$Y dt aodl ders

Cee here ty are

Oh P54 Mone dor tert ®

yh ene ae : fi 54 Wh aeelee aah ay RE seh

haf holele!

¥ heh tale E 14™

hee) 4 a skeen | =

Tale baiabe be! in : sh eho?n inal ’ ighe int th eo WO gl per *!

a RESTS FeT aah eee! st boleh 8 she ha Rohe tab ie of Ose erig’ paper sat

» rareletelete Reh Litebal abebahat se’

™ Ronnie poate itt

d SL apeliedtetnt

oie

poe te

We Vohedels lahore tufaie tard Marans hel ote nared

why pte we at Weert al

Mista ee tearees en

or PE part «

Vie *

holatatete tei AoBebetahchen: tec tee ha, cok

header

Pd aa wt ta ty -

lofetel 52h betas bay

Meh ibe ta

si?

pe lyn hig patted tet Uber hb yh

+ pile eh shh aE hy Mri.

pveltepet pre's

4 egg A on

eviahty:

LAs

ete fete : whee a ehe”

a'e let pen Pele

eee otek Leh 4 an Lio ey

Le ee oo

pipte' tole

gates

evils viel aly

ah aioe #

iyi y lene : aig oF lealgiehe™'¢

«4%

ibe)

Lia

tb hewel ateled

WELT ai

Peer

Vie eet ee ete erin

Mistrhert

atale hehehe tulie’

TT chahnbe!

ST he Rothe trtal

1a ee say eta ae ity

ty La hehedel "

thr

hy ek ed ee e

ste Miahadbe tian Catal, ja. ted

rd wy si

wl RV ote * al

Pere fire a

aps

roy

torah, ee

AHisedab eerie

‘alae Wh Hy dr con) a

AEM A Tt rm

va tg ee Arie ew

eee hare

ap Py AE at yer

Peo

rn Oh be ped ee ND he Urey de eh

} da remap ot edly pte te) ae aree be FCEO I rig bey he totes!

Ty Oe Tat che dam Maha ReAb Nahe Solin te telinte Ate Rcae ied Lo,

SENN bey ae Me PaXelie! abel st at href te shat *

‘ Pe a8) f

HE yt ay

Ps aka ahs hs ee

se Pee

4h t

ad Nh vie ey

iad athe he a!

Dial BY aad 9

te nye

with

kth

eats

ccs

bal

gts ofe' ey

Pa Pues Mili SheMale heod

yaa tele tal

Fafa! th ‘tie Roda

Perr LT we

Bi oy

0 ee epetefal

+ hate”

a ur

4 NN Sth) yy tp

a Nedet a Mi Amol ee an eTe

Na Mercht meh be toe

i a ae

Pe |

dkoheh pak htibl hada hhebeee

ee ek Be Oe ee te A

Eg peed leet beep et Ph ee 0@ Oe

th ee) ran plore

Boye

feiy shake

een eb rah We

rie itea

be Petal fares

betuliehehehalatemd beat chem ete!

eye hte M Shy on) a

bape ep reer ain!

9, Shy

pw At

Mae Meahy a

0 te arte

ere ee

, AP mera

ohbte (abe witht

he 1 i a

4 CNY Pw tp oer etl eee

tre pet ow

we cated

rears, | ed

2 BS rie Ores

Pedtreh ge tre? ore

mer ate

shel Gwe

ty Pepe ie eras,

1g RO ferry eer

ican

wy +

a? 9°s miptete | xi

oie bey “4

pere ‘

y \iadeas pedal

oe Baas Aid

“te tetety ee ree

o wtpte She

Lie she ibe

eit aad Seca

"+e

wrebea aby als

Votes elermig” F

a es

veg tem eg iphe tee ba a died

rg @ te Fes ee ae

EE EEO——E—EO———EEOEOEOOEOOEOEOEOOEOE—E——E————————EEOEOEOOEE——————oOl eee

il I il) I

Library

Arnold Arboretum

Harvard University

BYBPYPYPVPY\PVPVUP9 YUU AUAUMM™ AU AV

QDMA AMAMAAMAMAAWAUAAA AAA APAIAAIM

THE

GARDENS BULLETIN

SINGAPORE

VV PYVYUPPYAPP PBA AAYAAUMM UA.

Vol. XVI Ist August, 1958

BAP BPP DA.MA AMAA A229

CONTENTS

HoLttuM, R. E.: Bamboos of Malaya : ’ ; 1

Prowse, G. A.: Eugleninae of Malaya . : : 136

SINCLAR, J.: A Revision of the Malayan Myristicaceae . : 205

To be purchased at the Botanic Gardens, Singapore

Price $25

PVPUPUMPVVPBVUPBUPYUUPUPUPY\PY\PYPV\PYUPVPYPY\PYUPYUPYUPUPU PAUP AMPA UI A”

-AULD ARBOR

RECE! VED =7US \

FEB 12 1959 y)

y a

ee A ee

Published by Authority §

PRINTED = THE GOVERNMENT +s SRiviriac OFFICE, SINGAPORE, §

y A. G. BANFIELD, GOVERNMENT PRINTER §

1958 §

The papers in this volute hase. been ready f

printing for many months. Likewise there are pape

ready for volume 17. The Editor apologises to sub:

cribers and to authors for the prolonged oy F

publishing. This has been caused by reasons entirel

beyond his control and the control of the Goieamal n

Printer. In the present circumstances the publicatic no

of a complete volume at one issue is impracticable. oe

Volumes subsequent to this one will appear ne Gora

separately issued communications totalling a minimum bee tae ;

of 400 pages. This will allow appreciably quicker B eee 4

publication. The papers already in hand will appear af oe

very shortly. ac NS Sm

foe

. 1g b, ria

Re po ee Fee

het SEA

THE

GARDENS’ BULLETIN

SINGAPORE

POA A PUA AYAIAU AAA AUAUAI AAU AIA AAA

VOL. XVI Ist August, 1958

BPP OAYU PAU AUAY AYA AAUAAI AMAA AAA AAA

The Bamboos of the Malay Peninsula

By R. E. HOLTTUM

Scope of the present work

THE BOTANICAL classification of bamboos, as of other members of

the grass family, is based on characters of the inflorescence, flowers

and fruits. Botanists have accordingly described and named flower-

ing specimens. In some cases such botanists never saw the living

plants, and were not provided with information which would have

permitted them to prepare descriptions useful for the field botanist.

A field botanist finds that most of the bamboo plants which he

sees have no flowers. Therefore he often cannot name the plants

by consulting the botanical descriptions. But he finds that by care-

ful study of vegetative organs, particularly of culm-sheaths, he can

learn to recognize different kinds of bamboos with certainty. This

is the kind of knowledge possessed by the Malays who use bam-

boos in many ways every day of their lives.

Thus there are two classes of information about bamboos, and

in order to gain a satisfactory understanding of these plants the

two classes of information must be combined and correlated. The

object of the present work is to combine and correlate such in-

formation of these kinds as is at present available concerning the

bamboos of Malaya. An introduction is needed to explain the

nature of the characters which may be used to recognize and to

describe bamboos, because such a statement is not readily avail-

able elsewhere; and, as will appear in the course of this introduc-

tion, it appears to the present writer that a radical re-examination

of the basic ideas of the botanical classification of bamboos is

long overdue, so that botanically the following statement is an

attempt to formulate revised concepts, not merely a summary of

previous knowledge and opinions. A more detailed statement on

]

Gardens Bulletin, S.

the same subject has been published in the journal Phytomorpho-

logy, Vol. 6, pp 73-90.

The first botanist who made an effective attempt to study the

bamboos of Asia in the field was Kurz, who travelled widely in

Java and later in Burma. He recognized the importance of culm-

sheaths in the recognition of bamboos, and was the first to describe

and illustrate culm-sheaths for diagnostic purposes; his important

published works date from about 1864 to 1877. Gamble extended

the work of Kurz to cover, so far as he could, all the bamboos of

India, Burma and Malaya, and produced his monograph in 1896,

as volume 7 of the Annals of the Royal Botanic Garden, Calcutta.

