamina 8, as long as the petals ; anthers 2-celled. Ovary
inferior, 4-sided, ferruginous, 4-celled, tetrasporous. Styles 4,
equal to the stamina. Stigmas simple. Drupe oblong ovate, red,
containing a nut with 8 longitudinal furrows, and containing a
single seed. Seed oblong oval; embryo central in an ample
albumen.
Obs. The general habit of this species is very peculiar, and
has much the character of Australasian vegetation, to which
country the genus principally belongs.
ELODEA. ADANSON.
Hyperirincz, fuss.
This genus, which has been revived by a late author on
American Botany, appears to be abundantly distinguished from
Hypericum, and to form a good natural division. It is principally
characterized by having the stamina united into 3 phalanges,
which alternate with an equal number of nectaries. In the fol-
lowing species the placentation is peculiar; I know not whether
the American plants exhibit the same structure, as it is not men-
tioned in any description which I have seen, but if it should prove
on examination that they do, it should form part of the generic
character. Loureiro's Hypericum Cochinchinense, which un-
doubtedly belongs to Elodea, appears to be very nearly related to
my E. Sumatrana, and his description of the seeds seems to indicate
a structure similar to what I have observed. The Hypericum
DESCRIPTIONS OF MALA VAN PLANTS. 2 5 7
petiolatum of the same author seems also referable to this genus,
and to be different from Lianseus's H. petiolatum, whk;
native of Brazil. In all the species now referred to Elodea, the
generic distinction appears to receive confirmation from certain
differences of habit which may be remarked between them and
the true Hyperica, particularly in the colour of the flowers, which
in the latter is almost without exception yellow, but in Elodea is
often red.
ELODEA SUMATRANA. W. J.
Foliis subsessilibus oblongis attenuato-acuminatis glabris rigi-
diusculis, paniculis terminalibus foliosis, staminibus numerosis
triadelphis, petalis basi nudis.
Found at Tello Dalam, in the island of Pulo Nias.
A large shrub or small tree; branchlets rather compressed,
obscurely 4-sided. Leaves opposite, almost sessile, oblong,
tapering to the point, acute, broad at the base, entire, smooth ;
nerves proceeding from a middle rib, strong ; 6 or 7 inches in
length ; the surface appears by the aid of the microscope to be
dotted with opaque points. Panicles terminal, foliose, the lower
divisions being axillary ; oppositely branched and rigid. Flowers
dark red or purple. Bracts minute. Calyx 5-leaved, persistent,
leaflets ovate, smooth, the outer ones smaller. Corolla cup-shaped,
longer than the calyx, 5-petaled ; petals subrotund ; ungues naked,
without pore or scale. Nectaries 3, yellow, inserted below the
corolla and half as large as the petals, subrotund, doubled back-
wards upon themselves in such a manner as to form a sac which
opens behind near the base. Stamina numerous, their filaments
united for about half their length into three phalanges, which are
inserted 'alternately with the three nectaries; they are a little
shorter than the corolla ; anthers yellow, 2-celled. Ovary oblong,
3-celled, many-seeded. Styles 3, diverging. Stigmata 3, subro-
tund. Capsule oblong, 3-celled ; each cell containing several seeds
as long as the cell and attached to the bottom of the central
column ; they are thin and flat, disposed regularly one within the
other, forming concentric circles, which are particularly apparent
in the transverse section of the capsule.
Obs. This curious arrangement of the seeds is not a little
remarkable ; they lie one within the other like skins of an onion,
each occupying the full length and breadth of the cell, but
diminishing regularly in size from the outermost to the middle in
proportion to the different radius of the circle which it describes
round the common centre. They are attached one above the
other to the bottom of the cell at its inner angle. The leaves are
destitute of pellucid dots, and have their lateral nerves strongly
and distinctly marked. The nectaries which alternate with the
SECOND SERIES. VOL. II.
258 DESCRIPTIONS OF MALAYAN PLANTS.
stamina are very peculiar, being saccate, apparently by being
doubled backwards. This species differs from the following and
those of America in having no scales at the base of the petals, and
from the latter in having numerous stamina. It appears to be
nearly related to Loureiro's Hypericum Cochinchinense, which, as
already observed, belongs to this genus.
ELODEA FORMOSA. W. J.
Foliis petiolatis lanceolatis subtus glaucis pedunculis fasciculatis
axillaribus, staminibus numerosis triadelphis, nectariis acutis.
Kayo gaghak. Lampong. Sepadas Bunga. Malay.
Native of Sumatra.
A small tree with cinereous bark and smooth branchlets. Leaves
opposite, elliptic oblong, acute, very entire, smooth, glaucous
beneath, pellucidly punctate, two and a half inches long; the
nerves proceed from a midrib. Petioles slender. Peduncles
axillary, and from the axils of fallen leaves, fasciculate, i -flowered,
slender, smooth. Flowers white, with a slight rosy tinge. Bracts
several at the base of the peduncles. Calyx 5-leaved, smooth,
leaflets acute. Corolla 5-petaled, longer than the calyx ; petals
oblong, each furnished with a broad adnate scale a little above
the base. Stamina numerous, united into 3 phalanges. Nectaries
3, alternating with the stamineous fascicles, red, acute, carinate
behind, fleshy. Ovary 3-celled, each cell containing several flat
ovula lying one within the other, and attached by their bases to
the lower part of the axis. Styles 3, long. Stigmas capitate.
Capsules oblong, crowned by the persistent styles, 3-celled, many-
seeded. Seeds thin, flat, attached by their bases to a central
triangular column, on which they are inserted alternately in a
double series.
Obs. The arrangement of the ovula is similar to that observed
in the E. Sumatrana ; they are thin, attached by their bases to the
lower part of the cell, suberect, and concentrically disposed, but
are inserted rather higher on the axis of the cell than in the
former. This species agrees with those of America in having a
scale at the base of the petals, but differs in having numerous
stamina ; it therefore comes nearer to the E. Egyptica (Hypericum
Egypticum, Linn.).
TERNSTRCEMIA.
The Malayan species of Ternstrcemia exhibit a remarkable
agreement among themselves, at the same time that they differ
considerably from the rest of the genus. They have a trilocular
ovarium surmounted by 3 styles, which are inserted on the same
point, but are separate to the base. In some the corolla is mono-
DESCRIPTIONS OF If A I.. IV. LV PLANTS. 2 y>
q)etalous with monadelphous stamina, in others it is 5-petaled with
distinct stamina. The anthers are 2-celled and open at the top by
2 oblique pores ; this is probably the case with the whole genus,
though it has been omitted in the generic character, of which
it ought certainly to form an essential part. It seems doubtful
whether the monogynous species with bilocular fruit and definite
seeds ought to be united with those which have 3 styles, 3 cells
and numerous seeds ; but an examination of their ovaries and
placentation is necessary to decide the question. I have met witli
four species in Sumatra and the adjacent islands, two of which I
have already described in the first volume of the " Malayan Miscel-
lanies." Their common appellation in Malay is Ingor-ingor
karbau, or Buffalo's spittle.
TERNSTRCEMIA ACUMINATA. W. J.
Foliis obovato-lanceolatis acuminatis spinuloso-denticulatis
.glabris, floribus axillaribus solitariis polyandris, pedunculis squa-
mosis, fructu triloculari.
Found at Tappanuly on the West Coast of Sumatra.
Branches round, somewhat flexuose. All the young parts green,
with a few appressed scales. Leaves alternate, petiolate, obovate
lanceolate, attenuated to the base, terminating in a long acumen
or point, spinuloso-denticulate, smooth, with the exception of a few
appressed scales on the lower surface ; about a foot in length.
Petioles short, scaly. Peduncles axillary, solitary, i -flowered,
scarcely so long as the petioles, covered with small scales. Calyx
5-leaved, the 3 outer leaflets with appressed scalets. Corolla
white, 5-petaled, little longer than the calyx. Stamina many,
inserted on the base of the petals ; anthers large, truncate and
opening by 2 pores at the top. Ovary 3-celled, many-seeded.
Styles 3.
Obs. This agrees with the T. pentapetala in having the corolla
divided to the base, but the leaves are more acuminate, and the
flowers are solitary and axillary.
TERNSTRCEMIA SERRATA. W. J.
Foliis obovato-oblongis cartilagineo-serratis glabris, pedunculis
axillaribus binis, floribus monadelphis, laciniis corollae emarginatis,
fructu triloculari.
Frequent on the island of Pulo Nias.
A small tree. Young parts furnished with brownish scales.
Leaves alternate, petiolate, obovate oblong, acuminate, serrate
with irregular cartilaginous uncinate serratures, smooth, pretty
strongly nerved ; 7-8 inches long. Petioles brown, scaly.
s 7.
260 DESCRIPTIONS OF MALA VAN PLANTS.
Peduncles generally 2, axillary, i -flowered, slender, about an inch
long. Calyx 5-parted, whitish, leaflets unequal. Corolla white,
inonopetalous, quinquefid, longer than the calyx, cup-shaped, lobes
bifid or emarginate, generally oblique. Stamina shorter than the
corolla, and inserted on its base ; filaments united below ; anthers
oblong, bifid, 2-celled, each cell opening at top by an oblique
cucullate pore. Ovary hairy, 3-celled, many-seeded ; placentas
central. Styles 3, longer than the corolla, irregularly bent. Berry
3-celled, many-seeded. Seeds angled, foveolate.
Qbs. This differs from the other Sumatran species in having
firmer leaves, with stronger nerves and thickened callous serratures.
The peduncles are more slender, the styles longer, and the lobes
of the corolla obliquely notched.
TERNSTRCEMIA CUSPIDATA. W. J.
Foliis obovato-ellipticis acuminatis dentato-serratis, serraturis,
apice hamatis, fructibus 5-locularibus. pedunculis axillaribus 1-3
floris.
A tree, young parts ferruginous. Leaves petiolate, elliptic ovate,
attenuated to the base, broader above, sharply acuminate, serrated,
the narrow sharp toothlets generally curved or hooked at their
points, smooth, often marked with whitish glandular dots on the
nerves, veins,, and serratures ; 6-8 inches long. Peduncles axillary.
1-3 flowered, smooth. Calyx 5-parted, segments orbicular,
Corolla white, monopetalous, 5-parted. Stamina numerous ;
anthers opening by two gaping pores. Ovary subglobose, 5-celled,
ovula very numerous ; placentae from the inner angle of the cells.
Style very deeply 5-parted.
Obs. This species (received from Salumah during the printing
of the present sheet) comes very near to the T. serrata ; it differs
in having the leaves more sharply acuminate, with longer tooth-
like serratures, and rather shorter petioles, the peduncles fre-
quently bearing 2 or 3 flowers, and not so slender as in the former
and in the 5-celled fruit.
MILLINGTONIA. ROXB.
Calyx 5-phyllus, foliolis duobus exterioribus minoribus. Corolla
5-petala, petalis duobus minoribus squamiformibus. Stamina
quinque, quorum tria sterilia difformia basi petalorum majorum
inserta; duo fertilia basi minorum adnata, filamentis apice scyphum
gerentibus cui anthene bilobse insident. Ovarium nectario annulari
rinctum, biloculare, loculis disporis. Drupa mice plerumque
monosperma. Embryo erectus, curvatus albumine nuilo aut
]>arco.
DESCRIPTIONS OF MALAYAN PLANTS
Obs. It will be perceived that I have made a considerable and
material alteration in the terms of the generic description from
that given by Roxburgh ("Fl. Ind." i. p. 102), which I conceive
to be necessary towards explaining the true relations of the various
parts of the flower, and thereby affording the means of tr;i< iiv^
more correctly its natural affinities. The principal point is to
determine the real nature of what Roxburgh calls the nectarial
scales at the base of his petals. I have no hesitation in consider-
ing them as abortive stamina, which the examination of the flower
before expansion places, I think, beyond a doubt. In that state
the whole of the stamina connive over the pistil, the anther of the
fertile ones is turned inward, so as not to be visible, and there is
no considerable difference of appearance between them and the
sterile ones. The anther-bearing hollow of the fertile stamina is
applied to a corresponding hollow on the side of the sterile ones,
and at the time of expansion the former separate themselves with
a jerk and become erect, while the latter continue in their original
position incumbent over the pistil. The petals on which the
fertile stamina are inserted are much smaller and narrower than
the others, as if exhausted by the greater development of the parts
they nourish. These petals are called by Roxburgh outer lamina:
of the filaments, which is contrary to all common analogy, while
the other explanation might be supported by numerous examples
of a similar structure. Thus in place of a diandrous flower with
tripetalous appendiculate corolla and bifid stamina, we obtain 5
as the primary number of all the parts, only modified by the
partial abortion of 3 of the stamina.
MILLINGTONIA SUMATRANA. VV. J.
Foliis impari-pinnatis, foliolis 3-6 jugis ovato-lanceolatis, petalis
minoribus acutis, fructu ovato.
Found on the island of Pulo Nias.
It is a moderate-sized tree with grey bark. Leaves alternate,
pinnate with an odd one which is rarely wanting ; leaflets from 5
to 13, opposite, ovate, lanceolate, acuminate, entire, smooth, 6-9
inches long. Common petiole flat above and marginate, thickened
at the base. Panicles terminal, many-flowered, rather coarctate,
with stiff rigid divisions, slightly tomentose. Flowers white. Bracts
minute. Calyx small, 5-leaved, the outer 2 smaller, resembling
bracts. Corolla 5-petaled, the outer 3 large, subrotund, the
inner 2 much smaller, lanceolate, acute. Stamina 5, inserted
on the bases of the petals; 2 fertile, upon the smaller petals,
with broad filaments expanding at top into a kind of cup, on
which the anther rests, and to whose outer edge it is attached ;
the anther consists of 2 yellow lobes resembling masses of
262 DESCRIPTIONS OF MALA I 'AN PLANTS.
pollen which burst transversely. The 3 sterile stamina which
are inserted on the larger petals have thick filaments without
anthers, but marked with an oblong cup-like cavity on each side
corresponding to the cups of the fertile ones. Before expansion,
the 5 stamina connive over the pistil in such a manner that
the cup-like cavities are mutually applied to each other ; on
expansion, the fertile stamina separate with a jerk, by which the
pollen is in part dispersed, and the cup becomes erect with the
anther resting upon it; the other 3 never separate but remain
conniving over the pistil. Ovary embraced at the base by
a nectarial cup with 5 toothlets; ovate, 2-celled, each cell
containing 2 ovula attached to the centre of the partition.
Style short. Stigma small. Berry ovate, oblique or recurved,
somewhat less than an olive, containing a single i -seeded nut.
Nut obovate oblong, acute and curved at the base, carinate along
one side, and having a large umbilical hollow above the base
on the other, smooth, i -seeded. Seed obovate oblong, acute
at the base, covered with a dry loose brown skin ; albumen
none ; embryo glutinous on the surface, erect, doubled on itself.
Cotyledons thin, foliaceous, large, round ovate, reflected back
wards upon the radicle, and half embracing it laterally. Radicle
inferior, very large, thick, pointed, extending the whole length of
the seed and partly doubled up or curved at the top.
Ol>s. The cotyledons are wrapped round the embryo in such a.
manner as to give the whole somewhat of a chrysaloid appear
ance. This species has considerable resemblance to the M.
pinnata of Roxburgh, but differs in having unequally pinnate
leaves, with from 3 to 6 pair of leaflets, in having the smaller
petals entire and acute, not tridentate, in the nectarial ring
having 5 simple toothlets, not 3 bidentate angles, and in having
a large ovate fruit with a smooth, not rugose nut. The abortive
cell is generally observable near the umbilical foramen.
LAURUS INCRASSATUS. W. J.
Foliis ovato-lanceolatis venosis, pedunculis fructus incrassatis.
rubris.
Machilus medius. Rumph. " Amb." iii. p. 70. t. 41.
Jaring-jaring tupai. Malay.
J-'ound at Natal in the island of Sumatra.
A tree. Leaves alternate, petiolate, ovate lanceolate or lanceo-
late, acuminate, entire, very smooth, with lateral nerves proceed-
ing from a middle rib ; about 5 inches long. Petioles short.
1'cduncles axillary or lateral near the extremity of the branches,
shorter than the leaves, supporting a small panicle of flowers.
In the flower these peduncles and pedicels are slender and
DESCRIPTIONS OF MALA VAN PLANTS. 263
delicate, but as the fruit advances they become very much
thickened, fleshy and red. Perianth 6-parted. Stamina 9, the 3
inner ones glandular at the base and somewhat villous ; anthers
opening by longitudinal valves. Style short. Stigma capitate,
angled. Berry seated on the incrassated peduncle, and embraced
at the base by the divisions of the perianth a little enlarged,
about the size and shape of an olive, purple, i -seeded. Seed
oval, exalbuminous. Radicle superior, far within the edge of the
cotyledons.
Obs. I have met with another species at Bencoolen with
larger leaves, from 9 to 12 inches in length, in which the
pedicels alone are thickened, the peduncles remaining unaltered.
In this particular it agrees perhaps still better with Rumphius's
figure than the plant above described.
TETRANTHERA CORDATA. W. J.
N. O. Laurincz.
Racemis axillaribus, floribus umbellatis enneandris, filamentis
pilosis, perianthii limbo sexpartito, foliis cordatis subrotundo-
ovatis uninervibus costatis subtus ramulis pedunculis involucris-
que ferrugineo villosis.
West Coast of Sumatra.
A moderate-sized tree. Leaves alternate, petiolate, cordate,
sometimes sinuate-cordate, varying from subrotund-ovate to ob-
long oval, rather acute, smooth above, tomentose beneath, nerves
proceeding from a middle rib, veins transverse, subreticulate.
Peduncles axillary, shorter than the leaves, bearing a raceme of
involucred umbels. Involucres 5-leaved, leaflets roundish, to-
mentose without, deciduous. Umbels sessile on the involucre,
4-7-flowered; flowers pedicelled. Male. Perianth 6-parted.
Stamina 9, hairy, the inner 3 filaments furnished with large
glands ; anthers 4-celled. Female. Perianth 6-parted, segments
narrow. Sterile stamina 9, the inner 3 with large double
glands ; filaments pilose with long hairs. Style i, longer
than the stamina. Stigma dilated, sublobate. Berry oblong,
i -seeded.
KNEMA CxLAUCESCENS. W. J.
N. 0. MyristicecE. BR.
Glomerulis axillaribus 2-6 floris, floribus pedicellatis, baccis
oblongo-ovalibus subpulverulentis, foliis oblongis sursum attenuatis
subtus glaucis, antheris 12-15.
In the neighbourhood of Bencoolen.
A dioecious tree. The young parts covered with rusty down.