Gamble knew a great many Indian bamboos in the field, but he

did not visit Malaya, and therefore his descriptions of Malayan

bamboos were made upon the basis of specimens collected by

others, usually lacking culm-sheaths. Ridley later made consider-

able collections of bamboo specimens in many parts of Malaya,

and recognized the existence of several species not known to

Gamble, but his descriptions of these are very unsatisfactory, both

as regards vegetative and flower characters, so that the account of

Malayan bamboos in his Flora, Vol. 5, is not only unreliable but

almost unusable by anyone who has not Ridley’s own specimens

available for study. Backer wrote a most valuable work on the

grasses of Java (Handb. Fl. Jav., part 2), giving detailed descrip-

tions in Dutch of all the bamboos found there; I have found this

work very helpful, and was fortunate in being able to supplement

it by examination of bamboo plants in cultivation in the Botanic

Garden at Bogor in Java in the year 1951.

The account of bamboos in the present work is mainly based on

a study of the specimens and writings of Gamble and Ridley, as

preserved at Singapore and at Kew (also on some earlier works,

especially of Munro), and on my own observations of bamboos

in the field in various parts of Malaya and in cultivation in Singa-

pore. I wish to express my thanks to tke Director of the Royal

Botanic Gardens, Kew, for permission to use the relevant speci-

mens, manuscript notes and publications in the herbarium and

library at Kew, and also to Mr. C. E. Hubbard for placing his

unique knowledge of the family Gramineae at my disposal.

I have attempted to describe the bamboos occurring in Malaya,

as far as they are known at present, in terms of the species con-

cept, based on the work of former authors. But, as will appear, I

am by no means sure of the limits of species in some cases, and [

think it likely that there are some hybrid swarms which cannot be

treated in the conventional way, and which cannot be understood

without a good deal more field work. The present treatment is

Vol. XVI. (1958).

therefore a tentative one, and will need revision. But I hope at

least it will be a useful basis on which further work can be estab-

lished.

To the botanist, bamboos are of great interest in many ways

which have no immediate relation to the everyday usefulness of the

plants. But I believe that the establishment of a better understand-

ing of the inter-relationships of bamboos (i.e., of their basic bota-

nical classification) is of considerable ultimate practical import-

ance. The practical man, who is interested only in discovering the

useful kinds, and in exploiting their particular virtues, may think

that much of the present work is irrelevant for his purpose. But

the first thing a user of bamboos needs is the ability to distinguish

one kind from another; and the second thing is the need of names

which have a definite meaning. This work attempts to supply

these needs, so far as is at present possible, along with its contri-

bution to a more fundamental study of the plants.

Bamboos are used in so many ways by village people in Malaya

that they may be described as a necessity of life. Even in this

mechanical age their usefulness continues, and is likely to continue;

and it may well be that the peculiar properties of bamboos may

even find new uses for modern needs. A survey of the usefulness

of the particular kinds of bamboos found in Malaya is long over-

due; but such a survey must be dependent on a study by which the

various kinds of bamboos can be distinguished. The present work

is a contribution to that necessary study, and I hope it will be a

stimulus to the undertaking of the further necessary practical work,

which may be of great importance to the well-being of many

people in Malaya. The subject is however an intricate one, and its

study cannot be undertaken as a spare-time job; it needs the un-

divided attention of a specialist, considerable equipment, and an

adequate experimental plantation.

A recent publication by F. H. Hildebrand on the use of bamboos

in Java is frequently quoted in the present paper. The full title of

this publication is: Aanteekeningen over Javanese Bambu-sooten

(Rapport van het Bosbouwproefstation, Bogor, no. 66. 1954).

The ulustrations which occur in the text are all original, with

the exception of figs. 9C and 10B (adapted from Arber: Grami-

neae), and fig. 12. The drawings of culm-sheaths were mostly

made by Mr. Chan York Chye, formerly artist at. the Botanic

Gardens, Singapore. For some other drawings of vegetative parts

of plants I am indebted to Mr. J. N. Milsum. All drawings of

spikelets and their parts were made by myself. For permission to

reproduce fig. 12 (originally published in the Kew Bulletin) I am

indebted to H.M. Stationery Office, London.

Gardens Bulletin, S.

The occurrence of Bamboos in Malaya

In Malaya, bamboos are common everywhere, but they never

dominate the landscape. In the extreme north of Kedah and Perlis,

there are small areas of forest consisting almost entirely of bam-

boos, but these are not comparable to the extensive bamboo

forests found in the monsoon region further north. In the equable

climate of Malaya the gregarious flowering of single species of

bamboos in particular areas, at more or less regular intervals

of many years, which is such a striking phenomenon in

India, does not normally occur, and my impression is that the

truly native species of Malaya can go on growing indefinitely,

flowering sporadically from time to time or in some cases flowering

a little almost continuously. After an unusually heavy flowering,

a plant may die back for a time, but I have no evidence that this

occurs gregariously. Notes on the flowering behaviour of indivi-

dual species, so far as known, are given under each species in the

main body of this work.

The bamboos of Malaya can be divided fairly clearly into village

or cultivated bamboos and native or forest bamboos. A few native

species may also be planted, or at least encouraged, in the neigh-

bourhood of villages, but there are certainly some village bamboos

which are not native. The same situation occurs in Java, where the

species of Gigantochloa, well known as village bamboos, are appa-

rently not native. The distribution of Gigantochloa in Malaya

gives some support for this last statement. I have seen no native

plants of that genus in southern Malaya. The first appears in the

neighbourhood of Tampin (at the southern end of the Main

Range) and as one travels northwards the variety of native Gigan-

tochloas increases. In the north of Kedah this variety is quite

bewildering, and I believe that hybrid swarms exist. It is fairly

clear that the main centre of distribution of Gigantochloa lies

north of Malaya, probably in Lower Burma, where no intensive

study of bamboos has been made. So that the Gigantochloas of

Java, and some planted in Malaya, probably came from the

Burma region, and may be interesting records of the migrations of

men; but in the absence of a good knowledge of the bamboos of

Burma those records cannot be interpreted. It is strange that the

most distinctive and most useful Gigantochloas of Java seem

hardly known in Malaya (G. apus and G. maxima).

The native bamboos in Malaya are especially seen on the edges

of forest (beside roads made through the forest, and beside

rivers), on steep hillsides, and in clearings. They are nearly all

light-demanding plants, and cannot grow strongly in the shade of

Vol. XVI. (1958).

high forest. Such bamboos are extensively used, as for example in

the local industry of basket-making (to serve the needs of veget-

able growers) near the road from Tapah to Cameron Highlands.

Local users carefully discriminate the kinds of bamboo best suited

to their needs, whether for house-building (posts, walls and

floors), basket-making, raft-building, food, etc., but there has been

no systematic planting of bamboos of the most useful kinds, nor

any attempt at selection among allied species, such as occurred in

Java, except the plantation of Dendrocalamus asper by Chinese

market gardeners for its edible shoots. There would seem to be

considerable scope for the study of the bamboos native in the

northern part of Malaya, for selection of the best species or varie-

ties (which could be propagated vegetatively) and for the intro-

duction of some other species from neighbouring countries. It is

said for example that Gigantochloa apus is of outstanding value

in Java, but in Malaya it is not a common village bamboo; I have

only seen it at the Federal Experiment Station at Serdang, and at

the Botanic Gardens, Singapore (apparently imported by Javanese

employed in the Gardens).

There are some bamboos which cannot grow effectively in isola-

tion, because their culms are not strong enough to bear the weight

of their branches. Such bamboos only grow in forest, and their

development is encouraged by a little clearing such as has oc-

curred in the making of roads into the mountains; the most conspi-

cuous examples are Dendrocalamus pendulus and Schizostachyum

grande, both very abundant in the foot-hills of the Main Range. A

remarkable species of very restricted distribution, occurring at

higher elevations, is Bambusa wrayi. Young culms of these bam-

boos, being slender and unbranched, grow vertically upwards,

bending later with the weight of lateral branches, which rest on

any trees which may be near enough. There are also a few species

which are even more dependent on the support of trees, and

might be called climbers, namely Dinochloa scandens, Bambusa

paucifiora and the peculiar species I have called Racemobambos

setifera. The growth-habits of these plants have never been ade-

quately studied.