Leaves alternate, short-petioled, oblong, generally rounded at the
264 DESCRIPTIONS OF MALA VAN PLANTS.
base, attenuated upwards, acute, very entire, deep green and
shining above, glaucous beneath, the adult leaves nearly smooth,
the young ones furnished with short stellate pubescence on the
under surface; lateral nerves simple; about 7 inches long by
2 broad. Petioles somewhat rusty, a third of an inch in length.
Stipules none. Flowers 2-6, glomerate on a short axillary knob,
pedicellate ; pedicels as long as the petioles, ferruginously tomen-
tose. A minute bract about the middle of each pedicel. Male.
Perianth ferruginously tomentose without, deeply 3-parted, spread-
ing, segments round ovate, thick ; aestivation valvate. Stamineous
column central, slender, expanding at top into a peltate disc,
whose edge is divided into 12 or 15 rays, to the lower surfaces of
which are attached an equal number of 2-celled anthers. Female.
Fruit axillary, generally solitary, hanging, oblong oval, consider-
ably smaller than an olive, somewhat pulverulent and rusty, burst-
ing into 2 valves. Nut invested by a thin aril, which is laciniate
only at the top. Seed with ruminate albumen.
Obs. The seed has a pungent taste and slightly aromatic smell.
Mr. Brown has recognized the propriety of separating Knema
from Myristica.
CONNARUS. LINN.
This genus, with Cnestis, has been removed by Mr. R. Brown
from the Terebintaceae of Jussieu, and formed into a separate
and very natural family under the name of Connaraceae. They
are rather a numerous tribe in the Malay islands, and besides the
following species of Connarus and Cnestis, I have to add the new
genus Eurycoma, which appears to be sufficiently distinct from
both the former. I am doubtful whether the species which I
have referred to Cnestis really belong to that genus, as they have
all smooth capsules with arilled or carunculate seeds, or whether
they ought not to be separated from those whose capsules are
clothed with prurient hair. Some confusion appears also to have
existed between the species of Cnestis and Connarus, the ripe
capsules of the former being often solitary from the abortion of
the remaining ovaries, and I am much inclined to think that
Connarus santaloides, and mimosoides of Vahl in particular, are in
reality species of Cnestis, a supposition which is supported by the
analogy of the inflorescence, which is almost without exception
terminal in Connarus and axillary in Cnestis. This distinction is
of some importance between genera so nearly related.
CONNARUS FERRUGINEUS. W. J.
Eerrugineo-tomentosa. foliis pinnatis, foliolis oblongis coriaceis
subtus ferrugineo-villosis, paniculis terminalibus.
Bunga burutta. Malay.
DESCRIPTIONS OF MALA YAN PLANTS. 265
Native of Pulo Pinang.
A small sized tree. Branches round, covered with ferrugin-
ous wool. Leaves alternate, pinnate, leaflets 9, sub-opposite,
oblong lanceolate, acuminate, very entire, margins reflexed, coria-
ceous, green and tomentose above, ferruginously villous beneath.
Petioles round, villous, thickened at the base. Stipules none.
Panicles large, terminal, sometimes with a few axillary racemes.
Flowers numerous, white. Bracts roundish, often curved, ferrugin-
ously villous, as well as the calyces and the whole panicle. Calyx
5-parted, laciniae erect, oblong, acute. Corolla white, sprinkled
with red dots, 5-petaled, longer than the calyx, petals erect,
lanceolate. Stamina 10, erect, united at the base, the alternate
ones much shorter. Anthers ovate. Style shorter than the long
stamina. Stigma capitate, 3-furrowed. Capsule follicular, ferru-
ginous, rather inflated, oblique, gibbous behind, opening on one
side, i -celled, i-seeded. Seed bean-shaped, appendiculate at the
umbilicus. Umbilical appendage or caruncle large, and glandular.
Embryo dicotyledonous, conform to the seed, without albumen ;
radicle at a distance from the umbilicus.
Obs. This fine species is well distinguished by its thick
leathery leaves, and the ferruginous pubescence of their lower
surface and of the branches and panicles.
CONNARUS VILLOSA. W. J.
Villosissima, foliolis 5-7 lanceolatis longe acuminatis supra
glabris, paniculis terminalibus dense stellato-villosis ferrugineis.
Native of Sumatra.
The whole plant densely and ferruginously woolly. Branches
round. Leaves alternate, pinnate, leaflets 5 or 7, subopposite,
oblong lanceolate, narrowing towards the base, terminating in a
long acumen, entire, smooth above, villous beneath with stellate
pubescence \ about 6 inches long. In young leaves the upper
surface is covered with deciduous pubescence. Panicles large,
terminal, and from the upper axils densely villous, ferruginous.
Bracts long, linear, thick, curved, villous. Calyx 5-parted,
villous. Corolla 5-petaled, limb spreading. Stamina 10, united
into a ring at the base, the alternate ones shorter. Ovary densely
pilose with plumose hairs. Style longer than the stamina.
Stigma capitate.
Obs. This plant is covered with denser and rougher wool than
the preceding, particularly on the panicles, and the leaves are
much longer, acuminate, and not coriaceous.
266 DESCRIPTIONS OF MALA YAN PLANTS.
CONNARUS SEMIDKCANDRA. W. J.
Foliis pinnatis, foliolis 3-5 lato-lanceolatis subtus villosiusculis,
paniculis terminalibus axiilaribusque villosis, filamentis alternis
sterilibus.
Mangul, also Akar sedinka. Malay.
Abundant in thickets at various places on the West Coast o
Sumatra.
It is a small tree, with wrinkled bark ; the young shoots and
leaves are softly and ferruginously villous. Leaves alternate,
pinnate ; leaflets from 3-5, ovate lanceolate, acuminate, entire,
smooth above, slightly villous beneath, nerves lucid, 3-4 inches
long. Panicles terminal or from the upper axils, villous and
brownish. Flowers numerous. Bracts small. Calyx 5-leaved,
erect, reddish, tomentose. Corolla of a light blush colour, 5-
petaled; petals longer than the calyx, spreading at the limb.
Stamina, filaments 5, fertile, exsert ; 5 alternate ones, short, sterile,
all united into a ring at the base. Style somewhat shorter than
the stamina. Capsule tomentose, ferruginous, follicular, 2-valved,
i-seeded. Seed with an umbilical caruncle.
Obs. This is one of the most common species in Sumatra, and,
like all the rest of the genus, frequents thickets and copses, or
what is called by the Malays " Beiukar," rather than the great
forests.
CONNARUS GRANDIS. W. J.
Foliis pinnatis, foliolis quinis ovato-lanceolatis glabris, paniculis
terminalibus, capsulis magnis glabris.
At Tappanuly, in Sumatra.
A moderate-sized tree. Leaves alternate, pinnate; leaflets
generally 5, ovate lanceolate, acuminate, entire, smooth, 8 or 9
inches long. Panicles terminal, long, smooth. Capsules large,
oblique, red, smooth, follicular, bursting on one side, i -seeded.
Seed with a large umbilical caruncle.
Obs. I have not seen the flowers. It has larger leaves and
fruit than any other species that I have met with, and is further
distinguished by the smoothness of all its parts.
CONNARUS LUCIDUS. W. J.
Foliis pinnatis, foliolis glaberrimis nitidis emarginato-acuminatis,
paniculis terminalibus ferrugineis, calyce persistente.
Sumatra.
A small tree, with long divaricate subscandent branches. Bark
brown and wrinkled. Leaves alternate ; leaflets 5-9, ovate
lanceolate or elliptic oblong, terminating in a long linear acumen,
DESCRIPTIONS OF MALA VAN 1'LANTS.
which is emarginate at the point, entire, very smooth, shining and
lucid, 2-2$ inches long. Panicles terminal, small and delicate,
ferruginously tomentose. Flowers pale red. Calyx 5-leaved,
tomentose. Corolla 5-petaled, petals narrow. Stamina 10, mon-
adelphous at the base, the alternate ones short. Style i, longer
than the staminn. Capsule obovate, less oblique than usual in
the genus, embraced at the base by the enlarged persistent calyx,
smooth, bursting on one side, i-seeded. Seed attached nearly at
'the base, the umbilicus half embraced by the cup-shaped carun-
cula, which is rather smaller than usual.
Obs. This is a small delicate species, having smooth, shining
leaves with emarginate points ; the panicles are small and seldom
bring more than i or 2 fruit to perfection.
CNESTIS EMARGINATA. W, J.
Foliolis 5-7, acuminatis apice emarginatis racemis axillaribus
paucifloris, capsulis solitariis glabris, seminis umbilico caruncula
semiamplexo.
Found in the neighbourhood of Bencoolen.
A small tree with weak diffuse branches. Leaves alternate,
pinnate, leaflets 5-7, from ovate to oblong ovate, terminating in a
long acumen which is emarginate at the point, entire, very smooth,
the middle nerve pubescent underneath ; the upper leaflet is the
largest, and frequently 5 inches in length. Petiole thickened at
top and bottom, almost articulate under the terminal leaflet.
Racemis axillary, subsolitary, short, few-flowered. Pedicels alter-
nate, i -flowered; a bract at the end of each pedicel, small,
tomentose as well as the peduncle. Calyx 5-parted, smooth,
persistent. Corolla 5-petaled, petals oblong, acute. Stamina 10
distinct, the alternate ones shorter. Ovaries 5, smooth, with a
line of hairs along the suture. Styles 5, shorter than the stamina.
Stigmas emarginate. Capsule solitary, 4 ovaries aborting, em-
braced at the base by the thickened calyx, orange-coloured,
smooth, bursting on one side, containing a single black seed.
Seed furnished at the base with a cup-shaped orange-coloured
fleshy caruncle which partially surrounds the umbilicus. Embryo
inverse, without albumen.
Obs. The umbilical caruncle in this species is similar in shape
and situation to that observed in the Connarus lucidus, being
smaller than usual in this tribe.
CNESTIS FLORIDA. W. J.
Foliolis 3-5, rarius solitariis, oblongo-ovatis acuminatis glaber-
rimis, racemis fasciculatis axillaribus, seminibus arillo subinclusis.
Confer cum Connaro santaloide. Vahl, anne eadem ?
^68 DESCRIPTIONS OF MALAYAN PLANTS.
Found in Sumatra and the island of Pulo Nias.
A small tree, with somewhat rigid divaricate branches. Leaves
alternate, pinnate, leaflets 3-5, sometimes solitary, oblong ovate,
attenuated into a longish blunt acumen, very entire, very smooth,
rather rigid, shining above, veins reticulate, about 3 inches long.
Racemes axillary, fasciculate, slender, shorter than the leaves, the
lower pedicels 3-4 flowered. Calyx almost 5-leaved, erect, tinged
with red towards the base. Corolla 5-petaled. Stamina 10,
distinct, nearly equal, filaments flat and broader at the base.
Ovaries 5, oblong, erect. Styles i to each ovary. Stigmas simple.
Capsule solitary, the remainder aborting, ovate, pointed towards
both ends, somewhat oblique, smooth, bursting on one side,
i -seeded. Seed almost enclosed in a bright red lleshy aril, origi-
nating from the umbilicus, and in its expansion enveloping the
whole seed. Albumen none. Cotyledons plano-convex, solid.
Radicle remote from the umbilicus, as in Gcertner's Omphalobium.
CNESTIS MIMOSOIDES. W. J.
Foliis pinnatis subdecemjugis, foliolis ovato-oblongis emarginatis,
seminibus arillo subinclusis.
Connarus mimosoides. Vahl and Willd.
Found at Tappanuly.
I can scarcely entertain a doubt of this being the very plant
referred by Vahl to Connarus, and aptly named Mimosoides. Its
analogy with the preceding is very close, having the seeds simi-
larly enclosed in a large red aril, and the racemes axillary. I
have not seen the flowers, but the four abortive ovaries are quite
distinct at base of the perfect one. In all these three species only
one capsule ripens, in which particular, as well as having smooth
capsules and arilled seeds, they seem to differ from Cnestis.
EURYCOMA. W. J.
Pentandria Monogynia. N. O. Connaracecc. BROWN.
Calyx 5-partitus. Corolla 5-petala. Stamina quinque. Glan-
dulse decem staminibus alternse. Ovarium 5-lobum, lobis mono-
sporis. Stylus i. Stigmata quinque. Capsulee 3-5, folliculares,
glabree, monospermse. Semen nudum.
Polygama, foliis pinnatis fastigiatis, floribus paniculatis.
EURYCOMA LONGIFOLIA.
Kayu kabal. Malay.
Found at Tappanuly and Bencoolen, in Sumatra, and at
-Singapore.
DESCRIPTIONS OF MALA VAN PLANTS.
This is a small tree, whose branches are thick, rough with the
vestiges of fallen leaves and foliose at their summits. I
crowded at the extremity of the branches, 2 feet long, pinnated
with numerous leaflets, which are oblong lanceolate, acute, very
entire, very smooth, 2-3 inches in length. Panicles axillary, very
long. Flowers male and hermaphrodite on different plants.
Calyx small, 5-parted. Corolla longer than the calyx, purple,
tomentose without, with glandular hairs, petals erect with inflexed
.margins. Stamina 5, erect, shorter than the petals, alternating
with 5 pair of villous corpuscles, which are large and distinct in
the male flower, very small in the hermaphrodite. Ovary 5-lobed,
lobes monosporous, in the male very small and abortive. Style i,
short, curved. Stigmata 5, thick, recurved. Capsules from 3 to 5,
nearly ovate, smooth, bursting on one side, i-seeded. Seed naked
(without aril or caruncle), exalbuminous.
Obs. The corpuscles interposed between the stamina are
remarkable in the male flower, being roundish, erect, yellow bodies,
with somewhat the appearance of abortive anthers ; in the herma-
phrodite, however, they become simple scales. The genus differs
from Cnestis in the number of the stamina, the single style, and
the smoothness of the capsules, and from Connarus in the number
of the ovaries and stigmas, and the want of the umbilical
caruncula.
PERONEMA. W. J.
Didynamia Angiospcrmia. N, O. Verberacece. BR.
Calyx 5-partitus. Corolla tubo brevi, limbo irregulari 5-lobo,
laciniis secundis. Stamino duo, exserta ; rudimenta duorum ste-
rilium. Stigma refractum. Fructus siccus, 4-partibilis, 4-spermus.
Arbor, foliis pinnatis petiolo alato, panicula; terminal! opposite
corymbosa.
PERONEMA CANESCENS.
Sungkei. Malay.
A large tree, native of Sumatra.
Trunk straight, but little branched. Leaves opposite, pinnate,
nearly 2 feet long, with 7-9 pair of leaflets, which are alternate or
subopposite, lanceolate, attenuated to both ends, acute, somewhat
recurved, entire, smooth above, canescent beneath, veins reticulate
on the under surface, 8-9 inches long. Petioles winged, finely
and delicately tomentose, wings decurrent from the insertion of
the leaflets. Stipules none. The branches are crowned by a vast
terminal oppositely corymbose panicle, of which the ultimate
divisions are dichotornous, with a flower in the bifurcations ; the
whole is finely tomentose and hoary. Bracts small, acute. Flowers
inconspicuous, whitish. Calyx 5-parted, segments acute, erect.
270 DESCRIPTIONS OF MALA YAN PLANTS.
Corolla not much longer than the calyx, limb expanding, irregular,
5-lobed, segments secund, the two upper ones diverging, the
lowermost considerably longer than the rest. Stamina 2, reflexed
backwards between the upper segments of the corolla, filaments
subulate, thickened towards the base. Anthers long. Rudiments
of 2 abortive stamina. Ovary 4-celled, ovula erect. Style rather
longer than the stamina. Stigma simple, refracted. Fruit seated
on the calyx, villous, dry, separating into 4 portions, each of which
contains a single seed.
Obs. This is a valuable timber tree, the wood being hard and
tough, well suited for carriage shafts, which require to combine
strength and elasticity with lightness. When long buried in the
earth, it is said to become petrified. The genus is related to
Vitex, but is abundantly distinct therefrom.
RHODAMNIA. W. J.
Icosandria Monogynia. N. O. Myrta,
Calyx superus, quadrilobus. Corolla tetrapetala. Stamina
numerosa. Ovarium uniloculare, pluri ovulatum, placentis duobus
parietalibus. Bacca unilocularis oligosperma.
Arbuscula, foliis trinerviis, inflorescentia axillari.
RHODAMNIA CINEREA.
Frequent on the Western Coast of Sumatra and the islands
which skirt it. Its Malay name is Marpuyan.
There are two varieties of this species, the one of which is
larger than the other and has broader leaves, which are more
decidedly tomentose below. These differences are scarcely
sufficient for a specific distinction.
A small tree with greyish wrinkled bark and pilose branchlets.
Leaves opposite and alternate, petiolate, roundish- ovate in the
large variety, and broad lanceolate in the small one, acuminate,
very entire, 3-nerved, often with a less distinct pair near the
margin, smooth above, somewhat hoary beneath, pubescent,
particularly on the nerves, but in the small variety nearly smooth,
with little more than a glaucous tinge on the under surface.
Petioles short, tomentose. Stipules small, linear. Peduncles
short, axillary, i-flowered. Flowers white. Calyx tomentose,
persistent. Corolla twice as long as the calyx. Stamina inserted
on the calyx, almost as long as the corolla. Ovary i -celled, con-
taining many ovula attached to 2 parietal placentae. Style i,
erect. Berry reddish, subglobose, crowned with the calyx, i-
celled, containing a few seeds attached to the parietes, many of
the ovula proving abortive.
DESCRIPTIONS OF MALAYAN /'/
Obs. This genus, which is nearly related to Myrtus, appears to
be sufficiently distinguished by its ovary and plac entation, from
which, rather than from the fruit, the most important characters in
this family are to be derived. It is peculiar in having 3-nerved
leaves, in which particular it has a resemblance to Myrtus
tomentosa, but differs widely from that species in its fruit and
ovary.
ADINANDRA. W. J.
Polyandria Monogynia.
Calyx 5-partitus, persistens, basi bibracteatus. Corolla penta-
petala, petalis basi latis. Stamina 30, pluriseriata, subpolyadelpha,
interioribus brevioribus ; antheris bilocularibus apice mucronatis.
Stylus unicus, subulatus. Bacca supera, stylo persistente acumi-
nata, 5-locularis, polysperma, placentis ab angulo interiore loculos
bipartientibus.
Arborescens foliis alternis exstipularibus, floribus axillaribus.
ADINANDRA DUMOSA.
Daun saribu. Malay.
Abundant in thickets throughout Sumatra and various parts of
the Malay islands.
It grows to be a small tree ; the bark is dark brown, and the
branches are smooth. Leaves alternate, short petioled, elliptic
oblong, acute at both ends, sometimes rounded, with an obtuse
acumen at top, entire or obsoletely serrate, smooth, slightly glau-
cous beneath, almost veinless ; 3-4 inches long. Stipules none.
Peduncles axillary, subsolitary, i-flowered, shorter than the leaves,
recurved calyx bibracteate at the base, 5-parted, segments thick,
subrotund, overlapping each other. Corolla white, twice as long
as the calyx, erect or conniving, 5-petaled, petals ovate oblong,
broad at the base, acute. Stamina about 30, closely arranged
in several circles, the inner ones shorter; filaments divisible to
their bases, but closely pressed against each other, sericeously
pilose, particularly on their outer side; anthers of 2 parallel
lobes, adnate to the sides of the filament, which is prolonged into
a mucro at the summit. Ovary superior, smooth, 5-celled, poly-
sporous ; the cells are almost biparted by placentae which project
from the inner angle, and to whose edges the ovula are attached.
Style single, subulate. Stigma simple. Berry globose, embraced
at the base by the calyx, and acuminated by the persistent style ;
5-celled, many seeded.
Obs, In general habit and in the texture of the leaves this
plant has some resemblance to Dyospyros, but differs widely in
fructification.
272 DESCRIPTIONS OF MALA YAN PLANTS.
IXONANTHES. W. J.
Calyx 5-passim 6-partitus, foliolis subrottmdis. Corolla 5 vel 6-
petala, glutinosa. Stamina 10 vel 20. Nectarium germen cingens.
Stylus i. Capsula supera, calyce corollaque persistentibus cincta
ovato-acuminata, 5-locularis, 5-valvis, valvularum marginibus
introflexis. Semina singulo loculo duo, margini interior! dissepi-
mentorum affixa, compressa, in alam membranaceam producta.
Albumen semini conforme, embryone inverse foliaceo, piano.
Arbores, foliis alternis simplicibus, floribus, dichotome corym-
bosis axillaribus.
IXONANTHES RETICULATA.
Floribus decandris, foliis integerrimis.
Found at Tappanuly, on the West Coast of Sumatra.
A tree with smooth compressed branchlets. Leaves alternate,
petiolate, elliptic oblong, emarginate, somewhat attenuated to the
base, entire, smooth, firm and rigid, with thick revolute edges,
shining above, rather glaucous beneath, veins reticulate, about 3
inches long. Petioles short, flattened above. Stipules minute,
deciduous. Peduncles axillary on the younger shoots, much
longer than the leaves, smooth, dichotomous at the summit, with
a pedicel in the bifurcation, bearing generally about 7 flowers,
which are small and green. Calyx 5-parted, segments rounded.
Corolla glutinous as well as the calyx, 5-petaled, petals roundish.
Stamina 10; filaments inserted below the petals; anthers yellow,
2-celled. Ovary surrounded at the base by a yellow fleshy nec-
tarial ring, 5-celled, lo-seeded. Style erect. Stigma capitate.
Capsule surrounded at the base by the persistent calyx and corolla,
somewhat enlarged, oblong, pointed, smooth, 5-valved, 5-celled,
septa formed by the introflexed margins of the valves, cells
2 -seeded, but frequently only one comes to perfection ; they are
separated from each other by a ridge which projects from the
middle of the valves. Seeds compressed, oblong, angular, winged
at the lower end. Albumen conform to the seed. Embryo
inverse, central. Cotyledons flat, oval. Radicle superior, cylin-
drical, not so long as the cotyledons.
IXONANTHES ICOSANDRA.
Floribus icosandris, foliis crenatis.
Found in the interior of Bencoolen.
A tree. Leaves alternate or scattered, short-petioled, lanceolate
oblong, emarginate, dentato crenate, very smooth, shining above ;
about 6 inches long. Stipules small, deciduous. Peduncles axil-
DESCRIPTIONS OF MALA VAN PLANTS. 273
lary, nearly as long as the leaves, bearing a trichotomous umbel
or corymb of greenish flowers. Bracts small. Calyx 5-6-parted.
Corolla 5-6-petaled, glutinous as well as the calyx, petals spreading,
subrotund, pale, and somewhat transparent. Stamina 20, much
longer than the corolla, Nectarial ring crenate on the margin by
the compression of the filaments which are inserted round it.
Ovary 5-6-celled, each cell containing 2 ovula. Style a little
longer than the stamina. Stigma capitate. Capsule ovate, pointed
smooth, 5-6-celled, 5-6-valved, margins of the valves introflexed.
Seeds 2 in each cell, attached by their middle to the inner edge
of the valvular partitions, oblong, membranaceous to both ends,
bifid at the lower.
CHIONOTRIA. W. J.
Decandria Monogynia.
Calyx 5-partitus inferus. Corolla 5-petala. Stamina 10, erecta.
Ovarium 2 loculare 2 sporum, ovulis pendulis. Stylus i. Stigme
capitatum. Bacca monosperma. Semen exalbuminosum apice
umbilicatum. Cotyledonibus maximis convexo-planis, radicula
supera minima,
Frutex, foliis simplicibus oppositis pellucido punctatis, racemis
axillaribus.
Genus Aurantiis affine.
CHIONOTRIA RIGIDA.
Native of Pulo Pinang.
A shrub with corrugated grey bark. Leaves opposite, very
fihort-petioled, ovate lanceolate, acuminate, narrow at the base,
very entire, very smooth, pellucidly punctate. Stipules subulate,
acute. Racemes axillary, erect, rigid, branched, strict, shorter
than the leaves, pedicels short, rigid, many-flowered. Flowers
greenish, inconspicuous. Bracts very small. Calyx very small,
5-parted. Corolla a little longer than the calyx, 5-petaled. Sta-
mina 10, exsert, erect. Anthers incumbent. Ovarium superior,
2-celled, 2-seeded, seeds pendulous. Style thick, as long as the
stamina. Stigma capitate, obtuse. Berry of the size of a cherry,
snow-white, globular, and somewhat flattened, umbilicate, con-
sisting of a spongy farinaceous pulp, and containing a single large
round seed. Seed globose, attached superiorly and there umbi-
licate. Integument coriaceous, marked with veins which diverge
from the umbilicus. Albumen none. Embryo inverse, conform
to the seed. Cotyledons plano-convex, of a deep green colour,
somewhat rugose externally, and punctate on the inner surface.
Radicle superior, obverse to the umbilicus, short, straight, cylin-
drical, obtuse, covered with ferruginous down ; it is elongated into
a short conical plumule.
SECOND SERIES. VOL. II. T
2 74 DESCRIPTIONS OF MALA VAN PLANTS.
SPHALANTHUS. W. J.
Decandria Monogynia. N. O. Combretaceiz. BR.
Calyx tubolosus, hinc gibbus, deciduus, limbo 5-partito. Co-
rolla 5-petala, summo tubo calycis inserta et ejusdem laciniig
alterna. Stamina 10, corolla breviora. Stylus tubo calycis hinc
accretus. Ovarium uniloculare, ovulis paucis ab apice loculi
pendulis. Capsula 5-alata, monasperma, semine 5-angulato.
Semen exalbuminosum, cotyledonibus convexo-planis, radicula
minima conica.
SPHALANTHUS CONFERTUS.
Kayu sumang.
A shrub with round nearly smooth branches. Leaves generally
alternate, large and reflexly bifarious, short-petioled, ovate oblong,
acuminate, subcordate at the base, entire, very smooth. Petiole's
short, somewhat recurved. Stipules none. Spikes 1-3, terminal,
bending in an opposite direction from the leaves. Flowers
crowded, sessile. Bracts lanceolate acute, much shorter than the
flowers. Calyx superior, very long, tubular, gibbous on one side
below, reddish and somewhat tomentose without, limb 5-parted,
somewhat reflex, laciniae acute, broader at the base. Corolla
5-petaled, white at first, becoming red after expansion, a little
longer than the calyx, petals ovate oblong, acute. Stamina 10,
inserted in a double series on the calyx, erect, shorter than the
coralla ; anthers oblong, yellow. Ovary small, oblong, i-celled,
containing 3 pendulous ovula, attached by filaments to the
summit of the cell. Style green, filiform, rather longer than the
stamina, adhering to or concrete with the tube of the calyx on
one side along its whole length. Stigma simple. Capsule large,
not crowned with the calyx, oblong, with 5 membranaceous wings,
smooth, i-celled, i-seeded. Seed oblong, with 5 obtuse angles.
Integument membranaceous, easily separated. Albumen none.
Embryo conform to the seed. Cotyledons plano-convex, angled
exteriorly. Radicle conical, very small.
Obs. The structure of the seed is here different from what
generally obtains in the Combretacese, the cotyledons being solid,
not convolute.
PYRRHANTHUS. W. J.
Decandria Monogynia. N. O. Combretacea. BR.
Calyx 5-fidus, superus, persistens. Corolla 5-petala, calyce
longior. Stamina 5-10, erecta, corolla duplo longiora. Ovarium
DESCRIPTIONS OF MALA VAN PLANTS. 275
uniloculare, ovulis 3-5 pendulis. Drupa caryophylliformis, calyce
coronata ; nuce oblonga monosperma.
Arbor litorea inter Rhizophoras crescens ; foliis crassis ad apices
ramorum confertis, floribus subcorymbosis.
PYRRHANTHUS LITTOREUS.
Malay, Miri batu, and in Sumatra Kayu api-api.
Native of Sumatra and the Malay Peninsula, growing among
mangroves in salt swamps and near the mouths of rivers. It is
one of the most ornamental trees that occur in these situations.
It grows to be a large tree, generally with an irregular crooked
trunk. Leaves irregularly crowded at the extremities of the
branches, which are rough with their persistent vestiges, subsessile,
cuneiform, retuse, attenuated at the base into a very short petiole,
obtusely crenate, often nearly entire, smooth, thick and fleshy,
almost veinless. Stipules none. Racemes short, simple, terminal,
subcorymbose. Flowers pedicellate, crowded. Bracts 2, small,
acute, at the base of each flower. Calyx superior, 5-cleft, seg-
ments erect, thick, rather obtuse. Corolla crimson, 5-petaled,
petals spreading, twice as long as the. calyx, acute. Stamina
varying in number from 5 to 10, erect, twice as long as the corolla,
filaments red, subulate ; anthers oblong, purple, attached by the
middle. Ovary inferior, about the size and shape of a clove,
i -celled, containing from 3 to 5 ovula, which are pendulous from
the top of the cell. Style i. Berry or drupe somewhat com-
pressed, obtusely angled, crowned by the thick persistent calyx ;
nut oblong, with 2 prominent angles, i -seeded. Seed exalbumi-
nous. Embryo inverse. Cotyledons convolute.
Obs. The number of the stamina is very variable, 7 is perhaps
the most frequent ; 5 and 6 are common, but 10, the complete
number, is rare. The number of ovula varies also. The genus
is most nearly related to Laguncularia of Gaertner, but seems to
differ in its corolla and stamina. It has some resemblance to
Kada kandel (Rheed, "H. Mai." vi. p. 67, t 37), a figure which
has not, I believe, been quoted, and may possibly be another
species of this genus. Kayu api-api is the name generally given
to this tree in Sumatra, but is applied by Rumphius to his Man-
gium album ("H. A." iii. p. 115, t. 66), which is a species of
Avicennia, probably the A. resinifera of Forster, known in Sumatra
by the name of Pelandok kayu. It appears to be distinct from
A. tomentosa, having lanceolate acute leaves, white beneath, but
not tomentose, and the fruit being much smaller.
T 2
.276 DESCRIPTIONS OF MALAYAN PLANTS.
PHALERIA. W. J.
Octandria Monogynia.
Perianthium coloratura, tubulosum, inferum, limbo 4-partito.
Stamina 8, exserta. Ovarium biloculare, 2-sporum, ovulis pendulis.
Stigma capitatum. Bacca bilocularis, disperma. Semina exalbu-
minosa, embryone inverse.
Frutex, foliis suboppositis, floribus axillaribus.
This genus is related to the Thymelaese, but differs in having a
"bilocular ovary and fruit.
PHALERIA CAPITATA. W. J.
Native of Sumatra.
A shrub with smooth branches. Leaves opposite, or suboppo-
site, short petioled, ovate lanceolate, terminated by a long, sharp
acumen, entire, very smooth, 8 inches long. Petioles thickened.
Stipules none. Peduncles axillary, sometimes from the axils of
fallen leaves, very short, bearing a head or umbel of sessile flowers
which is embraced by an involucre composed of several oblong-
ovate leaflets or bracts. Flowers large and white, resembling
those of the jasmine. Perianth inferior, tube long, faux pervious,
smooth, limb 4-parted, segments ovate. Stamina 8, inserted on
the faux, exsert, rather long; anthers 2-lobed. Ovary embraced
by a thin white nectarial cup, oblong, attenuated into a style,
2-celled, cells monosporous, ovula attached to the summit of the
cell by a thread, which, passing along the back of the ovulum, is
inserted into its base, so that the ovulum seems as if doubled
upon its filament. Style a little shorter than the stamina. Stigma
capitate, papilose. Berries crowded, somewhat pear-shaped,
rounded above, acute at the base, cortical, 2-celled, 2-seeded.
Seed exalbuminous, embryo inverse, cotyledons plano-convex;
radicle small, superior.
PTERNANDRA. W. J.
Octandria Monogynia.
Calyx ovatus, limbo quadridentato. Corolla 4-petala. Stamina
octo, antheris introflexis, compressis, basi postice calcaratis, bilo-
cularibus, loculis longitudinaliter dehiscentibus. Ovarium calyci
infra adnatum, 4-loculare, polysporum, placentis parietalibus.
Stylus declinatus. Bacca polysperma.
Habitus Melastomarum, foliis oppositis trinerviis, floribus
paniculatis.
DESCRIPTIONS OF MALA VAN PLANTS. 277
PTERNANDRA CGERULESCENS.
Native of Pulo Pinang.
A large smooth shrub, with round branches. Leaves opposite,
short petioled or subsessile, ovate, acuminate, tapering at the base
into short petioles, very entire, very smooth ; coriaceous, paler
beneath, with 3 strong nerves, and 2 less conspicuous along the
margins ; the transverse veins are few and not prominent. Sti-
pules none, but the petioles are connected by an interpetiolar
line. Panicles oppositely corymbose, short, terminal, sometimes
also from the upper axils. Peduncles 4-sided, smooth. Bracts
small. Calyx united to the ovarium beneath, ovate, reticulately
squamous, almost entire or obsoletely 4-toothed. Corolla blue,
lighter at the margin, 4-petaled, petals ovate, acuminate, inserted
into the calyx. Stamina 8, blue; filaments nearly erect, incurved
at the apex. Anthers large, pointing inwards, compressed,
elongated behind into an acumen or spur, cells anteriorly gibbous
and bursting longitudinally. The anthers, before expansion, are
turned downwards, as in the Melastomse, but their points do not
reach much below the top of the ovary. Style declinate, about as
long as the stamina. Stigma conical and rather obtuse. Ovary
adnate to the calyx, 4-celled, polysporous, ovula attached to
convex parietal placentae. Berry 4-celled, many-seeded.
Obs. In general habit and appearance this plant has a close
resemblance to my Melastoma glauca, and at first sight appears
only to differ in having smaller flowers, and leaves with less
distinct nerves and veins. In the structure of the anthers, how-
ever, it differs essentially from Melastoma, and has some affinity
to Memecylon ; the fruit and mode of placentation differs from
both. The ovary might either be considered inferior, or superior
and adnate to the calyx; the analogy of Melastoma has led me to
assume the latter.
MEMECYLON PANICULATUM. W. J.
Foliis petiolatis ovatis obtuso-acuminatis, paniculis axillaribus
brachiatis.
Found at Tappanuly and on Pulo Bintangor, on the West Coast
of Sumatra.
A large shrub, with grey bark and smooth branches. _ Leaves
opposite, short-petioled, ovate or oblong ovate, terminating in a
rather obtuse acumen, entire, very smooth, shining above, paler
beneath, with pretty distinct nerves, which unite into a line near
the margin ; 7 or 8 inches long. Petioles short and thick. Sti-
pules none. Panicles axillary, sometimes from the axils of fallen
leaves, oppositely branched. Peduncles 4-sided, purplish ; there
278 DESCRIPTIONS OF MALA VAN PLANTS.
is generally a single i -flowered pedicel placed immediately below
each of the principal divisions of the panicle, springing as it were
from the same point. Flowers numerous, bluish. Bracts minute.
Calyx nearly entire. Corolla light blue, 4-petaled, petals broad,
acute. Stamina 8; filaments subulate; anthers blue, prolonged
behind into a thick spur, the upper surface of which is marked
with a nectariferous cavity ; cells on the anterior surface perpen-
dicular to the spur, which is nearly horizontal, bursting longitudi-
nally. Ovary i-celled, containing about 8 erect ovula attached
to a small protuberance in the base of the cell, its disc marked
with radii corresponding to the faces of the anthers, which are
incurved before expansion. Style subulate. Stigma acute. Berry
globular, i-seeded. Seed erect, exalbuminous. Cotyledons pel-
tate, hemispherical, their flat surfaces a little irregular or waved.
Radicle erect, rising perpendicularly between the cotyledons to
their centre, where it is inserted.
Obs. This peculiar structure of the embryo is different from
what obtains in all the other species of Memecylon that I have
examined, where the cotyledons in place of being solid and hemi-
spherical, are foliaceous and contortuplicate.
OCTAS. W. J.
Octandria Monogynia.
Calyx 8-partitus. Corolla 8-loba. Stamina octo, laciniis
corollas alterna. Stigma sessile, 8-radiatum. Bacca 8-sperma,
supera.
Frutex, foliis simplicibus alternis, spicis axillaribus.
OCTAS SPICATA.
Found at Tappanuly on the West Coast of Sumatra.
A shrub, with long branches, the young parts tomentose.
Leaves alternate, petiolate, lanceolate oblong, acuminate, entire,
smooth, 5 inches long. Stipules, small, acute. Spikes or
racemes 2 from each axil, rather shorter than the leaves, many-
flowered; pedicles in threes. Flowers small, white. Bracts
minute. Calyx small, 8-parted. Corolla monopetalous, spread-
ing, divided at the margin into 8 round lobes. Stamina 8,
as long as the lobes of the corolla ; anthers white, subsaggitate.
Ovary superior, globular, 8-celled, 8-seeded. Stigma large,
sessile, composed of 8 fleshy coadunate lobes. Berries about
he size of peppercorns, purple, containing 8 seeds, which are
ngled interiorly.
DESCRIPTIONS OF MALAYAN PLANTS. 279
CCELOPYRUM. W. J.
Octandria Monogynia.
Calyx 4-partitus. Corolla 4-petala. Stamina 8, alterna bre-
viora. Stigma obtusum subsessile. Drupa supera, nuce biloculari,
loculo exteriore lunato alterum fovente. Semen unicum, loculo
altero vacuo.
Arbor, ramis apice foliosis, foliis simplicibus, floribus racemosis.
CCELOPYRUM CORIACEUM.
T&gntang. Malay.
In forests in the neighbourhood of Bencoolen.
A tree with thick branches, which are foliose at their summit.