Most bamboos, including all village bamboos and those which

form dense clumps on the edges of forest, have rigidly erect culms,

the upper parts often curving outwards or the slender tips droop-

ing. The culms usually grow quite close together, and only branch

from their upper nodes. There are a few species which branch

from the lower nodes, notably the thorny bamboos, in which these

spreading lower branches bear stiff thorns. The general aspect of

Gardens Bulletin, S.

such a clump of a particular species, like the general aspect of a

tree, is something one learns to recognize without analysing its

individual characteristics; and in fact it is difficult if not impossible

to express clearly in words the peculiarities of habit which distin-

guish different kinds of bamboos. In particular, the precise way in

which the group of branches at each upper node develops, which

must be a characteristic feature of every bamboo, is difficult to

observe clearly and has in fact been little observed. Though the

process will not be easy, I believe such observation may be of

considerable interest.

There are however characters which can be more easily observed

and more clearly defined, and the next section will deal with these

characters. To understand them one must describe in some detail

the way in which a bamboo plant is constructed, and the way in

which it grows. After the description of the vegetative parts of

the plant will come a description and discussion of the flowering

parts.

Vegetative characters of Bamboos

Rébong. A new bamboo culm grows from a bud at the base of

an old culm. This bud grows slowly at first, forming the rudiments

of all the parts of a fully developed culm, as it were in a con-

tracted condition. This slowly developing young shoot is called a

rébong by Malays (fig. 1). It consists essentially of a short mas-

sive little-differentiated stem packed with food-materials and pro-

tected by numerous two-ranked overlapping rigid sheaths. When

this development is complete, the rébong may rest for a time,

until rainy weather provides the stimulus for further growth; but

in Malaya probably such resting often does not occur. Well grown

rébong are cut for use as food (bamboo-shoots).

The more rapid growth of the rébong into a fully developed

culm begins slowly, accelerates until the culm is about half its full

height, and then slackens gradually in rate; the maximum rate of

increase in length is about 30 cm. (one foot) per day in some

bamboos. The growth consists in the elongation of the very short

internodes present in the rébong and the consequent sliding of the

sheaths (which are attached to the whole circumference of each

node) within each other. The sheaths also expand somewhat and

become rigid, so that they can support the internodes, which need

to remain in a soft condition while they are elongating; the final

stage of elongation of an internode is confined to its basal region,

which is entirely enclosed and supported by the sheath when the

upper part of the internode is fully exposed and has begun to

develop its rigid mature condition.

Vol. XVI. (1958).

Rhizome, roots and culm. The basal nodes of the new culm

remain close together, near the surface of the ground (or even

below the surface) and this part of the culm also needs to grow a

little horizontally so that the new culm may be able to grow up

clear of the previous ones. The horizontal part of the new growth

bears roots and is called a rhizome. In some bamboos (e.g., spe-

cies of Arundinaria, not m Malaya) the rhizome continues to

grow horizontally for an indefinite length, the upright culms

formed by lateral buds upon it; but m all Malayan bamboos the

apex of each piece of rhizome soon grows upwards to form a new

culm. The collective rhizome of a Malayan bamboo is thus formed

by the first part of the growth of each new culm, and is said to be

sympodial (fig. 1).

[

——

_—

~—

SF tli,

. eB

{3/

(3% Ey

co == AS ‘

ot)

we aN WN

Ff x S iF . .

ssi NN

ta >

Fig. 1. Bases of three old culms, showing root-bearing rhizome of short

internodes at base of each, and a new shoot (rebong) beginning

growth. Diagrammatic; all sheaths removed except on new shoot.

Gardens Bulletin, S.

Roots begin to form at the nodes of the new piece of rhizome

(breaking through the bases of the sheaths) as soon as the new

culm begins to grow rapidly, and soon serve to provide its needs

from the earth (though the early stages of growth are of course

dependent on the roots on older rhizome-lengths). This sympodial

arrangement ensures that each new culm will have its own new

roots to supply its needs. Sometimes the first few nodes of the

erect part of the culm also bear roots (notably in Dendrocalamus

asper); whether such roots will grow beyond the stage of rudi-

ments probably depends on how much moist organic material

(dead leaves, etc.) is present round the base of the plant.

The young culm grows to its full height before branching. At

this stage the culm-sheaths cover the lower part of each internode,

and the upper part of the internode is exposed between the top of

the sheath and the next node. The exposed parts of the internodes

of young culms are sometimes covered with a waxy powder (how

this develops, and what is its nature, seem to have been little

investigated) and sometimes they bear more or less abundant

stiff hairs, usually appressed, variously coloured. The waxy

powder and the hairs disappear as the culms become old, and can

only be seen clearly on young culms; these characters may be of

considerable importance in recognizing species. The sheaths also

lose their diagnostic characters as they become old; they are des-

cribed in a separate section below.

The maximum diameter of culms, the thickness of their walls

(in the middle of an internode), the length of the longest inter-

nodes, the prominence or otherwise of nodes, are also important

characters; these can best be noted from old culms. Young plants,

or plants growing under poor conditions, have smaller culms than

mature and well-grown plants of the same kind. It is sometimes

not easy to judge whether one is dealing with a young or im-

poverished plant, or with a distinct species which never attains a

larger size.

Culm-sheaths. These are modified leaves, and have the same

parts as leaves, but the parts are differently proportioned and of

somewhat different structure (figs. 2 and 3). The important part

functionally is the sheath, which is large and rigid and protects

the growing culm which develops within it. The sheath is attached

at a node, the width of the base of the sheath being actually

greater than the circumference of the node, so that the two sides

overlap towards the base (the sheath is never tubular). When the

sheath is old, it usually separates from the node, leaving a scar.

The back of the sheath is usually covered with rigid hairs which

Vol. XVI. (1958).

are either appressed or loose; in either case they are easily de-

tached from the old sheaths and are irritant when in contact with

the softer parts of one’s skin, so that the study of sheaths is not

always a pleasant experience. The colour of the hairs is characte-

ristic, and sometimes a waxy powder is produced along with them.

The inside of the sheath is always smooth and shining. The colour

of those parts of young sheaths not covered by hairs is also

characteristic.

Prat |

A “at \ | | i ; : a i fi Qi

Ait! i jt ‘4 i

TL |

| V4 Met Set

Fig. 2. On right, generalized drawing of a culm-sheath in natural position;

the blade is erect, the auricles have bristles, and the ligule a

serrated edge. On left, the same culm-sheath spread out flat, the

upper drawing showing the outer face, the lower drawing the

inner face with complete view of ligule.

The sheaths are always in two opposite ranks, alternating. The

top of a sheath is more or less narrowed, and carries the blade

(called by Gamble the imperfect blade). The line of junction of

sheath with blade is usually quite distinct (in a few cases it is not

detectable externally), and when the blade is old it usually breaks

away more or less cleanly along this line. The line of junction is

sometimes almost horizontal, but more often it is upcurved, the

middle highest.

Gardens Bulletin, S.

Fig. 3. Schizostachyum longispiculatum. On right, top of old leaf-sheath

from which the blade has fallen, showing leaf-scar (dotted) sur-

rounded by callus, with ligule and auricle, x 4; on left, top of

a small culm-sheath, the blade reflexed, its base joined laterally

to the auricle, x 13.

The blades of sheaths at the base of a culm are quite small; on

the sheaths near the top of the culm they are leaf-like and green.

There is thus not one unique shape and size for a culm-sheath

blade on any bamboo. Blades on sheaths at about eye level or a

little higher (say 150-200 cm. above ground) have well developed

blades which are usually rigid and quite different in character

from the ordinary leaves, and it is these blades which are usually

described, because they show the best diagnostic characters. Such

blades have a characteristic shape (broadly or narrowly triangular,

or elongate), texture and colour; they are either erect or spread-

ing or reflexed; they usually have hairs (pale or dark) on their

upper surface (fig. 4).

The blade of a culm-sheath is usually contracted a little just

above the base, and then the base itself spreads out along the top

of the sheath to the full width available, or sometimes beyond the

full width. Thus there are extensions of the blade-base on each side

of the blade proper, and these extensions may be enlarged towards

their outer ends, the ends sometimes spreading and ear-shaped.

These lateral extensions of the base of the blade are called auricles

(little ears) and their characters are very important diagnostically;

their shape, size, colour, and the presence or absence of bristles

on their edges. The auricles are sometimes no more than a low

firm rim to the top of the sheath not occupied by the blade (e.g.,

Vol. XVI. (1958).

im species of Gigantochloa); m other cases, where the triangular

blade occupies the whole of the sheath-top. the auricles are more

or less rounded free lobes on each side of the base of the blade-

The auricles are not deciduous like the blade, but they are offen

fragile and are broken as the sheaths become old, or at least they

lose their bristles, so that old sheaths never show the auricle-

characters satisfactorily.

The ligule is an upgrowth from the top of the sheath on the

face towards the culm. This upgrowth is m contact with the culm.

and goes right across the mside of the base of the blade. The

ligule is always much thinner than the sheath itself, and sometimes

quite fragile. Its height, and whether it is entire or lacerated. or

whether it bears a fringe of fime hairs, are the characters which

need to be observed. The ligule is close to the auricles at its ends,

Fig. 4. Tops of young growing culms. to show differences of shape and

position of blades. A. Schizostachyum zollimgeri (blade rigid.

a —

erect, broad. convex): B. pendulus (

. partly refiexed): C. Gigantockloa scortechinin (

S leaf. spreading 2

Gardens Bulletin, S.

but it is always quite separate from the auricles. Careful observa-

tion is often necessary to distinguish between the bristles on the

auricles and bristles or hairs on the ligule. The ligule can only be

seen clearly on sheaths removed from young culms.

Branching and foliage leaves. Under every sheath on the main

culm is a bud, enclosed in a flat sheath of its own; the sheath

backs on to the culm itself, and has two inflexed edges which em-

brace the inner parts of the bud (fig. 5). Along the two sharp

keels so formed are usually stiff hairs, which may be diagnosti-

cally important; they have been insufficiently studied in Malayan

bamboos. This two-keeled sheath is called a prophyll, and is ac-

tually the first leaf on the branch of which the bud is the rudiment.

Such a two-keeled prophyll is normally present at the base of

every branch in the grass family (and in some other families of

Monocotyledons ).

(()) .

yas

Fig. 5. On left, bud at node on a culm (sheath removed), showing pro-

phyll with inflexed edges enclosing inner sheaths of bud (keels

of prophyll bearing short hairs). On right, Diagram to show

arrangement of branches at node of a culm; all sheaths removed

and lower internodes somewhat elongated. The plane in which

buds lie on any branch is at right angles to that of the branch

which bears it.

When the culm has reached its full height, the main lateral buds

begin to grow and to form branches (they usually remain dormant

at the lower nodes, though always present). Each branch has at

first very short internodes, like the culm itself, and then longer

ones; the thickness of the branch is naturally much less than that

of the culm. Usually one or more secondary branches grow from

the lower nodes of each main lateral branch, and the process

may be repeated on these secondary branches, so that from each

node will appear to come a tuft of small branches (fig. 5). The

larger of these branches may also bear some more small branches

at nodes further from the base. There is a leaf at every node, but

Yol. XVI. (1958).

the lower leaves consist of sheaths only, or have very small

blades; where the internodes are short, the sheaths overlap, and

the stem itself is not seen. The length of the branches, and

whether they spread or droop, may be characteristic features, but

have been little observed.

The ultimate branchlets are leafy, and usually not very long,

bearing perhaps 8—12 leaves. The internodes are shorter than the

leaf-sheaths, usually much shorter. The first few leaves on a

branchlet consist of short sheaths only. The blade of the first

foliage leaf is usually smallest (sometimes short and broad), and

has a broadly rounded base; the uppermost leaves are also often

a little smaller than the middle ones, and have usually a narrowly

cuneate (wedge-shaped) base. Each leaf-blade has a stalk; the

upper leaves usually have the longest stalks. For descriptive pur-

poses, one should take the middle leaves on a branchlet, noting

any peculiarities of the uppermost or lower leaves.

The leaves on the branches from the upper part of a large culm

are usually fairly uniform, and they are always smaller than the

leaves on sucker shoots (i.e., branches growing from nodes near

the ground, and more or less shaded); a young plant may also

have larger leaves than an old one. Thus in one species there may

be a considerable range of leaf-size; and a dried specimen may

not be a representative sample, because it may merely have been

from the branch most accessible to the collector. The top of a

main culm, which becomes gradually thinner upwards, also bears

foliage leaves, and so there is a transition between culm-sheaths

and foliage leaves on one single stem. The transition from a rigid

brown culm-sheath blade to a green leaf-like blade sometimes

occurs quite low down (e.g., in some species of Gigantochloa),

sometimes only quite near the top of the culm (Schizostachyum

zollingeri).

The blades of all leaves on a branchlet lie almost in a single

plane (unless the branchlet stands erect); that is, they are all

twisted at the base through a right angle, the twist occurring in

the stalk. This twisting is often accompanied by asymmetry of the

base of the blade and unequal development of the auricles.

A further asymmetry of the blade is caused by the fact that

bamboo leaves are rolled up while they are developing in the bud.

The blade on one side of the midrib is rolled inwards, that on the

other side of the midrib being wrapped round on the outside. The

inrolled half of the leaf-blade is always a little wider than the

other half, and its main veins are usually more widely spaced; its

edge also is usually almost smooth (except near the tip), whereas

that of the outer half is rough, bearing short stiff oblique teeth

Gardens Bulletin, S.

consisting of single cells with thickened walls; these teeth are like

a fine saw and can easily cut one’s fingers. (The leaves of some

other families of Monocotyledons are similarly rolled in the bud,

and show a similar kind of asymmetry; e.g., members of the

Banana and Ginger families).

Bamboo leaves have longitudinal veins like ordinary grass leaves.

There is always a pronounced midrib, and the rest of the blade

has evenly spaced main veins with several smaller veins between

adjacent main veins. The veins are close together and are usually

raised on the lower surface in dried leaves. In the non-Malayan

genera Phyllostachys and Arundinaria there are distinct cross-

veins uniting the longitudinal veins, exactly as in the Malayan

forest grass Leptaspis. In other bamboos, especially those with the

smaller veins very close together, cross connections between the

veins are less distinct, but always present. They can usually be

seen as translucent spots, more or less elliptical in shape, when a

fresh leaf is held up to the light. Though these cross-connections

are much more distinct in some bamboos than in others, they are

not easy to describe precisely as seen with a hand lens on the sur-

face of a leaf, and so they are not used in the present account.

They can be seen clearly in microscopic preparations, and details

will be given by Dr. C. R. Metcalfe in his publication on the ana-

tomy of the family Gramineae. Gamble described the number of

main veins as a possibly useful character in distinguishing leaves

of some bamboos, but I have not found it easy to use. Further

study of this character is needed.

Just as the blade of a culm-sheath is jointed to the sheath, a

foliage leaf-blade is also so jointed; the joint occurs at the base of

the leaf-stalk and is shown as a scar of distinctive shape when an

old leaf-blade has fallen. As in a culm-sheath also, there are auri-

cles which spread along the top of the sheath on each side of the

base of the leaf-stalk. Owing to the asymmetry of the base of the

leaf, the two auricles are often not equally developed. The auricles

sometimes bear bristles, and their shape and bristliness are some-

times distinctive, but in some bamboos these characters are so

variable (on the same plant) as to be unreliable for diagnostic

purposes. The sides of the top of the leaf-sheath, where the auri-

cles end, are connected across the front of the base of the leaf-

stalk by the ligule. This is usually quite small in bamboo leaves,

but is large and distinctive in a few species. Where the ligule is

tall, the base-line of the auricles is upcurved to meet it (as in

Gigantochloa ligulata, fig. 33).