Leaves alternate, petiolate, elliptic, obtuse or emarginate, entire
with reflexed margins, firm and leathery, smooth above, pale and
tomentose beneath, costate with strong parallel ribs or nerves, 10
-12 inches long. Petioles about 3 inches long, marginate and
flattened above. Racemes axillary, erect, shorter than the leaves,
branched ; flowers numerous, yellowish, small and inconspicuous,
in small racemules or spikelets. Bracts small, acute. Calyx
inferior, spreading. Corolla 4-petaled, petals longer than the
calyx, ovate. Stamina 8, the alternate ones shorter. Ovary
surrounded and nearly immersed in a large fleshy nectarial ring,
whose sides are angled by the compression of the filaments.
Style scarce any. Stigma obtuse. Drupe ovate, acute, smaller
than an olive, containing a single nut. Nut 2 -celled, cells unequal
and dissimilar, the outer and lower crescent-shaped, and embrac-
ing the other, which is smaller, oblong and always empty; the
larger cell contains a single conform seed.
Obs. The structure of the fruit is very peculiar ; the empty
cell is placed obliquely in the upper part of the nut, the fertile one
is, as it were, wrapped round the other. The extreme minuteness
of the ovary prevented me from satisfactorily ascertaining its
.structure.
PETROCARYA EXCELSA. W. J.
Heptandria Monogynia. N. O. Rosacecz. Juss.
Foliis oblongis acuminatis glabris, calycibus ore obliquis,
staminibus undecim fertilibus.
Kayu balam pangkat. Malay.
A large timber tree. Leaves alternate, short, petioled oblong,
acuminate, entire, smooth 4-5 inches long. Stipules longer than
28o DESCRIPTIONS OF MALAYAN PLANTS.
the petioles, deciduous. Racemes axillary and terminal, forming
a panicle towards the top, strict, erect, little branched; flowers
very short, pedicelled and appressed to the principal peduncle ;
the whole ferruginous and tomentose. Bracts broad, deciduous.
Calyx infundibular, ferruginous and tomentose, oblique at the
mouth, furnished with a ring of stiff hairs which point downwards,,
lowest on the side to which the fertile stamina and ovary are
attached, limb 5 -parted, subreflex. Corolla 5-petaled, inserted on
the mouth of the calyx and scarcely longer than its limb, petals
subrotund. Stamina n, fertile, twice as long as the petals,
inserted in one phalanx along the lower edge of the mouth of the
calyx, on the upper edge is a ring with 8 processes or abortive
stamina. Ovary adnate to the side of the calyx below the fertile
stamina, densely pilose, disporous. Style lateral, inserted near
the base of the ovary, as long as the stamina. Stigma simple.
Drupe enclosed in the enlarged calyx, which becomes adnate to it
and crowned by its persistent limb ; obliquely ovate, about the
size of a filbert. Nut smooth, i-seeded, with an abortive cell
generally above the fertile one. Seed curved, corresponding ta
the cell, albuminous ; embryo cylindrical inverse ; radicle superior,
clavato-cylindrical, longer than the ligulate cotyledons.
PETROCARYA SUMATRANA. W. J.
Foliis elliptico-oblongis subtus canescentibus, calycis ore
regulari, staminibus septem fertilibus.
A tree. Branchlets pilose. Leaves alternate, short petioled,
elliptic-oblong, 6-8 inches long, terminating in a bluntish acumen,
acute at the base, entire, the adult leaves smooth above, some-
what hoary, with close short wool beneath, the younger ones
covered with deciduous pubescence above, nerves prominent
beneath, veins reticulate. Petioles about a quarter of an inch in
length, stipules longer than the petioles, oblong, acute. Racemes
axillary and terminal, shorter than the leaves, tomentose ; pedicels
mostly 3-flowered, divaricate. Bracts rather large, concave, at
the base of the peduncles, pedicels and flowers. Calyx tubular or
campanulate, tomentose without, pilose at the faux, which is equal
and regular, limb spreading, 5-parted, segments acute. Corolla
5-petaled, white, petals inserted on the mouth of the calyx, and as
long as its segments. Stamina 14, of which 7 upper are fertile
arranged in one phalanx, and the opposite 7 abortive; fila-
ments short, flat, anthers roundish, 2-lobed. Ovary adnate ta
the upper side of the tube or calyx, pilose, 2-celled, containing z
erect ovula. Style lateral, inserted at the base of the ovary, as
long as the stamina. Stigma capitate.
Obs. These two species, though nearly related, present abun-
dant points of distinction. In the P. excelsa the leaves are
DESCRIPTIONS OF MALA VAN IV. ANTS. 28 1
smaller, smoother, and less strongly nerved, while the flowers are
larger, the racemes longer, more erect and compact, and the
stamina longer and more numerous than in the P. Sumatrana.
WORMIA EXCELSA. W. J.
N. O.
Fpliis ellipticis acutis denticulatis, pedunculis multifloris opposi-
tifoliis, pedicellis clavatis.
Kayu sipur. Malay.
In forests near Bencoolen.
A large tree. Leaves alternate, petiolate, from elliptic ovate to
elliptic-oblong, acute, denticulate or obsoletely serrate, smooth,
8-12 inches long. Petioles deeply channelled above. Peduncles
oppositifolious at the summit of the branches, many-flowered;
pedicels alternate, clavate. Flowers large, yellow, 3 inches in
diameter. Calyx 5-leaved, leaflets subrotund, concave, unequal,
Corolla 5-petaled, spreading, petals ovate oblong. Stamina very
numerous, the outer ones yellow, spreading, shorter than the inner,
which are purple, erect and recurved above ; anthers, lobes adnate
to the filament. Ovaries 6-8, connate, polysporous. Stigmas as
many, flat, recurved, diverging. Capsules 6-8, whitish, semi-
transparent, bursting at the inner angle, and then spreading,
containing no pulp. Seeds attached to the edges of the capsules,
enveloped in a red aril.
Obs. This is a large forest tree, which yields excellent timber,
the wood having some resemblance to oak.
WORMIA PULCHELLA. W. J.
Foliis obovatis integerrimis, pedunculis solitariis axillaribus
unifloris, floribus pentagynis.
Found at Natal.
A small tree. Branches round, rather smooth. Leaves alternate,,
petiolate, oblong obovate, rounded at top, with a short round point,
sometimes retuse, very entire, very smooth, thick and rather
coriaceous, about 5 inches long. Petioles smooth, channelled
and marginate above, less than an inch in length. Peduncles
axillary and sub terminal, solitary, i -flowered, angled, about 2
inches long. Bracts none. Calyx 5-leaved, leaflets subrotund,
smooth. Corolla 5-petaled. Stamina numerous. Ovaries 5,
collected into a globe, terminating in as many flat, reflexed
diverging styles. Stigmas thickened. Capsules 5, of a light
semi-transparent rose-colour, bursting at their angles, and then
spreading like a corolla. Seeds attached to the inner edges of the
282 DESCRIPTIONS OF MALA VAN PLANTS.
capsules, a few only coming to perfection, partly embraced by a
red pulpy aril, which originates from the umbilicus.
Obs. This species is very beautiful when in fruit, from the
delicacy of the colours which the capsules exhibit.
FICUS OVOIDEA. W. J.
Foliis cuneato-obovatis apice rotundatis, nervo medio dichotomo,
fructibus axillaribus binis pedunculatis.
Found at Singapore, and on several parts of the West Coast of
Sumatra and its islands.
A small tree, with smooth brownish bark. Leaves alternate,
petiolate, cuneato-obovate, rounded above, attenuated to the base,
very entire, very smooth, the middle nerve dichotomous ; from i j
to 2 inches long. Petioles nearly half an inch long, round, with a
slight furrow above, and covered with grey bark like the branchlets.
Peduncles in pairs, sometimes solitary, axillary, shorter than the
petioles, i-flowered. Involucres embraced at the base by 3 short
subrotund bracts, nearly globose, smooth, shut at the mouth by
scales, and containing numerous pedicellate florets. Seeds naked,
hard.
Obs. The leaves are peculiar in having the middle nerve
dichotomous, a character by which this species may be readily
distinguished from its congeners.
FICUS DELTOIDEA. W. J.
Foliis obcuneato-deltoideis apice latis v. retusis, nervo medio
dichotomo, fructibus axillaribus binis pedunculatis.
A small tree, native of Sumatra, and very similar to the
preceding, but having the leaves proportionally broader, more
decidedly deltoid, and retuse or truncate, not rounded at top;
the peduncles also are in pairs from the axils of the leaves and
longer than the petioles. The breadth of the leaves is generally
greater than their length in this species, which is not the case
with their preceding ; they are, however, precisely similar in their
leathery texture, and in having the nerve dichotomous and not
prominent.
FICUS RIGIDA. W. J.
Foliis ovatis lineari-acuminatis rigidis, fructibus pedunculatis
axillaribus globosis glabris.
Seribulan. Malay.
Sumatra, &c.
A tree, with grey cinereous bark and smooth branchlets.
DESCRIPTIONS OF MALA VAN PLANTS. 283
Leaves alternate, petiolate, ovate or obovate, with long linear
acumina which are obtuse or emarginate at the point, attenuated
to the base, 3-4 inches long, entire, firm and rigid, smooth,
shining above, rugose, with reticulate veins beneath ; nerves pro-
minent beneath, the lowermost pair springing from the base and
running along the margins until they anastomose with the upper
ones. Petioles brown with cracked skin. Berries 1-3, axillary,
pedicelled, pedicels shorter than the petioles, smooth. Involucra
globose, orange-coloured when ripe, smooth, with some whitish
spots as large as a currant. Florets numerous, pedicellate. Female
ones with a 4~5-parted perianth. Style inserted laterally. Seed
naked.
Obs. The bark of this species is fibrous, and I am informed
that it is employed in Menangkabau in the fabrication of a coarse
kind of paper.
JONESIA. ROXB.
N. O. Leguminosa.
Calyx tubolosus, basi bibracteatus, limbo 4-lobo. Petalla nulla.
Stamina 3-7, summo tubo calycis inserta. Ovarium pedicellatum,
pedicello calyci hinc accreto. Legumen oligospermum.
Frutices, foliis abrupte pinnatis, floribus fasciculatis.
The alteration I have here made in the terms of the generic
description from that given by Roxburgh will remove all obscurity
as to the true affinities of this genus, and establish its near relation
to Macrolobium. The bracteal leaflets (the diphyllous calyx of
Roxburgh) are found in both genera, though less conspicuous and
not coloured in Macrolobium, the stamina are similarly inserted
on the mouth of the tubular calyx, and are equally variable in
number ; the pedicel of the ovary is accrete to the calyx in both,
and the only difference consists in the presence or absence of the
single petal which is found in Macrolobium and is wanting in
Jonesia.
JONESIA DECLINATA. W. J.
Foliis 6-8 jugis, foliolis oblongis, floribus fasciculato-paniculatis
tetrandris.
Kayu siturun. Malay.
A small straggling tree found generally in thickets ; native of
Sumatra.
Branches depending, whence the native name. Leaves alternate,
composed of from 6 to 8 pair of leaflets, of which the lowest are
situated on the base of the petiole; they are opposite, from 10 to
12 inches in length, oblong, rounded at the extremity, but termi-
nating in a short thick recurved point, entire on the margin,
Smooth. Petiole roundish, thickened at the base. Stipule intra-
284 DESCRIPTIONS OF MALAYAN PLANTS.
petiolar, embracing the stem, broad at the base, ovate and pointed.
Flowers in lateral fasciculate panicles, 2 subrotund bracts below
each flower. Pedicels slender, the whole very smooth and deli-
cate, and of a light semi-transparent red colour. Calyx reddish
yellow, tubular tube narrow ; limb 4-parted, flat, segments sub-
rotund, about the same size as the bracts. Corolla none. Stamina
4, more than twice the length of the calyx and inserted on its
tube, their upper part deep red. Anthers deep purple, subrotund,
2-celled, each cell streaked with white. There are no rudiments
of abortive stamina. Germen pedicellate, pedicel accrete to the
tube of the calyx. Style long, red. Stigma round. Legume
pedicellate, flat, compressed, containing several seeds.
The large branches of delicate flesh-coloured flowers render
this a very beautiful shrub during the period of infloresence.
BAUHINIA EMARGINATA. W. J.
Foliis cordatis subrotundo-ovalibus glaberrimis acumine brevi
obtuso emarginato, floribus octandris, staminibus tribus superiori-
bus fertilibus.
Dadaub. Malay.
Native of Sumatra.
A strong woody climber. Lea.ves alternate, petiolate, cordate,
subrotund oval, terminating in a short blunt emarginate acumen,
very entire, 4 inches long, 7-9 nerved with reticulate veins, very
smooth. Petioles rather short. Cirrhi long, simple, revolute.
Racemes terminal or sometimes lateral, corymbose, many flowered ;
pedicels long, tomentose. Calyx 5-parted, tomentose, bursting
into 2 or 3 segments. Corolla large, 5-petaled, spreading, petals
nearly equal, unguiculate. Stamina 8 ; 3 superior fertile, longer,
with large 2-lobed anthers ; 4 inferior short, with small abortive
anthers; the fifth and lowest being a little longer, and entirely
sterile. Ovary tomentose. Style about the length of the fertile
stamina. Stigma peltate, round.
Obs. The form of the leaf is very peculiar, and readily dis-
tinguishes this species from the others.
BAUHINIA BIDENTATA. W. J.
Foliis cordatis acuminatis apice bidentatis glaberrimis, corymbis
terminalibus, floribus octandris, staminibus tribus superioribus
fertilibus.
Native of the Malayan forests, where it climbs over trees, and
shows its flame-coloured blossoms on their very summits.
Shrubby, climbing far over the trees in its neighbourhood;
bark brown ; branches round, flexuose ; branchlets covered with
ferruginous tomentum. Leaves alternate, petiolate, cordate, acute,
DESCRIPTIONS OF MALA VAN PLANTS. 985
bifid at the point (not 2-lobed), divisions approximate, with a
short thread interposed, very entire, 7-nerved, very smooth, the
younger ones rather silky beneath with ferruginous deciduous
hairs. Petioles thickened at the top and base. Tendrils simple,
revolute. Corymbs terminal. Pedicels clavate, striated, tomen-
tose. Calyx 5-parted, tomentose, for the most part bursting
irregularly into three divisions. Corolla orange-coloured, becom-
ing red after expansion, 5-petaled, petals nearly equal, subrotund,
unguiculate, spreading. Stamina 8, ascending, of which the .3
upper are longer and fertile, and the 3 lowest short and sterile.
Anthers subrotund. Ovary pedicellate, compressed, oblong, con-
taining from 6-8 ovula. Style declinate, incurved at the point.
Stigma large, capitate and glutinous.
Obs. This species is at once distinguished by the peculiar
form of the leaves, which are not 2-lobed as usual in the genus,
but have the apex divided so as to make the leaf terminate in 2
acute points. The flowers are large and showy.
INGA BUBALINA. W. J.
N. O. Mimosea. BR.
Inermis, foliis conjugato-pinnatis, foliolis bijugis glaberrimis,
capitulis paucifloris paniculatis, paniculis axillaribus et terminali-
bus, legumine recto cylindrico.
Bua karbau. Malay.
Sumatra, &c.
A tree, unarmed, with grey bark. Leaves alternate, conjugate
pinnate, leaflets 2-paired, ovate, with rather an obtuse acumen,
very entire, very smooth, nerves lucid, the upper pair of leaflets
the largest. Primary petiole short, thickened at the base, bearing
a gland at the point, secondary petioles without glands. Capitula
few-flowered, panicled. Panicles axillary and terminal, peduncled,
divaricate, shorter than the leaves. Bracts small. Calyx short,
tubular, 5-dentate. Corolla white, much longer than the calyx,
campanulate, 5-parted, segments spreading. Stamina many,
monadelphous at the base, long and white. Style filiform, as long
as the stamina. Ovary pedicellate. Legume dark green, straight,
cylindrical, about 4 inches long, thick, obtuse, many-seeded, fetid.
Seeds crowded, orbicular, piled one above the other and thus
flattened above and below by their mutual compression.
Obs. This species is nearly allied in habit and inflorescence to
the Inga Jiringa (" Mai. Misc." vol. i.), but differs in the shape of
the legume, which has a very offensive smell, but is eaten by
the natives in the same manner as that of the Petek (Acacia
graveolens, W. J.). Karbau in Malay signifies the Buffalo,
whence the specific name.
286 DESCRIPTIONS OF MALA YAN PLANTS.
INGA CLYPEARI^ W. J.
Inermis, ramulis acutangulis, foliis bipinnattis, foliolis lo-jugis
rhomboideis subtus tomentosis, paniculis terminalibus, leguminibus
contortis rubris.
Clypearia rubra. Rumph. "Arab." iii. p. 176, t. 112.
Jering munet. Malay.
A large tree. Branchlets smooth, acutely 5-angled, almost
winged. Leaves alternate, bipinnate ; pinnae about 4 pair ; leaf-
lets about 10 pair, rhomboidal, inequilateral, rather acute, entire,
smooth above, tomentose or silky and glaucous beneath, they are
of unequal size, the uppermost often 2 inches long. Petiole or
rachis acutely 4 or 5 angled, thickened at the base, eglandular.
Panicles large, terminal; peduncles fascicled. Flowers white,
pedicellate, in small capitula or heads. Calyx small, 5-parted.
Corolla much longer than the calyx, quinquefid. Stamina
numerous, monadelphous at the base. Style one. Legume red,
flat, 2-valved, spirally contorted, containing many subrotund
somewhat compressed black seeds.
Obs. This species, which agrees with that described by Rum-
phius, is found in forests in the neighbourhood of Bencoolen, but
I am not aware that it is there put to any particular use. These
two species, together with the I. Jiringa, might perhaps with equal
propriety be refered to Acacia, as the seeds are not arilled, though
the legume (as in I. bubalina) is fleshy and esculent ; the stamina
are those of Inga and the paniculate inflorescence is more frequent
in that genus than in Acacia. The distinction between these two
sections of the Linnean genus Mimosa is an artificial one, and
the characters of the present species are in some degree inter-
mediate between the two.
TABERN^MONTANA MACROCARPA. W. J.
Foliis ovato-ellipticis basi attenuatis, corymbis terminalibus
dichotomis, folliculis maximis subglobosis.
In the interior of Bencoolen.
A tree, branches smooth, somewhat compressed in contrary
directions between each pair of leaves. Leaves opposite, petiolate,
from elliptic-ovate to elliptic-lanceolate, tapering to the base,
broader above with a short point, very entire, very smooth ;
nerves transverse, uniting into submarginal arches; 10-12 inches
long. Petioles embracing the stem and uniting with the base of
the opposite one. Peduncles 3-4, terminal, dividing at their
summits into dichotomous corymbs. Flowers rather large, yel-
lowish. Calyx 5-cleft, erect, thick. Corolla much longer than
the calyx ; tube gibbous, almost globose at the base, narrowing
DESCRIPTIONS OF MALA VAN PLANTS. 287
upwards; limb rotate, 5-parted; segments oblong, oblique.