A feature not seen in culm-sheaths is shown by the tops of the

sheaths of bamboo foliage leaves. The top of the sheath, where it

Vol. XVI. (1958).

meets the leaf-stalk, is produced upwards into a rim, which is here

called a callus. The callus is low in the middle, and each half is

more or less raised and rounded, so that the callus may look like

a pair of auricles. The edge of the callus may be hairy or even

bristly, but I have not found these to be useful characters. The

only species in which the callus is sufficiently distinctive to be

diagnostic is Gigantochloa ligulata, in which one half of the callus

may be considerably elongated and quite thin, more like a ligule

in texture.

In the transition from culm-sheath blade to foliage leaf-blade,

the base of the blade becomes narrower until it becomes a stalk;

at the same time its direct connection with the auricles is broken,

and the raised callus at the top of the sheath, below the leaf-stalk,

is developed. The auricles always lie between the callus and the

sides of the sheath; they are in close contact with the callus but

quite distinct from it (fig. 3).

The Inflorescence of Bamboos and Grasses

Throughout the grass family, the individual flowers are very

small (they are often called florets) and are arranged in spikelets,

each spikelet protected when young by small empty outer (later

basal) bracts called glumes. It is generally considered that the

primitive grass spikelet contained several flowers (fig. 6A), but in

some existing grasses, and also in some bamboos, each spikelet

contains only one perfect flower, sometimes also one or more im-

perfect flowers. The unit of a grass inflorescence is thus a spikelet,

not a single flower. The arrangement of the spikelets on the bran-

ches which bear them varies in different kinds of grasses; all

arrangements can ultimately be explained as reductions from the

condition in which the spikelets are borne in a panicle (fig. 7A);

that is, the main axis of the inflorescence has branches, and these

branches again bear secondary (or even tertiary) branches which

end in spikelets (the main axis itself and the main branches also

each end im a single spikelet). A remarkable feature is that in

most grasses there are no bracts where the branches of the panicle

spring from the main axis; such bracts do however occur in a few

grasses, and they are very small. The bamboo genus Arundinaria,

which does not occur in Malaya, has inflorescences in the form of

panicles in which the branches are in the axils of small bracts, so

that in inflorescence-form there is no sharp distinction between

Arundinaria and some of the true grasses.

Gardens Bulletin, S.

In all the bamboo genera which occur in Malaya, except the

peculiar and apparently rare Racemobambos, the arrangement of

the spikelets is different. The only person who has described this

arrangement is McClure (Journ. Wash. Acad. Sci. 24: 541-548.

1934). But nobody appears to have considered how this arrange-

ment is related to that of Arundinaria. Bamboos are generally

Rig. 6. A, diagram of grass spikelet, with 2 empty glumes and six lemmas

each containing a flower. B, longitudinal section of A, floral

parts omitted. C, primary flowering branchlet of Bambusa, having

3 bud-bearing bracts, one empty glume, six lemmas containing

flowers, and a rudimentary terminal flower. D, section of C,

floral parts omitted. E, Bambusa heterostachya, primary flower-

ing branchlet showing prophyll backing axis at base, X mio,

face view of E, a later stage showing two secondary branchlets.

G, primary flowering branchlet of Dendrocalamus. H, section of

G (compare with D; in both cases axis of branch has been

turned through a right angle).

Vol. XVI. (1958).

regarded as the most primitive members of the grass family, and

Arundinaria, connecting bamboos to the main mass of true grasses,

as the least primitive among the bamboos. One should logically

therefore regard the other tropical bamboos (Bambusa itself,

Schizostachyum, etc.) as the most primitive genera; and one

ought therefore to see how the Arundinaria type of inflorescence

can have arisen from the Bambusa type (or how both originated

from some more primitive type). I propose to attempt to do this

in the following discussion, and will begin with the inflorescence

in Bambusa, afterwards discussing possible transitions between

this and the inflorescence in Arundinaria; finally we shall come to

a discussion of the individual flower and its parts.

Fig. 7. A, diagram of small panicle of spikelets of a grass (outlines of

spikelets dotted); note absence of bracts at branchings. B, Diag-

ram of spikelet-tuft of Bambusa, with all internodes elongated

and all spikelets in the same plane.

A young branch of Bambusa which is to bear flowers has a bud

in the axil of each sheath (the blades on the sheaths are absent or

much reduced); the sheath usually falls as the bud develops. The

bud grows out into a short primary branchlet with very short inter-

nodes, a small bract at every node, and a two-keeled prophyll

backing the parent branch at the base (fig. 6C, 6E). The lowest

bract is quite small, then a few others are of gradually increasing

size, and the rest are all alike. The lowest bracts have buds in their

axils, as have the lowest leaf-sheaths on a vegetative branch, and

these buds soon grow out to form short branches like the first one

(fig. 6D, 6F); these secondary short branches behave in the same:

way and each may produce a few tertiary short branches. Thus.

Gardens Bulletin, S.

develops a tuft of short bract-covered branches, the pattern of the

whole being exactly like that of the branching at a node on the

main culm. '

Each of these short branches has flowers in the axils of the

upper bracts, which are called lemmas. Each flower is enfolded by

a second bract called a palea. Between the lemmas and the basal

bud-bearing bracts are usually one or two empty bracts (on secon-

dary and tertiary branches there may be no bud-bearing bracts).

‘These empty bracts correspond to the glumes of ordinary grasses;

but it will be noted that in bamboos there is nothing to distinguish

a glume from a bud-bearing bract or from a flower-bearing bract

(lemma) except the fact that it is empty. The end of the short

flowering branch clearly corresponds to a grass spikelet; but where

does the spikelet begin?

One supposes that the primitive bamboo had several flowers at

the end of each of the short flowering branches (this is the condi-

tion of Bambusa) and it is not surprising to find that the upper-

most of such flowers is often imperfect. There are other bamboos

in which there are very few flowers, or even only one, on each

short branch; one must suppose that these have evolved from

those with several flowers by a process of reduction.

In Bambusa, the flower-bearing axis of the short branch (i.e., of

the spikelet) is jointed below every node. This axis is called the

rachilla. There may also be a joint below one or more of the empty

bracts below the flower-bearing part. In grasses, the position of

such joints is very precise, and is uniform in each genus or group

of genera. In Bambusa and other Malayan bamboo genera there

is much less regularity, and the question needs further study.

Another kind of reduction can also occur in bamboo spikelets,

as compared with the condition of Bambusa. In this case the

rachilla is very short and is not jointed below the nodes, which are

sometimes only about 0-5 mm. apart. In such a spikelet there may

be several flowers. If the lemmas were all alike, their tips would

all be very close together; but along with the shortening of the

rachilla an arrangement has evolved whereby the outer (lower)

lemmas are shorter than the upper ones, and the floral parts also

(fig. 6G, 6H). Thus a spikelet of Dendrocalamus does not look very

different from some spikelets of Bambusa until one comes to

dissect it.

Now how can we rationalize a comparison between the group of

short flowering branches at one node found in a bamboo with a

panicle of spikelets as found in a grass? If all the lower internodes

of a tuft of short flowering branches in the bamboo were much

elongated, as far as the empty glumes in each case, we should have

Vol. XVI. (1958).

something very like a panicle of spikelets, each spikelet on a stalk

and either terminal on a branch or in the axil of a bract with a

prophyll immediately above the bract (fig. 7B). If we eliminate

these branch-bearing bracts and the prophylls adjacent to them,

and reduce the empty glumes at the base of each spikelet to two,

we have exactly the condition of a grass panicle (there are differ-

ences in the floral parts of bamboos and grasses also, and these

will be considered later).