Stamina 5, within the tube. Ovary double. Styles 2, shorter
than the stamina. Stigma small. Follicles 2, baccate, as large
as citrons, red, diverging, subglobose, exuding a milky juice when
cut, with a ridge along the middle and one at each side which
unite in a short blunt point, i-celled, many-seeded ; the cell is
recurved into the form of a crescent. Seeds contained in red
fleshy arils or lobules which are angled by mutual compression,
oblong, chrysaloid, hollowed on the one side with incurved
rounded edges, convex on the other and longitudinally corrugated.
Embryo contained in a conform albumen ; cotyledons flat, round,
cordate; radicle centripetal, cylindrical, longer than the coty-
ledons.
FAGRCEA CARNOSA. W. J.
Foliis subrotundo-ovatis mucronatis caraosis, floribus terminali-
bus solitariis.
In the neighbourhood of Bencoolen.
A parasitic shrub growing on trees, with smooth greyish bark
and somewhat dichotomous branches. Leaves opposite, petiolate,
subrotund with a short reflexed point, entire with reflexed
margins, very smooth, thick and fleshy. Petioles compressed,
embracing the branch and furnished with an intrapetiolar ligula or
stipule. Flowers terminal, solitary, nearly sessile, embraced at
the base by a few sheathing bracts. Calyx 5-parted. Corolla of
a dull yellowish white colour ; tube about 4 inches long, expand-
ing into a 5-parted limb. Stamina 5, rising a little above the tube ;
anthers large. Style little more than half the length of the tube.
Stigma 4-lobed. Berry as large as a small egg, seated on the
persistent calyx, ovate, rather pointed, 2-celled, many-seeded;
seeds nidulant.
Obs. This is the fifth species of Fagrcea that I have met with
in the Malay Islands ; the others have been already described in
Roxburgh's "Flora Indica." The F. racemosa grows to be a
small tree, and the F. volubilis, doubtfully proposed by Dr.
Wallich as a distinct species, is the same plant. The F. auricu-
lata is a large shrub, and from the size of its flowers is the most
splendid of the genus. I originally met with it at Singapore, but
have since found it also at Tappanuly. The following particulars
may be added to the description given by Dr. Wallich.
F. Auriculata. Flowers terminal, generally 3, rarely 5, on short
thick pedicels, each embraced by 4 opposite calyculate bracts, of
which the outer 2 are the smallest. Corolla very large, yellowish-
white. Stamina inserted near the bottom of the tube. Stigma
large and flattened. Ovary 2-celled, polysporous ; the edges of
the placentae revolute. Fruit as large as a duck's egg, acuminated
by part of the persistent style ; seeds numerous, nidulant.
288 DESCRIPTIONS OF MALA VAN PLANTS.
IXORA NERIIFOLIA. W. J.
Foliis linearibus acuminatis glabris, corymbis terminalibus.
Bunga Saluang. Malay.
Native of the West Coast of Sumatra.
A shrub, with round smooth branches. Leaves opposite, short
petioled, linear, tapering to the point, acute, about 9 inches long,
by little more than half an inch broad, entire, with revolute edges,
very smooth. Stipules interpetiolar, subulate, longer than the
petioles. Corymbs terminal, erect, trichotomous. Flowers red.
Bracts small, acute. Calyx small, 4-toothed. Corolla tube long,
slender; limb spreading, 4-parted, segments lanceolate, acute.
Stamina 4, alternate with the laciniae of the corolla. Style a little
longer than the tube. Stigma clavate. Fruit a berry.
Obs. The long narrow leaves readily distinguish this species ;
it is a handsome, delicate shrub.
LECANANTHUS. W. J.
Pentandria Monogynia. N. O. Rubiacea. Juss.
Calyx campanulatus, ampliatus, coloratus, irregulariter divisus.
Corolla tubo brevi, limbo 5-partito. Ovarium biloculare, poly-
sporum, placentis centralibus convexis. Stylus bifidus. Stigmata
2, linearia crassa. Fruticosa, floribus capitatis involucratis termi-
nalibus, sestivatione valvata.
LECANANTHUS ERUBESCENS.
Found in the interior of Bencoolen.
A small erect shrub ; stem 4-sided, 2 of the angles acute.
Leaves opposite, short-petioled, ovate lanceolate, acute at both
ends, rather attenuated to the point, entire, smooth; about 8
inches long. Stipules interpetiolar, large, ligulate, carinate towards
the base. Flowers pale red, densely aggregated within the hypo-
crateriform cup of the involucre, forming a head which is terminal,
nearly sessile, and turned backwards. Involucre monophyllous,
entire. Pedicels none. Calyx superior, coloured, tomentose,
thick and fleshy, much wider than the corol, expanding into from
2 to 4 irregular unequal obtuse lobes ; the calyces of the outer
flowers are often so much produced on one side as to seem bila-
biate. Corolla, tube short, segments 5, acute, thick. ^Estivation
valvate. Stamina 5, inserted on the tube ; anthers large. Ovary
crowned with a prominent nectarial ring, 2-celled, polysporous ;
ovula arranged round central, semi-cylindrical placentae. Style
bifid. Stigmata 2, thick and linear.
DESCRIPTIONS OF MALA VAN PLANTS. 289
PSILOB1UM. W. J.
Pentandria Monogynia. IV. O. Rubiaceic. Juss.
Calyx patens, 5-partitus. Corolla tubo brevi, limbo 5-partito.
Stamina basi corollae inserta. Stigma clavatum, lo-alatum,
turn. Fructus cylindricus siliquae formis, foliolis calycinis
tentibus coronatus, bilocularis, polyspermus. Semina duplici serie
axi affixa.
Fruticosa, pedunculis axillaribus paucilloris, zestivatione valvatu.
PSILOBIUM NUTANS.
Found in the interior of Bencoolen.
Stem erect, 4-sided, with rounded angles. Leaves opposite,
petiolate, lanceolate, attenuated to both ends, acute, entire, smooth.
Stipules interpetiolar, broad, acuminate, carinate. Peduncles
axillary, drooping, bearing from 3 to 6 flowers. Bracts forming u
kind of involucre at the base of the very short pedicels. Calyx
superior, very large, composed of 5 leaflets or very deep segments,
which are veined with red. Staminas. Filaments short ; anthers
long, erect. Style short. Stigma long, exsert, oblong-ovate,
longitudinally ic-winged, the 5 alternate wings smaller. Fruit
long, cylindrical, siliquose, crowned with the large persistent calyx,
2-celled, many-seeded ; seeds arranged in a double series in each
cell.
OPHIORRHIZA HETEROPHYLLA. W. J.
Foliis oppositis subrotundo-ovatis, altero nano.
Found in the interior of Bencoolen.
This species is readily distinguished by the peculiarity of one
of the opposite leaves being always dwarf or abortive ; the other
is subrotund-ovate, with a bluntish acumen, smooth, pale and
whitish beneath. The stem is erect and tomentose. Flowers
in a small terminal cyrne. Capsule compressed, obcordate.
QUERCUS RACEMOSA. W. J.
Foliis lato-lanceolatis integerrimis glaberrimis, spicis masculis
paniculatis, fructibus spicatis nuce umbilicato-depressa, calice
fructus tuberculato.
Punning-punning bungkus. Malay.
Native of Sumatra.
A large tree, with brownish bark. Branches smooth. Leaves
alternate, short petioled, ovate-lanceolate, acuminate, attenuated
to the petiole, very entire, very smooth, nerves well-marked and
SECOND SERIES. VOL. II. U
2 9 o DESCRIPTIONS OF MALAYAN PLANTS.
reddish beneath ; 6-8 inches long. Stipules small, linear. Male
spikes numerous, panicled, terminal, and from the axils of the
upper leaves which are crowded round the thickened extremity of
the branch, slender, hoary. Flowers sessile, aggregated. Female
spikes at first terminal, becoming afterwards lateral by the shooting
up of the branch. Flowers numerous, dense, sessile. Male.
Calyx 6-parted, segments acute. Stamina 15-20. The centre of
the flower is occupied by a densely villous disc. Female. Calyx
rugose, turbinate, umbilicate. Ovary 3-5-celled, each cell con-
taining 2 ovula attached by a thread to its summit. Acorns large,
depressed, umbilicate, with a short mucro. Cup flat, embracing
the nut for about half its height, nearly an inch in diameter, rough
with angular imbricated tubercles, which are large towards the base,
and become small towards the edge.
Obs. This is a very splendid species, from the great size of the
racemes and acorns. Punning-punning is the generic appellation
of the oaks in Malay; in the .Rejang dialect they are called
Pasang.
QUERCUS URCEOLARIS. W. J.
Foliis elliptico-oblongis acumine gracili integerrimis glaberrimis,
fructibus spicatis, calyce fructus subhemisphserico limbo patente.
Native of Sumatra.
A tree, with rough bark. Leaves alternate, petiolate, elliptic-
oblong, terminated by a long slender acumen, very entire, smooth,
coriaceous, pale beneath ; 8-9 inches long. Fruit on lateral
racemes. Acorns rounded and flattened at top, umbilicate in the
centre and mucronate with the 3 short persistent styles, rather
perpendicular at the sides, half embraced by the calyx, which is
cup-shaped, marked on the outer surface with small acute scaly
points concentrically arranged, and whose margin expands into a
spreading, nearly entire, waved limb. The ovary is 3-celled, each
cell containing 2 ovula, and is lodged in the bottom of the large
funnel-shaped calyx. The acorn contains a single exalbuminous
seed placed in a little obliquely.
Obs. The spreading limb of the cups forms a good distinctive
character, and renders this a very remarkable and curious species.
ARECA TIGILLARIA. W. J.
Frondibus pinnatis, foliolis acutis, spadicibus ramosis, flore
unico femineo inter duos masculos, fructibus globosis.
Nibong. Malay.
Abundant in Sumatra and the Malay Islands, where it is much
used in the construction of houses, &c.
DESCRJPTIONS OF MALA VAN PLANTS. 291
Trunk erect, generally thicker than that of the common Pinang
(Areca catechu), armed, particularly on tl.
straight, slender, flattened spines. Fronds pinnate linear,
acuminate, reflexed at the edges so as to make the upper surface
convex, smooth, with a few brownish scales on the middle nerve
of the younger ones ; they diminish in size to the top of the frond
and the last 2 are partly united at their base. Stipes of the frond
scaly while young, compressed, grooved above, the sheaths armed
like the trunk. Spadix within the sheath of the frond, embracing
the stem, flattened at the base, much branched ; flower bearing
branchlets about 2 feet long, drooping, the lower ones 3-4 together,
the uppermost solitary or in pairs. Spathe single, completely
enclosing the spadix before expansion, compressed, 2-edged, deci-
duous, partial spathes none. Flowers sessile, i female between 2
males, the latter considerably the largest and deciduous. Male.
Hermaphrodite. Perianth 6-parted, the outer leaflets small, the
inner much longer, and acuminated with fine points. Stamina 6.
Anthers sagitate. Ovary small, surmounted by 3 linear styles.
Female. Perianth 6-parted; leaflets nearly equal, rounder and
shorter than those of the male. Stamina none. Ovary monospo-
rous. Styles none. Stigmata 3. Fruit globose, about the size
of a carbine bullet, of a deep purple colour when ripe, with a
glaucous tint, containing under a reddish pulp a single smooth
globular nut. Nut i -seeded, having a thickened whitish scar on
the side, and a small areola at the base opposite to the embryo.
Seed solid ; albumen ruminated. Embryo basilar, short, cylin-
drical, obtuse.
Obs. This differs from the common Areca in the disposition
of the flowers on the spadices, and in having the nut contained
under a pulpy and not a fibrous covering. In A. catechu the
ovary is likewise monosporous.
ENCHIDIUM. W. J.
Moncecia Monadelphia. N. O. Euphorbiacca. Juss.
Calyx 5-partitus. Corolla 5-partita. Nectarium glandular
decem. Mas. Filamentum columnare, lo-antheriferum ; antheris
radiatim patentibus. Femina. Ovarium trilobum. Styli 3.
Stigmata 6.
Flores masculi et feminei in eadem spica.
ENCHIDIUM VERTICILLATUM.
Arbor spiculorum. Rumph. "Amb." iii. p. 167. t. 106.
Not unfrequent on hills in Sumatra and the Malay Islands.
A large shrub. I have not met with any that had attained to so
great a size as mentioned by Rumphius. The leaves are arranged
u 2
292 DESCRIPTIONS OF MALA YAN PLANTS.
in a kind of irregular verticils at different distances along the
branches, as exhibited in the figure quoted ; on the young shoots
they are sometimes irregularly disposed along the whole length ;
they are petiolate, lanceolate, acuminate, very entire, very smooth,
firm and somewhat leathery, of various length, generally about 6
inches long by 2\ broad. Petioles from i to 2-i- inches long,
flattened above, striated. Spikes from among the upper verticils
of leaves, bearing both male and female flowers, the former
lowermost, all pedicellate. Calyx 5-parted. Corolla purple to-
wards the centre, 5-parted, furnished with 10 callous nectaries or
glands at the base. In the male the filament is columnar, bearing
10 anthers which diverge in a radiated circle round the summit.
Thefemate has a 3-lobed ovary surmounted by 3 styles with bifid
stigmata.
Obs. There can be little doubt of the identity of this plant
with Rumphius's Arbor spicularum, of which he says he was never
able to procure the flower. I have seen great numbers of these
plants in the woods, but only once was successful in observing
the flower, and have never met with the fruit. As the spike, how-
ever, fortunately contained both male and female flowers, its
characters have been sufficiently determined to assign its proper
place. It comes nearest to Cluytia, but differs in the corolla and
in having 10 anthers with filaments united into a central column.
Both its fructification and habit appear to distinguish it from all
the present genera of the Euphorbiaceous family.
ANTIDESMA FRUTESCENS. W. J.
Frutescens, foliis oblongo-ovalibus basi rotundatis supra glabris,
racemis terminalibus et axillaribus subpaniculatis geminis solita-
riisque, nectarii glandulis quinis cum staminibus alternantibus.
Bencoolen.
A small dioecious shrub, not exceeding a few feet in height.
Branchlets tomentose. Leaves alternate, petiolate, oblong oval,
rounded and sometimes subcordate at the base, acute, sometimes
terminated by a short mucro, or awn, entire, smooth above, sub-
tomentose beneath, chiefly on the nerves ; 3 inches long. Stipules
long, subulate, acute. Racemes axillary and terminal, geminate
and solitary, somewhat panicled, tomentose ; when geminate, the
outer raceme is simple, and the inner branched ; male racemes
generally longer than the leaves, female ones shorter. Pedicels
solitary. Bracts shorter than the pedicels. Male. Calyx ^ 5-
parted, tomentose. Nectary of 5 yellow pilose glands alternating
with the stamina. Stamina 5 ; filaments much longer than the
calyx ; anthers bifid, cells bursting transversely on the summits of
the lobes. Pistil abortive, pilose. Female. Perianth 5-parted.
Ovary superior, villous, oblong ovate, compressed, i-celled, vesi-
DESCRIPTIONS OF MALAYAN PLAK <
cular, containing 2 ovula, which arc attached close together to
one side near the top, and hang forward into the cell, whki:
great part empty and inflated. Styles 2, i often biful. Drupe
subglobose, purplish, about the size of a peppercorn; n>
seeded.
Obs. It has considerable resemblance to Roxburgh's A. pubes-
cens ; that, however, is a tree, while this is a small shrub. The
most important difference appears to be in the nectary of the male
flower.
SALACIA. LINN.
This genus seems to require a little elucidation. It was origi-
nally referred to Gynandria, the fleshy nectary on which the
stamina are inserted, having been mistaken for the germen, and
the real ovary, on account of its smallness, having escaped the
observation of Linnaeus and Loureiro. This is now, I believe,
generally admitted ; there can therefore be no doubt of the identity
of Roxburgh's Johnia with Salacia, and his I. salacioides agrees so
well with S. chinensis, particularly in having entire leaves, that it is
questionable whether they are not the same, for it is to be observed
that in most of the species the leaves are only subopposite and may
occasionally on the same tree be found both opposite and alter-
nate. Tonsella prinoides (Willd. "Berl. Ges. Nat. Fr. Mag." iv.)
is also without doubt a true species of Salacia, if it be not in fact
the same plant as the Johnia Coromandeliana (Roxb. "Flor. Ind."
i. p. 173). Calypso salacioides of Aubert du Petit Thouars agrees
exactly with these in the structure of the flower but differs in hav-
ing many-seeded berries. Some of the species of Tonsella appear
likewise to have polyspermous fruit, but those which have definite
seeds are probably true species of Salacia. It may be questioned
whether the distinction founded on the number of seeds be really
of generic value where the agreement is so exact in all other
respects, especially if it should be found that a gradation exists
from the one to the other in the fruit of the different species.
This, however, can only be determined by an accurate examination
of the ovaries and fruit of the various plants, at present ranged
under Tonsella.
In the natural arrangement Salacia undoubtedly bears the
greatest affinity to Hippocratea, it being scarcely possible to
distinguish the two genera when only in flower. It also agrees in
many particulars with the Celastrinoe, but differs in having exalbu-
minous seeds. The union of the Hippocraticeae and Celastrinai
has, however, been suggested by Mr. Brown in his remarks on the
Botany of Terra Australis. Under the above view the genus will
be characterized as follows :
Calyx inferus 5-fidus. Corolla 5-petala. Stamina, 3, disco
294 DESCRIPTIONS OF MALA VAN PLANTS.
carnoso inserta. Ovarium 3-loculare, loculis 1-2 sporis, ovulis axi
affixis. Bacca 1-3 sperma.
Frutices vel arbusculae, foliis suboppositis simplicibus.
I have met with 2 species in Sumatra, i with anthers sessile on
the nectary, which agrees very nearly both with S. chinensis and
Roxburgh's I. salacioides ; the other with anthers supported on
filaments and nearly related to I. Coromandeliana Roxb.
VITIS RACEMIFERA. W. J.
Tetrandra, foliis quinatis, foliolis spinescenti-serratis subtus
incanis, cirrhis oppositifolis racemiferis, racemis compositis long-
issimis, baccis dispermis.
Akar charikan, or Bayur akar. Malay.
Native of Sumatra.
A large, strong, woody climber. Branches round, villous.
Leaves alternate, quinate, leaflets pedicellate, oblong obovate,
acute, subspinoso-serrate, the serratures being formed by the spin-
escent termination of the nerves, smooth above, hoary beneath,
frequently with a ferruginous shade. Petioles villous. Cirrhi
opposed to the leaves, very long, simple or bifid, when bifid i
branch becomes the peduncle. Racemes very long, compound,
consisting of numerous densely flowered racemuli inserted on a
peduncle formed of the thickened tendril. The whole raceme is
often a foot and a half in length. Peduncles ferruginously villous.