I suggest that in the original bamboo (which was also the ori-

ginal grass) groups of flower-bearing branches occurred at the

ends of leafy branches (as in some species of Schizostachyum),

instead of at the nodes of a leafless branch. This would icad

straight to a grass panicle at the end of a leafy branch, by way of

the condition found in Arundinaria. By reducing the number of

branches and putting each spikelet on a short stalk, we could also

reach the condition of Racemobambos; and by eliminating every-

thing except the terminal spikelet we reach the condition of Schi-

zostachyum terminale.

Part of the evolution of bamboos has consisted in the evolution

of the large hollow woody culm, with its many branches and rela-

tively small leaves. I suggest that along with this vegetative evolu-

tion has gone the development of small groups of reduced flower-

ing branches at each node. But the basic floral structure must have

been evolved at a very early stage, and represents almost certainly

a reduction of floral parts. The most primitive (tropical) bamboos:

probably retain most nearly the floral pattern of the original plants.

from which the whole grass family evolved. We will next consider

the details of this structure.

Spikelet and Flower structure, apart from the ovary

As explained above, the primary flower-bearing branch of a

bamboo has at its base small bracts with buds in their axils, then

empty glumes, and then lemmas having flowers in their axils. The

part from the glumes onwards corresponds to a true spikelet; the

whole has been termed a pseudo-spikelet by McClure. The buds in

the axils of the lower bracts on a pseudo-spikelet may grow out

into pseudo-spikelets much like the first one, or some of them

only into true spikelets. We are here concerned with true spikelets,

and with floral parts found in the axils of their lemmas.

In many bamboos, it is not at all easy to observe just how many

empty glumes are present; and as may be understood from the

above remarks the number of such glumes on branches of different

orders may not be constant. In some bamboos there may be a

joint below the true glumes, in others only above them. Statements

Gardens Bulletin, S.

about the glumes of bamboos are thus often unreliable and the

whole subject needs critical study. We will ignore them in the

further discussion in this section, and deal only with the rachilla,

the lemmas, and the floral parts.

The rachilla, where the internodes are not very short, is usually

more or less zig-zag, and the shape and hairiness of its internodes

may be important. In Bambusa the last flower, which terminates

the spikelet, is reduced, both as regards the lemma itself and its

fioral parts, but the rachilla-internode which bears it is not mark-

edly different from the other internodes, except that it may be

shorter (fig. 8A). In Schizostachyum the rachilla-internode which

bears the reduced terminal fiower is much more slender than the

rest; and in some cases the reduced flower is so rudimentary as to

be hardly recognizable, or even the whole rachilla-internode above

the uppermost perfect flower may be completely aborted. In some

species of Dendrocalamus the uppermost flower is perfect, and

ends the spikelet. In Gigantochloa the terminal flower is always

represented by a long narrow lemma and nothing else. There is

no doubt that the way in which the spikelet terminates is often an

important character in the discrimination of relationship among

bamboos.

A B G D

Fig. 8. Bambusa ridleyi. A, top of spikelet, showing terminal imperfect

flower, and uppermost perfect flower (enclosed by lemma ) and

rachilla, x 3/2. B, a lower internode of the rachilla, with one

lemma (pulled back a little) and its palea. C, back of a palea,

showing veins and thickened fringed keels, x 23. D, front of

same palea, showing infolded edges.

Each bamboo lemma is a short sheathing bract, comparable to

the sheath of a leaf, usually without any indication of a blade,

though it may terminate in a short stiff (almost thorn-like) point

Vol. XVI. (1958).

which presumably represents a reduced blade. There are always

many veins in a lemma (this is a difference from grasses) and

usually some cross-veins; the midrib is distinct. There may be

hairs on the edge or back of a lemma, and these are often of

diagnostic importance.

When one removes a lemma one finds within it a two-keeled

prophyll which backs on to the axis of the spikelet and the next

higher lemma on the same side (fig. 8B). This prophyll is called

a palea. It is of exactly the same nature as the prophyll at the base

of every branch in a bamboo plant, but more delicate in structure

than most such prophylls. In cases where paleas are exposed when

a spikelet is mature (as in those species of Bambusa and Schizo-

stachyum with long rachilla-internodes) the palea is of a much

tougher nature than in cases (e.g. Gigantochloa) where it is totally

protected by the lemma. The two keels of a palea (as of other

prophylls) are usually fringed with hairs; the back and the edges

may be hairy also. The parts of the palea between the keels and

edges are infolded and embrace the true floral parts. There are

always some veins between the keels of the palea (absent in the

paleas of grasses), and more between each keel and edge (fig.

8C, 8D).

In the case where a perfect flower terminates the spikelet (as in

some species of Dendrocalamus) the palea of that flower lacks

the keels, or almost lacks them. In the case of Schizostachyum

(fig. 11A) where there is a very slender last internode of the

rachilla, bearing a much reduced final flower, the palea of the

uppermost (usually sole) perfect flower is very slightly two-keeled,

but retains a doubly pointed apex. The absence of the two keels

in such cases is probably to be explained simply by the fact that

there is no axis nor any further flower behind the palea to exert

pressure during development; a normal palea is the shape of the

space allowed for it by the development of the other parts of the

spikelet, its back being hollow to fit those parts, its edges sharply

infolded also to fit them.

The perianth (sepals and petals) of an ordinary flower are re-

presented in most bamboos by three small members called lodi-

cules. They are usually largest in the genus Schizostachyum, but

also comparatively large in Bambusa ridleyi. As may be seen in

some of the transverse sections drawn by Dr. Agnes Arber (The

Gramineae, fig. 31 p. 90; fig. 38 B1 and D2, p. 115) the lodicules

in a Schizostachyum flower are symmetrically placed in the position

one would expect three petals to occupy (fig. 9C). The lodicules

are always delicate structures without any fibrous cells. The basal

part of lodicules is rather thick and the swelling cells in this

Gardens Bulletin, S.

region in some cases at least cause a separation of lemma and

palea sufficient for the emergence of the developing stamens and

stigmas. In Bambusa the lower part of two of the lodicules is

sometimes very much swollen (fig. 9A), in Schizostachyum slight-

ly (fig. 9B). The distal part of lodicules is usually thin and

translucent, often fringed with hairs. The lodicules of grasses often

consist of the fleshy basal part only. In most bamboos there is

some inequality in the lodicules, two being alike (and sometimes

asymmetric, one more or less the mirror image of the other) and

the third smaller, symmetrical. In some species of Schizostachyum

there are transitions between stamens and lodicules, in the same

kind of way that there are transitions between petals and stamens

in many other more normal flowers; this is part of the evidence

for the petal-nature of lodicules. In the genus Ochlandra (Southern

India and Ceylon), closely related to Schizostachyum, there are

often more than three lodicules; this is connected with the fact

that there are also more than six stamens. In most species of

Gigantochloa and Dendrocalamus there are no lodicules, and they

are also sometimes lacking in species of other genera.

Fig. 9. A, Bambusa ridleyi, one of the two larger lodicules, and the single

smaller one, * 5. B, Schizostachyum brachycladum, one lodi-

cule, x 6. C, transverse section of floret of Cephalostachyum

virgatum, showing palea enfolding 3 equal and symmetrically

placed lodicules, 6 stamen-filaments, and top of ovary (after

Arber).

The usual number of stamens in bamboo flowers is six. In

Arundinaria and most grasses there are three stamens, this being

considered as a reduction from six and a further indication of the

Vol. XVI. (1958).

more highly specialized nature of Arundinaria. In some species of

Schizostachyum there are transitions from stamens to lodicules,

and then there may be only 4 or 5 stamens, the other one or two

being wholly or partly transformed into lodicules.

The anthers are always long, and their length when mature may

be a useful diagnostic character, but im the cases of Gigantochloa

and Dendrocalamus, where the lower paleas are much shorter

than the upper ones on the same spikelet, one needs to observe

the length of the anthers in the uppermost perfect flowers. The

connective of the anther is often prolonged at the apex to form a

sterile tip projecting beyond the ends of the pollen-sacs; this sterile

apex may bear hairs or papillae (fig. 101). In some species the

extent of development of the anther-tips is variable (fig. 10F, G,

H).

Fig. 10. A, Bambusa ridleyi, ovary with style and part of stigmas, x 10.

B, Gigantochloa scortechinii, transverse section of ovary, show-

ing position of ovule and vascular strands (after Arber). C,

diagrammatic longitudinal section of ovary of Bambusa. D,

same, after fertilization. E, diagrammatic longitudinal section of

fruit of Bambusa. F, Bambusa heterostachya, tips of 2 anthers.

G, Bambusa klossii, tip of anther. H, B. ridleyi, same. I, Gigan-

tochloa wrayi, same.

The filaments of bamboo stamens, as of grass stamens, are at

first very much shorter than the anthers, and elongate rapidiy

when the flower opens to become delicate threads longer than the

Gardens Bulletin, S.

anthers. In some bamboos the filaments are not separate, but col-

lectively form a tube which when extended surrounds the ovary

and style, or sometimes two or three filaments are joined. This

fusion of filaments (or rather, the development of a tube in place

of the separate filaments) occurs in several bamboo genera which

in my opinion are not closely related. Mlle A. Camus has proposed

a classification of bamboos in which such genera are placed in a

distinct sub-tribe (Arch. Mus. Hist. Nat. Paris V1, 12: 601. 1935).

This seems to me a quite unnatural arrangement; it involves taking

some species of what I would call Schizostachyum and putting

them into a different sub-tribe from the other members of the

genus, and alongside Gigantochloa which has an ovary and a

spikelet of quite different structure. In some species (especially of

Dendrocalamus) it is very difficult to decide whether the expanded

filaments form a tube or not, because they are so delicate; in other

cases (e.g., Gigantochloa) the tube is much firmer and quite easy

to observe.

The Ovary and Fruit in Bamboos

The ovary in bamboos, as in grasses, is unilocular, but it con-

tains clear evidence of its derivation from the typical tricarpellary

ovary shown in many families of Monocotyledons. Dr. Arber (The

Gramineae, p. 122, fig. 45 B1) shows the existence of vascular

bundles corresponding to three carpel midribs, and also the fused

marginal veins of the carpels. The single ovule is joined to one of

these marginal veins, the vascular tissue in this case being much

enlarged to supply the needs of the developing seed (fig. 10B).

As regards the external form of the ovary, there are two prin-

cipal types among Malayan bamboos. In the Bambusa type of

ovary the top of the ovary-wall is thickened and often forms a

(usually hairy) swelling wider than the hollow basal part of the

ovary; the three stigmas are either borne directly by this swollen

top of the ovary, or there may be a style of varying length (fig.

10A). In most species of Gigantochloa and Dendrocalamus the

style is continued into a single stigma.

In Schizostachyum (also in Ochlandra and Melocanna outside

Malaya) a very different type of ovary occurs, and its detailed

structure needs examination from microscopic preparations. The

ovary is continued upwards into a gradually tapering, rigid struc-

ture which is a little longer than the palea and bears three short

spreading stigmas at its apex (fig. 11C). The cavity of the ovary

is apparently quite small and basal, as in Bambusa. The upward

extension appears to be comparable to the swollen fleshy top of

Vol. XVI. (1958).

the Bambusa ovary; it is in effect a style, and I have so called it

as others have done, but there is no external mark of separation

between ovary and style. At the time of flowering, this rigid style

appears to contain a free central strand of tissue which connects

the stigmas with the ovary proper, and I have seen nothing like

this in other bamboos (fig. 11D). Dinochloa has such very short

spikelets that the ovary with its style are also necessarily very

short, and a comparison between this ovary with that of other

bamboos is not easy to make. I believe that the ovary in Dino-

chloa is similar to that of Bambusa, not to Schizostachyum, next to

which Dinochloa has usually been placed.

A B Cc

Fig. 11. A-D, F, Schizostachyum brachycladum. A, upper perfect floret and

Tudimentary floret on slender extension of rachilla; back of

palea slightly grooved near tip, x 4. B, lower perfect floret with

tachilla-internode; palea grooved from top to bottom. C, ovary

and stigmas. D, diagrammatic longitudinal section of base of

ovary. F, diagrammatic longitudinal section of a pseudo-spike-

let, for comparison with fig. 6 D & H (only one perfect floret).

E, ripe fruit of S. gracile, enclosed by lemma and palea, x 2.

As an ovary develops into a fruit, it is naturally the portion

containing the ovule which enlarges most. The various structures

formed by the top of the ovary, which serve to bear the stigmas,

are usually only slightly enlarged (fig. 10C, D, E). As the cavity

of the developing fruit enlarges, the attachment of the developing

seed to the ovary wall (i.e., the hilum) elongates, so that the

hilum of a ripe seed extends almost throughout its length. The

embryo of a ripe seed is usually small and round, at the base of

Gardens Bulletin, S.

the seed on the side opposite to the hilum; the rest of the seed

consists of endosperm. The whole seed is elongate and more or

less cylindrical to narrowly fusiform in shape. The exceptions to

this rule do not occur in Malaya; they are the bamboos with large

fruits of the genera Melocanna, Ochlandra, Melocalamus, and

Dinochloa. In these fruits the pericarp becomes thick and fleshy

throughout. In the only species examined in detail, Melocanna

bambusoides, the scutellum of the embryo grows at the expense

of the endosperm until it fills almost the whole seed. A cursory

examination of a fruit of Melocalamus indicates a similar struc-

ture; the fruit of Ochlandra has never been investigated. In Melo-

canna and Melocalamus the seed germinates while the fruit is still

attached to the parent plant; the growth of the scutellum is a

consequence of this fact.

The upper part of the pericarp in all Malayan bamboos is dis-

tinctly thickened and is free from the top of the seed. The pericarp

(ovary wall) becomes thinner towards the base, so that the sides

of the seed in the ripe fruit have quite a thin covering, through

which the position of the embryo can usually be seen when the

fruit is dry and slightly shrunken. In my experience, it is usually

possible to separate the thin lower parts of the pericarp from the

seed, as a sort of skin, in a freshly gathered fruit which has

reached its full size but has not hardened. Munro’s proposed divi-

sion of bamboos into those with a fleshy pericarp and those with

a thin adherent pericarp, with Dendrocalamus in the former divi-

sion and Bambusa in the latter, is thus quite unjustifiable, though

it has been repeatedly copied for nearly a century. I can see no

essential difference in structure between the fruits of Dendrocala-

mus and Bambusa.

Munro saw few fruits, the only one in Dendrocalamus being of

D. strictus. Had he seen the fruits of other species which are clearly

allied to D. strictus (e.g., D. pendulus Ridl.), he would have

realized, as Gamble must have done, that the lower part of the

pericarp in most Dendrocalamus species is quite thin, so that the

position of the embryo shows externally, a fact indicated in several

of Gamble’s figures. But Gamble seems to have been determined

not to alter the scheme of classification of Bentham in Genera

Plantarum, a scheme copied almost completely from Munro,

though Gamble certainly realized that his further knowledge, of

more than one kind, invalidated that scheme (the generic descrip-

tions of Gamble do not is all cases agree with the statements about

those genera in the keys to sub-tribes and to genera, which he

Vol. XVI. (1958).

took from Bentham). I wrote a note calling attention to this un-

satisfactory state of affairs, and suggesting a tentative new scheme

of classification, in 1946 (Journ. Arn. Arb. 27: 340-346). Since

writing that note, I have come to the conclusion that Bambusa,

Gigantochloa and Dendrocalamus are more nearly related than I

realized in 1946; that the position of Dinochloa is probably wrong;

that Oxytenanthera is a distinct genus, wholly African, differing

in the top of its ovary from Dendrocalamus; and that the sug-

gested distinction of “Oxytenanthera” (i.e., of Asiatic species re-

ferred by Gamble and Ridley to that genus) from Dendrocalamus

is probably not valid.