Flowers sessile on the partial peduncles, small, green. Calyx
minute, embracing the base of the corolla, quadridentate. Corolla
deeply 4-parted. Stamina 4, anthers yellow. Ovary surrounded
by a fleshy ring, tetrasporous. Style scarce any. Stigma thick,
Berry of the shape of an olive and nearly as large, purple, juicy,
2-seeded.
Obs. This would be a species of Cissus according to the
Linnean division, but that genus has now been united to Vitis by
Mr. Brown, as they differ in nothing but the number of parts.
RHOPALA OVATA. W. J.
Foliis subsessilibus ovatis utrinque acutis integerrimis, pedicellis
brevissimis cum calycibus ovariisque levissime tomentosis.
Found at Tappanuly.
A small tree. Leaves alternate and opposite, almost sessile,
broad ovate, acute, sometimes acuminate, entire with revolute
edges, very smooth, nerves distinct; 10 inches long by 6 broad.
Petiole none, save the thickened base of the middle nerve. Ra-
cemes below the leaves from former axils. Pedicels 2-flowered;
a bract at the base of each and at the subdivisions. Perianth,
together with the pedicels, slightly tomentose or nearly smooth.
Nectarial scales 4.
DESCRIPTIONS OF MALAYAN PLANTS. 295
LINOCIERA ODORATA. W. J.
Diandria Monogynia. N. O. Oleina.
Foliis lanceolatis utrinquc acutis glabcrrimis, paniculis axillari-
bus foliis brevioribus.
At Natal and on Pulo Mosella.
A large shrub, with subdichotomous branches. Leaves sub-
opposite, short-petioled, oblong-lanceolate, acute at both ends,
entire, smooth and coriaceous ; 4-5 inches long. Panicles axil-
lary, opposite, much shorter than the leaves ; peduncles opposite,
3-5 flowered. Flowers subsessile, fragrant. Bracts small, oblong,
Calyx 4-parted. Corolla white, almost 4-petaled, petals long,
linear, united by pairs, by means of the filaments, slightly coher-
ing at the other divisions. Stamina 2 ; anthers large, emarginate
at the apex. Ovary 2-celled, each cell containing 2 linear pen-
dulous parallel ovula. Style scarce any. Stigma bifid.
The following species have been discovered since the printing
of this paper, and may be here briefly noticed :
ADINANDRA SYLVESTRIS. W. J.
Baccis trilocularibus.
Suka beranak. Malay.
A large forest tree, found at Moco Moco.
PTERNANDRA CAPITELLATA. W. J.
Floribus axillaribus capitellatis.
Found at Moco Moco.
PTERNANDRA ECHINATA. W. J.
Pedunculis axillaribus terminalibusque, calycibus ovariisque
echinatis.
A large tree, found at Kataun. The leaves are 3-nerved in al
the species.
PSILOBIUM TOMENTOSUM. W. J.
Tomentosa, floribus axillaribus subsessilibus.
At Kataun. The fruit is baccate.
296 DESCRIPTIONS OF MALA VAN PLANTS.
NOTE.
Since the foregoing article On Malayan Plants was printed, Sir J. D.
Hooker has kindly supplied the Editor with the following references,
while the Hon. D. F. A. Hervey has added some valuable corrections
and suggestions concerning the Malay names. The former are here
given in Italic, the latter in Roman type.
p. 211. Psychotria Malayana. (Flora Brit. Ind. iii. 165.) Bayam
badak, i.e., rhinoceros spinach, probably relished as food by
that animal.
Rondeletia corymbosa. Grcenia Jackii, W. and A, (Ib.
iii. 41.)
p. 212. Phyteuma begonifolium, Roxb. Pentaphragma bcgonifoliiim,
Wall. (Ib. iii. 437.)
Curculigo Sumatrana. (Wight, Ic. PL Ind. or. t. 2042.)
Kalapa, cocos nucifera, piiyu, a fish so named ; but this is
probably a mistake for piiyuh, the larger of the two quails
found in the Archipelago, which frequents this plant,
p. 213. Loranthus coccineus. (P"lora Brit. hid. v. 206.)
Loranthus ferrugineus. (Ib. v. 210.)
p. 214. Nephelium lappaceum, Linn. (Ib. i. 687.) There is a
variety called rambutan pachat (i.e., leech rambutan) dis-
tinguished from the ordinary variety by the smaller size
of the fruit, and the way in which the soft spines of the shell
curl over, looking like the leech on his way to attach himself
to some fresh point. It is used medicinally with other
remedies in small-pox.
Sapindus rubiginosus, Erioglossnm edule, Blumc. (Ib. i.
672.) Kulit layu, withered, faded bark,
p. 215. Melia excelsa. (7#. i. 544.)
Microcos tomentosa. Grewia paniculata, Ro.rb. (Ib. i. 393.)
p. 216. Mimosa jiringn, W. J. Pithecolobium lobatum, Benlli. (lb.\\.
305.) Mimosa Kaeringa, Roxb. Buah jering, a fruit eaten
by Malays. Other varieties are jering tupai (squirrel j.),
Pith, oppositum, and j. hantu (spirit j.), Pith, bigeminum.
p. 217. Clerodendrum molle. C. villosum, Bhcmc. (Ib. iv. 595.)
p. 218. Gmelina villosa. (Ib. iv. 582.)
Vitcx arborea. V. pubescens, Vahl. (Ib. iv. 585.) The wood
is similarly used in the Peninsula,
p. 219. Sphenodesme pentandra. (Ib. iv. 602.)
Sterculia coccinea, Roxb. S. laevis, Wall. (Ib. i. 357.)
This is an error of Jack's, it is not Roxburgh's coccinea,
which is a N. Indian species.
p. 220. Sterculia angustifolia. S. rubiginosa, Vent. (Ib. i. 358.) Filet
makes this a variety of the Sundancse hantap. Roxburgh's
S. angnst. is a different species, referred to S. Balan-
ghas, 'L.
NOTE.
297
p. 220. Calla humilis. Chamacladon hunrilc, Mi,,. (EnrUr Manoe
Arac. t 345.)
p. 221. Clala angustifolia. Chamacladon angustifolium, Schott.
(Ib. 34,4..)
Calla nitida. Aglaoncma nitidum, Knnth. (Ib. 438.)
Flacourtia inermis. (Flora Brit. hui. i. 192.) Variety of the
fruit known as "rokam" or "rukam," probably ' rokam
manis," or the sweet variety, also described as Flacourtia
rukam. There are other varieties viz., " r. asam " (Flacourtia
sapida), and "r. si-pat" (F. jangomas).
p. 222. Rottlera alba. Mallotus albus, Muell. Arg. (De Cand. Prodr.
xv. 2, 965.) Balik angin, turn wind, meaning that turns up
its under side with the wind, and sh,ows the whiteness of it.
p. 223. Didymocarpus crinita. (Flora Brit. Ind. iv. 351.) Temu'.
There are several other varieties viz., tCinu kunchi
(Kaempferia pandurata), " t. giring " (Curcuma viridiflora),
and '* t. lawat " (Curcuma zerumbet).
p. 224. Didymocarpus reptans. (Id. iv. 352.) Temu kunchi.
Didymocarpus corniculata. (De Cand. Prodr. ix. 265.)
p. 225. Didymocarpus frutescens. Didissandra frutcscens, Clarke.
(Flora Brit. Ind. iv. 355.)
Sonerila erecta. (Ib. ii. 530.) Sambau, according to Johor
aboriginal tradition, one of the first plants seen by the first
parents of mankind. Used medicinally,
p. 226. Sonerila Moluccana. (Ib. ii. 537.)
p. 227. Rhopala attenuata. Helicia attcnuata, Bl. (Ib. v. 190.)
Rhopala Moluccana. Helicia pettolaris, Benn. (Ib. \. 190.)
p. 228. Ixora pendula. (Ib. iii. 141.)
Epithinia Malayana. Scyphiphora hydrophyllacea. (Ib. iii. 125.)
p. 229. Morinda tetrandra. M. umbellata, Linn. (Ib. iii. 157.)
Morinda polysperma. Lucinea morinda, DC C. (Ib. iii. 93.)
p. 230. Euthemis leucocarpa. (Id. i. 526.) There is a very hard
timber tree, named pelawan ("lawan" to resist), of which
there are several varieties in hill, plain, and swamp,
according to Malays. This shrub, E. leucocarpa, which
Jack gives as " pelawan beruk," or the p. of the cocoa-nut
monkey (Simius nemestrinus), is mentioned by Filet as mdta
pelandok," eye of the Moschus javanicus, or an allied
species, from the berry, which, when ripe, is of a lustrous
black, but earlier of a bright scarlet, and perhaps white in
the younger stages.
p. 231 . Euthemis minor. (Ib. i. 526.) Filet gives, as the Malay of this,
"putat ayer," indicating a swampy habitat, but Jack has no
hint on this point.
Celastrus bivalis. Microtropis bivalvis, Wall. (Ib. i. 614.)
p. 232. Leucopogon Malayanum. (Ib. iii. 477.) Called "tt-ruta
Bangka according to Filet.
p. 233. Rauwolfia Sumatrann. (De Cand. Prodr. viii. 337.) Used
medicinally in conjunction with many other plants : <; stSmbu
badak," horn of rhinoceros; "tampal," is a piece, to mend
or to patch, but I am unaware of its application here.
298 DESCRIPTIONS OF MALA VAN PLANTS.
p. 234. Tacca cristata. (.Miguel, Flora Ind. Bat. iii. 578.) " Puar
lilipnn,''' at Palembang, according to Filet, and would no
doubt be a ' ; puar ; ' of some kind in the Peninsula. These
"puar," mostly zingibers, are many of them used medi-
cinally.
p. 234. Vcratrum Malayanum. Veratronia Malayana, MiqueL (Flora
Ind. Bat. iii. 553.)
p. 235. Memecylon coeruleum. (Flora Brit. Ind. ii. 559.) " Kulit
nipis," thin bark.
p. 236. Laurus parthenoxylon. Cinnamomum parthenoxylon, Meissn.
(Ib. v. 135.)" Kayu gadis," virgin wood. Filet gives this as
Parthenoxylon porrectum, Bl., Nat. Fam. Laurinea?, a sort of
pseudo- sassafras.
p. 237. Gomphia Sumatrana. (Ib. i. 525.)
p. 238. Murraya paniculata, W. J. Murray a exotica, Linn., Van. (Ib.
i- 53-) Filet gives it as " Kemuning Japan," or, as a Malay
would say, "Jipiin."
Aglaia odorata. (Ib. i. 554.) Filet, " Kemuning China."
p. 239. Rhizophora caryophylloides. Bruguiera caryophylloides,
Blume. (Ib. ii. 438.) Filet calls it'" Kandaka (or gendaga)
nasi," i.e., boiled rice case.
p. 240. Acrotrema costatum. (Ib. i. 32.)
p. 241. Lagerstrcemia floribunda. (Ib. ii. 577.)
Ternstrcemia rubiginosa. SauraiyaJackiana,Kort. (Miguel,
Flora Ind. Bat. i. 2, 479.)
p. 242. Ternstrcemia pentapetala. Sauraiya tristata, De C. (Flora
Br. Ind. i. 287.)
Elosocarpus nitida. (Ib. i. 401.)" Buah manik," jewel fruit.
p. 243. Monocera petiolata. Elccocarpus Integra, Wall. (Ib. i. 408.)
p. 244. Monocera ferruginea. (Ib. i. 409.)
Tetracera arborescens. (Miguel, Flora Ind. Bat. i. 2, 9.)
p. 245. Uvaria hirsuta. (Flora Brit. Ind. i. 48.)
Careya macrostachya. Barringtonia macrostachya, Kurz.
(Id. ii. 509.)
p. 246. Clerodendrum divaricatum. (Miguel, Flora Ind. Bat. ii. 882.)
HedychiumSumatranum. (Ib. iii. 608.) " Gandasuli hutan," the
jungle or wild gandasuli, the ordinary being H. coronaium.
" Ganda " seems to be Sanskrit for odour, which occurs in
several Malay plant names, e.g., "gandapura" (Abelmoschus
moschatus, Filet) and "gnndarusa" (Gandarussa vulgaris,
Nees, or Justicia gandarussa.) " Suli" in this name comes,
it is suggested, from the Sinhalese " sulinga," spiral (Rigg,
quoted by Favre).
p. 247. Alpinia elatior. (Ib. iii. 606.) " Bunga" flower ; " Kinchong,"
the meaning of this is not easy to fix, as it is not certain
hovv the word should be sounded; it might be " Kinchang,"
" Kcnchang," or " Kinchong " i.e., strong, fine, or unfor-
tunate.
p. 248. Alpinia capitellata. (Ib. iii. 607.)
Globba ciliata. (Ib. iii 592.)" Puar amas," golden puar. Filet
calls it "puar amut," speaking of the Sumatran variety also.
NOTE. 299
p. 249. Aristolochia hastata. A.Jackiana,Steud. (Miguel, Flora Ind.
Bat. i. i, 1067.)
p. 250. Begonia caespitosa. (De Cand. Frodr. xv. i, 397.)
Begonia orbiculata. (Id. xv. I, 398.)
p. 251. Begonia sublobata. (Ib. xv. i, 353.)
p. 252. Begonia fasciculata. (De Cand. Prodr. xv. i, 522.)
Begonia pilosa. (Ib. xv. i, 398.)
p. 253. Begonia bracteata. (Ib. xv. i, 316.)" Bunko " = Bengkok or
Bongkok. [See note above, p. 57.]
Begonia racemosa. (Ib. xv. i, 322.) " Lay.ing-la> ang," means
both the swallow, and a flying kite ; " simpei '' is a hoop,
circle or loop, also the name of a monkey (Semnopithecus
melalophos). Probably the name means " monkey's kite,''
referring to the round capsules of the female. It is common
for Malays to give names of this kind to plants. Filet
calls it Diploclinium racemosum.
Begonia geniculata, B. isoptera, Jack. (Ib. xv. i. 320.) Filet
(No. 6910) gives this as Diploclinium bombycinum, Bl.
" Rumput udang-udang," shrimp or prawn grass. The
Malays call many plants grass e.g., Ruellia repanda is called
" rumput mas," or golden grass, the Mimosa pudica is called
" rumput kamaluan," the bashful grass, c. &c.
p. 254. Sonerila heterophylla. (Miguel, Flora. Ind. Bat. i. I, 582.)
Rhododendron Malayanum. (Flora Brit. Ind. iii. 462.)
p. 255. Vaccinium Sumatranum. (Miguel, Flora Ind. Bat. ii. 1063.)
p. 256. Haloragis disticha. Anisophyllea disticha, Hook. (Flora Brit.
Ind. ii. 442.) " Kayu," wood, ' Kanchil," small deer,
(variety of Moschus javanicus), the cunning animal in Malay
tales, like the fox in others.
p. 257. Elodea Sumatrana. Cratoxylon Sumatranum. (Miguel, Flora
Ind. Bat. i. 2, 516.) Tello Dalam, i.e., fi Telok dalam,'
deep bay.
p. 258. Elodea formosa. Cratoxylon formosum, Benth. and Hook. f.
(Flora Brit. Ind. i. 258.) " Kayu gagak," crow tree;
" pedas," pungent, "bunga," flower. This would apparently
indicate that there is another plant of the same name which
is not supposed to flower.
p. 259. Ternstrcemia acuminata. Not taken up in later works.
Ternstrcemia serrata. (Miguel, Flora Ind. Bat. \. 2, 488.)
p. 260. Ternstrcemia cuspidata. Not taken up in later works.
Salumah is probably " Sri Lt-mak," one of the districts in
the Menangkabau country in Sumatra.
p. 261. Millingtonia Sumatrana. Meliosma Sumatrana, Hook. f.
(Flora Brit. Ind. ii. 6.)
p. 262. Laurus incrassatus. Dehaasia microcarpa, Blume. {Ib. v.
126.) " Jcring-jering tupei," the squirrel jering. ^Filet gives
it as Pithecolobium oppositum, the ordinary jering as P.
lobatum, jcring hantu as P. bigeminum, N.O. Mimoseac.
p. 263. Tetranthera cordata. Litsoea cordata, Hook.f. (Ib. v. 177.)
Knema glaucescens. (?) Myristica glaucescens. (Ib. v. in.)
p. 264. Connarus ferrugineus. (Ib. ii. 51.)
300 DESCRIPTIONS OF MALAYAN PLANTS.
p. 265. Connarus villosus. (Miqttet, Flora Ind. Bat. i. 2, 666.)
p. 266. Connarus semidecandrus. (Flora Brit. Ind. ii. 52.) Filet calls
it Karabu. If it is called "akar," it should be a creeper.
Connarus grandis. (Ib. ii. 53.) Sundanese native name
" Kilaja," Filet.
Connarus lucidus. (Miquel, Flora Ind. Bat. i. 2, 666.)
p. 267. Cnestis emarginata. Not taken up in later ivorks.
Cnestis florida. (?} Rourea simpticifolia. (Ib. i. 2, 659.)
p. 268. Cnestis mimosoides. (?} Rourea concolor, Blumc. (Flora
Brit. Ind. ii. 49.
Eurycoma longifolia, W. J. (Ib. i. 521.) " Kayu kdbal " the
invulnerable wood, whether used superstitiously or because
the wood is hard. Jack gives no hint. Filet gives the
Sumatran designation, ' babi kurus,' thin pig, but without
stating whether the porcine tribe value the fruit for anti-
Banting properties.
p. 269. Peronema canescens, W. J. (Flora Brit. Ind. iv. 599.)
Javanese name " Kilangir " (Filet). It is also used for fences
and for sheaths of common knives.
p. 270. Rhodamnia cinerea, W. J. R. trineura, Blumc. (Ib. ii. 468.)
Filet gives it as " marampuyan," and also mentions
R. concolor, and says <; beide hooge boomen," so that the
specimen Jack saw could not have been full grown. The
"merpoyan" of the Peninsula gives a hard wood, and is
used in carpentering, and also, with other ingredients,
medicinally in diarrhoea.
p. 271. Adinandra dumosa, W. J. (Ib. i. 282.) In the Peninsula
there is a shrub called " ribu-ribu," which seems to corre-
spond with the description here given, but the berry is red,
and Jack is silent as to the colour. It is used medicinally
in a variety of ways, and is called " mcrkasih," by the
aborigines of Johor. The Ligodium scandens is called
"puku" or "rumput saribu," the epithet " ribu," thousand,
having reference doubtless in this case, as in the text, to the
countless number of leaves.
p. 272. Ixonanthes reticulata, W. J. (Ib. i. 417.)
Ixonanthes icosandra, W. J. (Ib, i. 416.)
p. 273. Chionotria rigida, W. J. (/) Glycosmis pcntaphylla, Corr.
(Ib. i. 500.) Called " biarang " in Bangka.
p. 274. Sphalanthus confertus, W. J. Quisqualis densiflora, Wall.
(Ib. ii. 460.)
p. 275. Pyrrhanthus littoreus. Lumnitzera coccinca, Wd. and A. (Ib.
ii. 452.) "Miri " for "kemiri" (?) "bfitu/' stone, meaning
hard variety.
p. 276. Phaleria capitata, W. J. Drymispcrmum phaleria, Meissn.
(Miquel, Flora Ind. Bat. i. i, 884.)
p. 277. Pternandra ccerulescens. (Flora Brit. Ind. ii. 551.)
Memecylon paniculatum. (Miquel, Flora Ind. Bat. i. I, 572.)
p. 278. Octas. Genus not taken up by later authors.
p. 279. Ccelopyrum. Genus of unknown affinity.
Ccelopyrum coriaceum. Filet gives " Tarantang " simply as
NOT/ . 3 o,
' Iludianania auriculata," IJI.,jN. O. Anacarcli.iccrv, and refers
to B. macrophylla, which t-.ivc name
of ' mrdan^ sangka." Then there are B. scss.
B. splendens, native names "t. ayam " an 1 "t. 1 urong"
respectively. Leaves and root of >vrer,
Malacca), used medicinally.
p. 279. Petrocarpa excelsa. Parinarium (?) Jackinnum, Benth. (Flora
/!>:(. Ind. ii. 312.) " Balam" is applied to one of thegutta-
(gctah) producing trees, and generally to a icd variety of any
species. " Pangkat '"' means elevation, rank, and may refer to
the situation or the good quality of this variety.
p. 280. Petrocarpa Sumatrana. Parinarium costatum, Blttmc. (?), (Ib.
ii. 309.) Filet ^ives the Sumatran name of this as ' taij.is "
(tayas), N.O. Chrysobalanese.
p. 281. Wormia excelsa. (Miguel, Flora Ind. Bat. i. 2, 10.)
Wormia pulchella. (Flora Brit. Ind. i. 36.)
p. 282. Fictis ovoidea. Urostigma ovoideum. (Miguel, Flora Ind.
Bat. i. 2. 345.)
Ficus deltoidea and Ficus rigida. Not taken up in later works.
" Sen," the complexion, glorious, illustrious ; " bulan,''
the moon : this may refer to the appearance of the fruit,
bark, or the leaves in the moonlight,
p. 283. Jonesia. Saraca, Linn.
Jonesia declinata. Saraca declinata, Miguel. (Miguel, Flora
Ind. Bat. ii. 84.)" Siturun," from " turun," to descend,
p. 284. Bauhinia emarginata. (Flora Brit. Ind. ii. 278.)
Bauhinia bidentata. (Ib. ii. 278.) Malay name " akar ka-
katup ; " used medicinally, the root being boiled with those
of other plants, and the decoction drunk for diarrhoea,
p. 285. Inga bubalina. Pithecolobium bubalinum, Benth. (Ib. ii. 304.)
p. 286. Inga clypearia. Pithecolobium clyperia, Benth. (Ib. ii. 305.)
"Jering munyet" *'.., the monkey variety of jering, pro-
bably because that animal feeds on the fruit. Filet
gives the Bangka name as "kabu-kabu" (which, in the
Straits, is applied to the tree producing the cotton-pods),
and says of it, "they make large canoes of the trunk, which
are light, but not very durable, because the wood absorbs too
much water and thus does not easily become dry again ; they
also make shields or " salowakkos " of it. The bark is used
for the tanning of fishing-nets."
Tabernsemontana macrocarpa. (Ib. iii. 649.)
p. 287. Fagrcea carnosa. (Ib. iv. 82.)
Fagrcea auriculata. Filet only gives the Chinese name " Fan-
nyin-won " as the native equivalent.
p. 288. Ixora neriifolia. Not taken up in later works. Whether it be
"bunga," flower, or "kayu" or "poko," tree, makes no
difference ; the name "saluang," is taken from a fish which in
form and dimensions resembles the leaves of this shrub.
Lecanthus erubescens, W. J. (Ib. iii. no.)
p. 289. Psilobium nutans, W. J. (Miguel, Flora Ind. Bat. ii. 199.)
Ophiorrhiza heterophylla. (Ib. ii. 175.)
Quercus racemosa. Oucrcus spicata^ Smith. (Dc Cand. Prodr.
302 DESCRIPTIONS OF MALA VAN PLANTS.
xvi. 2, 85.) "Prning-pening bungkus." The spelling in the
text was a rough attempt to convey the sound which the
word has in the Mi-nangkfibau dialect. "Empening" and
" hempening" are other forms.
p. 290. Quercus urceolarius. (Id. xvi. 2, 89.)
p. 290. Areca tigillaria. Oncosperma filameniosa, Blumc. (Miguel,
Flora hid. Bat. iii. 13.) It is also used for floors and
for boat decks, being split up into fine laths ; poles of
this also, with sharpened ends, arc often used with fatal
effect in riots.
p. 291. Enchidium. Trigonostemon, Blumc.
Enchidium verticillatum. (Ib. i. 2. 363.)
p. 292. Antidesma frutescens. (De Cand. Prodromus, xv. 2, 250.)
p. 294. Vitis racemifera. Not taken up by later authors. Filet
identifies" bayur akar" with Canthium glomerulatum, Miq..
N.O. Rubiaceac.
Rhopala ovata. Helicia ovata, Benn. (Miquel, Flora Ind.
Bat. i. i, 984.)
p. 295. Linociera odorata. (Ib. ii. 554.)
Adinandra sylvestris. Not taken lip by later authors. " Suka
boranak" i.e., fond of having children.
Pternandra capitellata. (Flora Brit. Ind. ii. 551.)
Pternandra echinata. Kibassia simplex, Korth. (Ib. ii. 533.)
Psilobium tomentosum. Not taken up by later authors.
GENERAL AND GEOGRAPHICAL
INDEX.
ACHEH, Achfn, i. 214, 215
'Adat sagala raja-raja Malayu, ii. 46
Ambergris island, i. 222
Ant-eater, ii. 200 ff.
Arjuna vijaya, ii. 89
Aru, i. 216, 217
Attar of roses, i. 261
Ayer etam, i. 12
kiti, ii. 58
BAB AD, ii. 91
Bade, wadah, ii. 140 ff.
Bale, ii. 99, 141
Bali, i. 138, 139, 183, 184; 11.69-200
meaning, ii. 70; language, 71;
its ingredients, 74 ff. ; litera-
ture, 77 ff. ; religion, 97 ff. ;
places of worship, 100 ff. ;
the gods worshipped, 102 ff. ;
Siva's attributes, 104 ; the cre-
ation, 114 ff. ; religious cere-
monies and offerings, 12 1 ff. ;
dress of panciitas, 124; dress
of the gods, 126 ; feasts, 127 ;
details of worship, 130 ff . ;
rishis, 136; trimurti, 137; cre-
mations, i37ff.; castes, 151 ff. ;
Brahmans, 154 ff. ; Kshatri-
yas, 158 ff. ; Wesyas, 160 ff. ;
princely families, 162 ff. ;
further remarks on the castes,
183 ; feudal system, 184 ;
^udras, 1 86 ; caste in Java,
189 ; calculation of time, 191 ;
calendar, 193
Bali-sangraha, ii. 70
Banjermasin, i. 227
Banka, i. 202
Bantam, i. 179, iSl
Barata Yudda, ii. 86 ff.
Baruna, ii, 109
Batu Bayas, i. 1 1
Birtam, i. II
Lanchong, i. n
Beaju, i. 228
Bela, ii. 146
Bencoolen, ii. 57 ff.
Benko, Bengkok, ii. 57
Bidayatu-lhidayat, ii. 48
Billiton, i. 148, 151, 201
Bomakavya, ii. 89
Brahma, ii. 105
Brahmans, ii. 154 ff.
Brunei, i. 223
Bukit China, i. 4
Jalutong, i. 12, 1 6
Kandfs, ii. 60
Merah, i. 12, 16, 17
Tangah, i. 13
Burning of widows, ii. 145 ff-
Bustanu-l'ariffn, ii. 41
ssalatin, ii. 15
CAMPHOR-BAROS, i. 260
Cassowary, i. 262
Chang-kwang, i. 133
Character of Balinese, ii. 189
Charitra nabl Allah Musa, ii. 39
Chinese geographical literature, i. 1 26
DAFTAR shajarah Charibon, ii. 22, 56
Duraka Juru, i. 13, 18, 19
Dvipas, ii. 117
EXPIATORY feasts, ii. 128
FANTSUR, i. 164 ; ii. 51
Farquhar Collection of Malay MSS.
ii. 45-49
Funan, i. 239
3 o 4 GENERAL AND GEOGRAPHICAL INDEX,
GANA, Ganesa, ii. in
Giau-chi, i. 127, 128, 205
Grisse, i. 173, 179, 180
Grooved rocks, i. 25
Gunong Bau, i. 22
Bubu, i. 10
Belumut, ii. 60
Benko, Journey to, ii. 57 IT.
Jcrai, i. 12
HAKANG, i. 179, 181, 182
Hamzah, his works, ii. 51
Harivangsa, ii. 90
Hikayat Ahmad Bisnu, ii. 35
Bakhtiyar, ii. 39
Bayan Budiman, ii. 6
Bikermaditya, ii. 38
Budak miskin, ii. 53
Barma Shahdan, ii. 12
chabut tunggul, ii. 30
Charang Kulina, ii. 13, 56
Chikat Waning Pali, ii. 19
Dalang Indra Kftsuma, ii. 19
Panguda Asmara, ii. 17
Daniar Bulan, ii. 6
Dewa Mandu, ii. 31
endang Malat Rasmi, ii. 20
Fatimah kawan, ii. 39, 54
Ghulam, ii. 40
Hang Tuah, ii. I
Ind&ra Kryangan, ii. 36
Putra, ii. 10
Isma Yatim, ii. 14
Kalflah wa Damanab, ii. 28
Khojah Meimun, ii. 6
Maharaja 'Ali, ii. 42
Boma, ii. 13, 52
Mesa Indara Dewa Kitsumn,
ii. 32
Lari Kasumah, ii. 30
Naga Barsaru, ii. 21
P&landuk Jinaka, ii. 41
Pandawa lima, ii. 18
Jaya, ii. 3, 52
Pangeran K&suma Agung, ii. 17
Panji Wila K&suma, ii. 21
partan islam, ii. 32
Putri Bilkis, ii. 39
Johor Manikam, ii. 37
raja Babi, ii. 134
Iskandar, ii. 46
raja-raja Pasei, ii. 41
Ranga Ari} a Kuda, ii. 5
S&ri Rama, ii. 19
Shah Kobad, ii. 22
Shamsu-lbarri, ii. 38
Si Miskin, ii. 35
tamimu-ddari, ii. 34
Iluiku, i. 232
Human sacrifices, ii. 129, 145
ICHNEUMON, ii. 205 flT.
Indra, ii. 108
Indragiri, i. 200
JAVA, Chinese accounts of, i. 131
Jilahati, i. 184
Jih-nan, i. 128
Johore, i. 254
KALA, i. 241, 243
Kalah, i. 243
Kalang, i. 149
Kaling, i. 138, 140, 183
Kandali, i. 185, 192, 193, 200
Kanyoh, i. 226
Karimata, i. 157, 236
Kataun, ii. 64, 65
Kaulan, Kolan, i. 201, 236
Kavvi language, i. 282 ff. ; ii. 73 IT.
Kelantan, i. 257
Kenhangrok, ii. 91
Kianchou, i. 239
Klaebang, i. 5, 7
Kora, i, 241, 243
Kubera, ii. no
Kukang, i. 168, 169, 195, 197, 199,
200
LAMBRI, i. 169, 219, 220, 221
Land-shells of Pinang, i. 87 ff.
Langga, i. 135
Langpi, i. 140, 141
Law-books, ii. 93
Laye, Lais, ii. 59, 62
Lignum-aloes, i. 260
Lingga, i. 203
Litai, i. 208, 219
Lubu Puar, ii. 58
MAHABHARATA, ii. 84 ff.
Majapahit, Mojopait, i. 149, 171 ; ii.
77, 7Q, 159
Makota sagala raja-raja, n. 15
Malacca, i.' 1-9, 243-254.
Malagasy language, i. 263-286 ; its
affinities, 263 ; grammatical
structure, 266 ; idiosyncrasy,
267 ; Sanskrit and Arabic in-
gredients, 269 ; phonetic sys-
tem, 271
Malat, ii. 97
Malay Archipelago, Notes on, from
Chinese Sources, i. 126-262
Malay character, i. 17
GENERAL AND GEOGRAPHICAL
Malay Manuscripts, Account of, ii.
1-56
Pantuns, ii. 65 ff.
Malayan Amphibia and Reptilia, i.
72 ff.
Plants Described, ii. 209-302
Mantras, Account of the, i. 286-307 ;
other native tribes, 287 ;
origin, 288 ; traditions, 289 ;
habits and customs, 290 ;
weapons, 292 ; character, 292 ;
games, 293 ; ceremonies, 295 ;
religion, 297 ; government,
301 ; language, 303 ; missions,
35
Marbukit, i. 22
Marigi, ii. 58
Mausu, i. 224, 257
Mayitung, i. 202
Ma'zijat rasul allah, ii. 32, 39
Milikii, Moluccos, i. 183, 237
NAKUR, i. 208, 218
Nuruddin, his works, ii. 49
PADANDAS, ii. 99 ff., 156
Pahang, i. 255, 257
Palembang,i. 163, 168, 184, 185, 188,
199
Pamendanga, ii. 92
Panditas, ii. 157
Passier Ries, i. 21
Pekalongan, i. 166
Perak, i. 10
Permatang Pau, i. 12
Pinang, i. 9, 20, 87 ff.
Poli, i. 203, 205, 207, 242
Polo, i. 222
Porcupine, ii. 207 ft.
Prye, i. 12, 19, 20
Pulo Kindi, i. 10
Riman, i. 10
rondo, i. 222
Sejahat, i. 23
Tarn (Ktam), i. 22, 39
Tikang, i. 22, 26
Ubin, i. 21 ff.; its rocks, plutonic
and volcanic, 45
Puni, 5. 225, 229, 257
Punjong, ii. 58
Purohita, ii. 157
PyahTrubong, i. 12
RAFFLES' Collection of Malay MSS.,
ii. 1-45
Ramayana, ii. 80 ff.
Rangga Lawe, ii. 91
Rejak Bessi, ii. 58
SECOND SERIES. VOL. II.
Rejang, ii. 58, 62, 63
Rindowati, n. 10
SALANGOR, i. 10
Salat Tambroh, i. 22
Salsalah raja-raja di tanahjawa, ii. 2
Samarkand f, ii. 53
San-bo-tsai, i. 187, 192, 193, 197
Sarbaza, i. 187, 200
Sarfbu-masail, ii. 38
Satya, ii. 146
Serawi, ii. 62, 63, 69
Sha'ir AngSreni, ii. 40, 56
Bidasari, ii. 7
buang, ii. 48
ikan, ii. 35
Tambara, ii. 10
Javan Tamasa, ii. 53
johan anak raja Perak, ii. 48
Ken Tambuhan, ii. 8, 40
Kumpani Wolanda, ii. 46
prang Angres di Batawi, ii. 43
S&ri Buniyan, ii. 9
Silindung dalima, ii. 9, 54
Shajara Malayu, ii. 16
Shamsuddin, his works, ii. 52
Shrew-mouse, ii. 203 f.
Sillebar, Selebar, ii. 62
Simpang ayer, ii. 58
Sivaites, ii. 98 ff.
S'kodo, i. 29
Smaradahana, ii. 88
Soli, i. 166
Sukitan, i. 179
Sulu, i. 225, 227
Sumana Santaka, ii. 89
Sumatra, i. 162, 184. 208, 21 1
Sunda language, ii. 64, 69
Sungei Baru, i. 19
Jara, i. 19
Kalim, i. 19
Labu Marijam, i. 19
Lamau, Lemau, Limau, ii. 58
Surabaya, i. 171, 179
Sutasoma, ii. 90
TijU-sSALATfN, ii. 1 6, 30
Tanjong Agung, ii. 58
Jangy, i. 21
Kling, i. 3
Pamudang, i. 34
Pongal, i. 21 ' m
Sanei, ii. 61
Tajam, i. 32, 33
Tello Anou, ii. 6l
Tiehli, i. 184
Timor, Timun, i. 236
Tiongkalo, i. 182, 237
306 GENERAL AND GEOGRAPHICAL INDEX.
Tuban, i. 171, 179
Tulloh Kumbar, i. 10
Tumapol, i. 149, 162, 165
Tunsun, i. 239, 240
Tuturs, ii. 93
UNDANG-UNDANG, ii. 26
raja Malaka, ii. 46, 49
Usana Bali, ii. 70, 92
Java, ii. 92
VIVAHA, ii. 88
Vishnu, ii. 106
WES Y AS, ii. 1 60
Wriga Garga, ii. 191, 199
YAMA, ii. 109
Yortan, i. 179
INDEX OF LATIN TERMS.
ABLABES flaviceps, i. 84
Acrotrema cospatum, ii. 240
Adinandra dumosa, ii. 271
sylvestris, ii. 295
Aglaia odorata, ii. 238
Aglaonema nitidum, ii. 297
Alpinia alatior, ii. 247
capitellata, ii. 248
Alycseus gibbosulus, i. 95
Amonaum biflorum, ii. 210
Anisophyllex disticha, ii. 299
Antidesma frutescens, ii. 292
Areca tigillaria, ii. 290
Aristolochia hastata, ii. 249
jackiana, ii. 298
AquUaria agallocha, ii. 260
BARRINGTONIA macrostachya, ii. 298
Bauhinia emarginata, ii. 284
bidentata, ii. 284
Begonia bracteata, ii. 253
csespitosa, ii. 250
fasciculata, ii. 252
geniculata, ii. 253
isoptera, ii. 299
orbiculata, ii. 250
pilosa, ii. 252
racemosa, ii. 253
sublobata, ii. 251
Bruguiera caryophylloides, ii. 298
Bulimus atricallosus, i. 1 14
interrupts, i. 114 .
CALAMARIA stahlknechti, i. 80
Calla angustifolia, ii. 221
humilis, ii. 220
nitida, ii. 221
Cantoriana, i. 109
Careya macrostachya, ii. 245
Celastrus bivalvis, ii. 231
Chamaeladon humile, ii. 297
angustifolium, ii. 297
Chionotria rigida, ii. 273
Cinnamomum parthenoxylon, ii. 298
Clausilia filicostata, i. 116
penangensis, i. 115
Clerodendrum divaricatum, ii. 246
molle, ii. 217
villosum, ii. 296
Cnestis emarginata, ii. 267
florida, ii. 267
mimosoides, ii. 268
Coelopyrum coriaceum, ii. 279
Connarus ferrugineus, ii. 264
grand is, ii. 266
lucidus, ii. 266
semidecandra, ii. 266 .
villosus, ii. 265
Cratoxylon formosum, ii. 299
sumatranum, ii. 299
Curculiga sumatrana, ii. 212
Cyclophis tricolor, i. 83
Cyclophorus borneensis, i. 89
malayanus, i. 88
Cyclostomacea, i. 88
DEHAASIA microcarpa, ii. 299
Dendrophis caudolineatus, i. 84
Didissandra frutescens, ii. 297 J
Didymocarpus corniculata, ii. 224
crinita, ii. 223
frutescens, ii. 225
rep tans, ii. 224
Draco fimbriatus, i. 80
quinquefasciatus, i. 79
Drymispermum phaleria, ii. 300
ELJEOCARPUS integra, ii. 298
nitida, ii. 242
Elodea formosa, ii. 258
sumatrana, ii. 257
Enchidium verticillatum, ii. 291
Ennea bicolor, i. I2O
Epithinia malayana, ii. 228
Erioglossum edule, ii. 296
Euprepes olivaceus, i. 79
Eurycoma longifolia, ii. 268
308
INDEX OF LATIN TERMS.
Euthemis leucocarpa, ii. 230
minor, ii. 231
FAGRCEA auriculata, ii. 287
carnosa, ii. 287
Ficus deltoidea, ii. 282
ovoidea, ii. 282
rigida, ii. 282
Flacourtia inermis, ii. 221, 297
GLOBBA ciliata, ii. 248
Glycosmis pentaphylla, ii. 300
Gmelina villosa, ii. 218
Gomphia sumatrana, ii. 237
Gonyosoma oxycephalum, i. 84
Greenia jackii, ii. 296
Grewia paniculata, ii. 296
Gymnodactylus pulchellus, i. 79
HALORAGIS disticha, ii. 256
Hedychium sumatranum, ii. 246
Helicacea, i. 98
Helicarion permolle, i. 105
Helicia attenuata, ii. 297
ovata, ii. 302
petiolaris, ii. 297
Helix similaris, i. 113
Hipsirhina alternans. i. 86
Hystrix cristata, ii. 207
torquatus, ii. 208
INGA bubalina, ii. 285
clypearia, ii. 286
Ixonanthes icosandra, ii. 272
reticulata, ii. 272
Ixora nerufolia, ii. 288
pendula, ii. 228
JONESIA declinata, ii. 283
KIBASSIA simplex, ii. 302
Knema glaucescens, ii. 263
LAGERSTRCEMIA floribunda, ii. 241
Lagocheilus, i. 96
striolatus, i. 97
trochoides, i. 96
Laurus incrassatus, ii. 262
parthenoxylon, ii. 266
Lecananthus erubescens, ii. 288
Leucopogon malayanum, ii. 232
Linociera odorata, ii. 295
Litsaea cordata, ii. 299
Loranthus coccineus, ii. 213
ferrugineus, ii. 213
Lucinea murinda, ii. 297
Lumnitzera coccinea, ii. 300]
MACROCHLAMYS stephoides, i. 104
Mallotus albus, ii. 297
Manis brachyura, ii. 201
macrura, ii. 201
Megalomastoma sectilabrum, i. 94
Melia excelsa, ii. 215
Meliosma sumatrana, ii. 299
Memecylon coeruleum, ii. 235
paniculatum, ii. 277
Microcos glabra, ii. 216
tomentosa, ii. 215
Microcystis palmicola, i. 105
Microtropis bivalvis, ii. 297
Millingtonia sumatrana, ii. 261
Mimosa jiringa, kaeringa, ii. 216, 296
Monocera ferruginea, ii. 244
petiolata, ii. 243
Morinda polysperma, ii. 229
tetrandra, ii. 229
umbellata, ii. 297
Murraya exotica, ii. 298
paniculata, ii. 238
Myristica glaucescens, ii. 299
NEPHELIUM lappaceum, ii. 214
OCTAS spicata, ii. 278
Olibanum, ii. 261
Oscosperma filamentosa, ii. 302
Ophiorrhiza heterophylla, ii. 289
mungos, ii. 205
Ophites albofuscus, i. 85
subcinctus, i. 85
Opisthopotus penangensis, i. 92
solutus, i. 93
Oxycalamus longiceps, i. Si
PARINARIUM costatum, ii. 301
jackianum, ii. 301
Pentaphragma begonifolium, ii. 296
Peronema canescens, ii. 269
Petrocarya excelsa, ii. 279
sumatrana, ii. 280
Phaleria capitata, ii. 276
Philomycus, i. 116
pictus, i. 118
Phyteuma begonifolium, ii. 212
Pithecolobium bubalinum, ii. 301
clypearia, ii. 301
lobatum, ii. 296, 299
Psilobium nutans, ii. 289
tomentosum, ii. 295
Psychotria malayana, ii. 21 1
Pternandra capitellata, ii. 295
cocrulescens, ii. 277
echinata, ii. 295
Pupa, i. 119
orcella, i. 120
palmira, i. 120
Pupina aureola, i. 94
Pupisoma, i. 119
INDEX OF LATIN TERMS.
39
I'yrrhanthus litoreus, ii. 275
QUERCUS racemosa, ii. 289
spicata, ii. 301
urceolaris, ii. 290
ensiflora, ii. 300
RAN A fusca, i. 76
lymnocharis, i. 77
plicatella, i. 77
porosissima, i. 78
Rauwolfia sumatrana, ii. 233
Rhizophora caryophylloides, ii. 239
Rhodamnia cinerea, ii. 270
trineura, ii. 300
Rhododendron malayanum, ii. 254
Rhopala attenuata, ii. 227
moluccana, ii. 227
ovata, ii. 294
Rhysota cymatium, i. 98
Rondeletia corymbosa, ii. 211
Rottlera alba, ii. 222
Rotula bijuga, i. 101
Rourea simplicifolia, ii. 299
concolor, ii. 300
SALACIA, ii. 293
Sauraiya jackiana, ii. 298
tristata, ii. 298
Sapindus rubiginosus, ii. 214
Saraca declinata, ii. 301
Scyphiphora hydrophyllacea, ii. 297
Simotes bicatenatus, i. 82
cruentatus, i. 82
catenifer, i. 82
Sitala carinifera, i. 103
Sonerila erecta, ii. 225
heterophylla, ii. 254
moluccana, ii. 226
Sorex musaraneus, ii. 203
Sphalanthus confertus, ii. 274
Sphenodesme pentandra, ii. 219
Sterculia angustifolia, ii. 22O
coccinca, ii. 219
kevis, ii. 296
rubiginosa, ii. 296
Storax liquida, ii. 261
Styphelia, ii. 232
TABERN>EMONTANA macrocarpa, ii.
286
Tacca cristata, ii. 234
Tanarius major, i. 261
Ternstroemia acuminata, ii. 259
cuspidata, ii. 260
pentapetala, ii. 242
rubiginosa, ii. 241
serrata, ii. 259
Tetracera arborescens, ii. 244
Tetranthera cordata, ii. 263
Trachia penangensis, i. 112
Trigonostemon, ii. 302
Trimeresurus wagleri, i. 87
Trocomorpha castra, i. 108
timorensis, i. 109
UROSTIGMA ovoideum, ii. 301
Uvaria hirsuta, ii. 245
VACCINIUM sumatranum, ii. 255
Vaginulus birmanicus, i. 121
Veratronia malayana, ii. 298
Veratrum malayanum, ii. 234
Veronicella birmanica, i. 122
Vitex arborea, ii. 218
pubescens, ii. 296
Vitrina nucleata, i. no
Viverra mungos, ii. 204
Vitis racemifera, ii. 294
WORM i A excelsa, ii. 281
pulchella, ii. 281
ZINGIBER gracile, ii. 209
INDEX OF MALAYAN AND CHINESE
TERMS.
Aji, i. 158, 189
Akar charikan, ii. 294
kakatup, ii. 301
sedingka, ii. 266
Aksara g'de, ii. 72, 76
murda, ii. 72, 76
Anak-tumiang, i. 292
Anting-anting, ii. 134
Apus kupak, ii. 126
Aquilaria agallocha. i. 260
Aturan, ii. 132
BABAD, ii. 91
Babadong, ii. 126
Babandong, ii. 134
Babedatti, ii. 126
Babing, i. 294
Bade, ii. 140
Badong, ii. 82
Baju, i. 290
Baju-panjang, i. 290
B'kel, ii. 161, 187
Balai, bale, i. 3 ; ii. 62, 99
Balam pangkat, ii. 301
Bale, ii. 139
Balian, ii. 136
Balik angin, ii. 222, 297
Bambu ribut, i. 294
Banten, ii. 132
dagan, ii. 143, 148
Bapang, ii. 126
Batin, i. 288, 298, 301, 302
Batu-kapala, i. 130
Bayam badak, byumbada, ii. 211, 296
Bayur akar, ii. 294, 302
Begandai, ii. 63
Bela, ii. no, 146
Bcladan, i. 172
Bemban, ii. 66
Beo, i. 174, 206
Berduwi, ii. 63
Biarang, ii. 300
Bindang, i. 4
Biola, i. 294
Buah jring, ii. 216, 296
karbau, ii. 285
manik, ii. 242, 298
Bunga saluang, ii. 288, 301
burutta, ii. 264
kgnchong, ii. 247, 298
yarum, ii. 228
CATI, i. 177
Chaler, ii. 126
Chan-pi, i. 196
Charu, ii. 132
Chechandian, ii. 134
Chiang-chin-hsiang, i. 220
Ch'ien, i. 177
Chingkau, ii. 6l
Chinkani, i. 301
Chintayn, ii. 67
Chitseh, i. 253
Chuundung, ii. 133
gripada, i. 135, 161
DADAUB, ii. 284
Daun saribu, ii. 271
Dinar, i. 210
Dnkun, ii. 136
Durian, i. 209
Duung, ii. IOI, 104
FAN-NYIN-WON, ii. 301
Fuyung, i. 187
GAMBANG, ii. 63
(iandasuli hutan, ii. 246, 298
G'de, ii. 168
G'dong chantel, ii. IO2
tarik, ii. 102
Gc-kuei, i. 246
Ggndaga nasi, ii 298
Glang i. 290
INDEX OF MA LA VAN AND CHINESE TERMS. 31 1
Glang batis, ii. 126
kana, ii. 126
kupak, ii. 134
Glung chandi, ii. 126
kurung, ii. 126
Gomuti, i. 138
Guduha, ii. 104, 125, 134
pawilan^an, ii. 134
Guling buntut, ii. 129'
Gusti, ii. 130, 151
HlPO-BATANG, i. 292
Hsiang-chen hsiang, i. 261
Hwo-siang, i. 240
IDA, ii. 151
Inga jring, ii. 285
Ingor-ingor karbau, ii. 259
ATA, ii. 124
awat, ii. 126
awi pikan, i. 7, 9
ennang, i. 302
ring-jgring tupei, ii. 262, 296
dring munit, muiiet, ii. 286, 299, 301
uru-krah, i. 295, 302
KABAHIA, i. 290
Ka.bu-ka.bu, ii. 301
Kadigjayan, ii. 87
Kahyangan, ii. 101
Kain kasoh, i. 299
Kajang, i. 166, 211, 244, 254
Kalapa puyuh, ii. 212, 296
Kalintang, ii. 63
Kalung, ii. 126
Kamben, ii. 126
Kampulan badan, ii. 137
Kampara, i. 204, 206
Kanda, ii. 81, 84
Kandis, ii. 59
Kan-man, i. 260
Kapas, kapeh, ki-pei, i. 142, 185,
206, 229, 259
Kara, i. 143
K'au-ni, i. 217
Kavi, ii. 73
Kavin, ii. 73
Kayu api-api, ii. 275, 300
balam pangkat, ii. 279
briang, ii. 218
gadis, ii. 236, 298
gagak, ii. 258, 299
kabal, ii. 268, 300
kanchil, ii. 256, 299
sipur, ii. 281
siturun, ii. 283, 301
sumang, ii. 274
Kekavin, ii. 73
KOmuning, ii. 238, 239, 298
Kidung, ii. 77, 91, 137
Kiiat bah 11, ii. 134
Kilin, i. 249
Kladiayer, ii. 220
Kobang, i. 177
Koorkup, ii. 221
Koyan, i. 15, 19
Krd, ii. 61
Kramat, ii. 57
Kranti, i. 294
Kris, i. 292
Kulak, i. 178
Kulang kaya, i. 246
Kulit layu, ii. 214
Kulit nfpis, ii. 235, 296, 298
Kunchah, i. 19
Kuniet, i. 295
Ku-pa, i. 206
Kun-tun-lu-lin, i. 143
Kwan, i. 249
LADANG, ii. 61
Lalakon, ii. 3
Landak, ii. 208
Langsap, i. 174
Larak, i. 176
Layang-layang simpei, ii. 253, 299
L6ban, ii. 218
Lemu, i. 300
Li, i. 128
Liak, ii. 114
Liao-ko, i. 174
Lichi, i. 244
Lineher, ii. 126
Lintangan, ii. 194
Lo-ki-lien, i. 143
Lumut, ii. 60
MANGI-MANGI chengke, ii. 240
Mangul, ii. 266
Manis, ii. 201
Mangku, ii. 130
Marpuyan, ii. 270, 300
Mata p^landok, ii. 297
Mavinten, ii. 132
Mdyin mandrah, i. 1 8
Mayong, i. 1 8
Mdang-sanka, ii. 3C
Mengure glung, ii. 127
M^ngkudii kcchil, ii. 229
Mcnpuyan, ii. 300
Mntada, ii. 232
M^rkasih, ii. 300
Miri batu, ii. 275, 300
Mutiara, i. 143
NATO, i. 178
Nakhoda, i. 246
312 INDEX OF MALA VAN AND CHINESE TERMS.
Nalih, i. 19, 178
Nangka, i. 245
Natar, ii. 128
Nibong, ii. 290
Nipa, i. 211
Nyambri, ii. 63
ORANGkaya, i. 246, 288
piitih, i. 4
Orlong, i. 1 8, 19
PABRISSIAN, ii. 140
Padanda, ii. 99
Palanduk, ii. 41
Panah, ii. 100, 104
Panataran, ii. 101, 104
Pangalasan, ii. 8
Pangaskaran, ii. 140
Pangastanan, ii. 101
Pangeran, i. 224
Panghulu, i. 15
Panluang, ii. 197
Pantun, ii. 13, 35, 49, 63
Panggoling, ii. 201
Papudukan, ii. 126
Pa-ra-man, i. 145
Parang, i. 292
Pararyangan, ii. 101
Parmata, ii. 134
Pasang, ii. 290
Patitis, ii. 126
P'au, i. 150
Pauh, ii. 226
Pavan, i. 300
Pedang, ii. 101, 104
Pglandok kayu, ii. 275
Palawan, ii. 297
Pdning-pening bungkus, ii. 289, 290,
301
Permatang, i. 4, 12
Petek, ii. 285
Petinggi, i. 211
Pi, i. 193, 253
Pien, i. 204
Finding, i. 290
Pinka, i. 140
Pi-pa, i. 174
Po-ho, i. 144
Po-ho, bahara, i. 210
Po-ho-pi-ni, i. 144
Po-lut, i. 261
Prakulit, ii. 122, 131
Pu, i. 1 88
Pfiar amas, ii. 248, 298
lilipan, ii. 234, 297
Pfltat aygr, ii. 297
Futu, ii. 169
RAGA, i. 293
Rambutan, ii. 214, 296
Rokam, rukam, ii. 297
Ronron, ii. 126
Rumbing, ii. 127
Rumput udang-udang, ii. 253, 299
SADKAHYANGAX, ii. 100, 104, 107
Sagung, ii. 180
Sahala, i. 253
Sambau, ii. 297
Sambuk, ii. 101, 104
Samir, ii. 126
Sampat, ii. 126
Sanggar, ii. 101, 134
Sarbacane, i. 292
Sarong, i. 172, 260, 290
Sa-tien, i. 191, 259
Satya, ii. no, 146
Sawah, i. 4 ; ii. 6r
Sawit, ii. 126
Sembu badak, ii. 233, 297
Senenan, i. 176
Sepddas bunga, ii. 258, 299
Seramba, ii. 63
Seribulan, ii. 282, 301
Sheng, i. 178
Shew-chu, i. 2
Siamang, ii. 61
Siburu, ii. 237
Silimpal, ii. 126
Simpai, ii. 61
S'ingo-ingo, ii. 241
Sin-tsai, i. 143
Siturun, ii. 301
S'kar taji, ii. 126
Slendang, i. 172; ii. 63
Sloka, ii. 76, 78
So-fu, i. 260
Subong, ii. 127
Suka branak, ii. 295, 302
Summow, sambau, ii. 226
Sumpitan, i. 292
Sungkei, ii. 269
TABU, i. 292
Tael, i. 177
Tambedana, ii. 126
Tampal badak, ii. 233, 297
Tandak, ii. 122
Tanggal, ii. 197
Tangkal, i. 300
Tarantang, ii. 279, 300
Tayas, ii.*3Or
Tazi, i. 139, 142, 145, 222
Tekan, ii. IQ4
Tfimu, ii. 223, 297
kechil, ii. 224
kunchi, ii. 297
Teratap, ii. 297
INDEX OF MALAYAN AND CHINESE TEA'M
T'ieh, i. 206
Ti-mi, i. 143
Titiran, ii. 68, 149
Todak, ii. 25
Togog, ii. 132
To-lo, i. 193
To-lo-ni, i. 260
Tongkok,
Tou, i. 178
Toya tirtn, ii. IOO
Trang-teja, ii. 127
Tuak, ii. 100, 104, 107
Tugu, ii. 132
Tu-man, i. 260
Tumbak, ii. 100, 104
Tumiang, i. 292
Tutur, ii. 74, 91
UBAS-UBAS, ii. 104
Unka piiteh, i. 287
Unting-unting, ii. 217, 296
Unting-uming bsar, ii. 220
WADAH, ii. 140
Wawalen, ii. 122, 130
Wayang.i. 1 8
VAM-PA, i. 209
ZANGGI, i. 140
END OF VOL. II.
PRINTED BY BALLANTYNE, HANSON AND CO.
LONDON AND ED1NHI 1: II
SECOND SERIES. VOL. IF.
i
o
si >
M
UNIVERSITY OF TORONTO
LIBRARY
Acme Library Card Pocket
Under 1'at. " Kef. Index
Made by LIBRAE! BUREAU