As above noted, the genus Dinochloa is peculiar, and I have

always felt doubtful of its position along with Schizostachyum.

Fruits of D. scandens in Java, and of D. andamanica Kurz have

the pericarp quite thin towards the base, thickened a little and

beaked at the top, quite as in some species of Bambusa; the em-

bryo is larger than usual in proportion to the size of the seed, but

I see no fundamental difference. A species of Dinochloa from

Borneo and the Philippines has however spherical fruits with

fleshy pericarp. The aspect of the spikelets in some Dinochloas is

very like those of the slender subscandent Bambusa species of

Malayan mountains. I am inclined to put Dinochloa near Bam-

busa, not near Schizostachyum, but a further study of fruits is

needed for a better understanding of the genus.

Scheme of Classification of Malayan Bamboos

Top of ovary narrowed gradually upwards

to form a rigid hollow style longer than

the palea at flowering, the stigmas di-

verging from its top = .. 1. Schizostachyum

Top of ovary more or less fleshy, not rigid

nor hollow, variously shaped, bearing

the three stigmas directly, or more or

less abruptly attenuate to a slender

style which bears 1-3 stigmas

Spikelets forming the distal parts of short

branches which are borne in close

tufts at the nodes of a stem

Rachilla-internodes distinctly elongate,

with a joint below the lemma-at-

tachment at each node

Paleas all alike, not deeply bilobed 2. Bambusa

Gardens Bulletin, S.

Lowest palea deeply bifid .. 3. Thyrsostachys

Rachilla-internodes not elongate, not

jointed below each lemma if seve-

ral flowers present, or flower single

without an extension of the rachilla

beyond it

Spikelets less than 5 mm. long, with

one flower; ovary-top gradually

attenuate and bearing three

stigmas ay: .. 4. Dinochloa

Spikelets more than 5 mm. long,

usually with more than one

flower; ovary-top gradually atte-

nuate to a hairy style which

usually bears one stigma

Uppermost or sole perfect flower

with unkeeled (or incom-

pletely 2-keeled) palea, pa-

leae of all other perfect

flowers 2-keeled; a terminal

short imperfect flower present

or not : . 5. Dendrocalamus

All perfect flowers with 2-keeled

paleae; a terminal imperfect

flower consisting of a long

narrow lemma (as long as

uppermost normal lemma)

always present .. .. 6. Gigantochloa

Spikelets borne singly in a short raceme at

the end of a leafy branch .. 7. Racemobambos

General key to Peninsula bamboos based on vegetative characters

Culms bearing many spreading thorny branches at lower nodes

Bambusa

Culms usually without branches at lower nodes; such branches, if

present, not thorny

Exposed parts of internodes of young culms bearing throughout

many appressed pale hairs, lacking white waxy powder

Auricles forming a low firm rim (dark green when young)

along the top of the sheath on each side of the blade;

bristles on auricle-edge scattered .. Gigantochloa apus

Vol. XVI. (1958).

Auricles of different shape, bristles always close and numer-

Gmmeaeme ther edges ; ./.. 2). diwiee sss Schizostachyum

Exposed parts of internodes of young culms quite smooth, or

covered with white waxy powder; or if hairy, the hairs near

top of internodes only, not spread evenly throughout

Fully developed culms not strong enough to bear the weight

of their branches, more or less climbing in habit, curved

over from base if not supported by surrounding trees

Longest internodes 200 cm. long ...... Bambusa wrayi

Longest internodes much shorter .

Young culm-sheaths bearing much white waxy powder

among the felted pale hairs; young culm itself

usually covered with white waxy powder

Dendrocalamus

Young culm-sheaths and young culms lacking waxy

powder

Bases of fallen culm-sheaths persisting as leathery rims

of uneven width; young unbranched culms weak,

sometimes trailing on ground........ Dinochloa

Culm-sheaths falling to leave an even scar; unbranched

young culms usually rigid and erect

Leaf-auricles 4-10 mm. tall, taller than wide, bear-

ing bristles to 20 mm. long... . Bambusa ridleyi

Leaf-auricles otherwise, usually much smaller

Culms only about 1 cm. diameter (?); on moun-

tains in Johore (culm-sheaths unknown)

Racemobambos

Culms stouter; on mountains of Main Range and

Meru Of Wialave: oa Se ik Bambusa

Fully developed branch-bearing culms rigidly erect near the

base at least, usually almost erect for greater part of

their height

Lowest internodes of old culms persistently covered with

close felt of fine brown hairs; commonly cultivated

Dendrocalamus asper

Lowest internodes of old culms not so covered

Leaf-auricles 4-10 mm. tall, taller than wide, bearing

| bristles to 20 mm. long ........ Bambusa ridleyi

Leaf-auricles otherwise, usually much smaller

Slender bamboos on limestone rocks in the north

Dendrocalamus

Gardens Bulletin, S.

Bamboos in other parts of Malaya

Culm-sheath blades on middle sheaths quite erect,

or spreading only slightly

Culm-sheath ligule 10-25 mm. tall including its

fringe of. bristles; auricle a low firm rim,

bristly OP MOE.fs.)3.)... <). cee Gigantochloa

Culm-sheath ligule shorter; in most cases auricle

otherwise

Cuim-sheath ligule not over 5 mm. tall, its edge

entire or bearing fine short hairs

Culm-sheath auricles forming a low firm rim

along top of sheath each side of blade

G. ridleyi

Culm-sheath auricles otherwise

Culm-sheath auricles not over 4 mm. tall

and wide, often much smaller, usually

bristly

Successive culm-internodes forming a zig-

zag pattern .. Bambusa ventricosa

Successive culm-internodes in an almost

straight line

Hedge bamboo; culms to 2 cm. dia-

meter; leaf-blade velvet hairy and

glaucous beneath

Bambusa glaucescens

Larger bamboos; leaves not glaucous

beneath

Old culm-sheaths very persistent,

thin and papery; leaves com-

monly to 12 cm. long and 8

mm. wide, glabrous beneath

Thyrsostachys siamensis

Old culm-sheaths not thin and

papery; leaves commonly

larger, more or less_ hairy

beneath |

Dendrocalamus strictus —

Culm-sheath auricles much larger

Bambusa

Culm-sheath ligule (on well-grown plants) over

5 mm. tall, its edge bearing tapering bris-

tles . skau een GE. Bambusa heterostachya —

Vol.. XVI. (1958).

Culm-sheath blades on middle sheaths spreading or

reflexed

Slender bamboo of high mountain ridges on Main

Range; culm-sheath blade quite reflexed,

commonly 10 cm. long and 3 mm. wide

Bambusa magica

Lowland bamboos, native or introduced

Very large bamboo, culms 18-25 cm. diameter

near base .... Dendrocalamus giganteus

Smaller bamboos

Culm or sheaths (or both) when young bear-

ing white waxy powder

Culm-sheath blades of middle sheaths

green and leaf-like

Gigantochloa scortechinii

Culm-sheath blades of middle sheaths

rigid, pinkish, not leaf-like

Dendrocalamus

Culm and sheaths when young lacking a

white waxy powder .... Gigantochioa

SCHIZOSTACHYUM

Schizostachyum Nees, Agrost. Bras. 534. 1829.

Type species: S. blumii Nees, l.c., from Java. McClure in

Blumea 2: 86-94. 1936.

Culms short to tall, the top usually very slender and drooping;

walls (except in the non-Malayan S. caudatum) relatively thin;

surface of young culms bearing closely appressed white hairs;

branches usually rather short and densely tufted at the nodes.

Culm-sheaths with broad or narrow erect or reflexed blade; auri-

cles small or large, always with slender curved bristles; ligule

often with a short fringe of hairs or slender bristles, never with

long bristles. Spikelets in groups at the distal nodes of a branch

the base of which is leafy, or on wholly leafless branches, in S.

terminale solitary and terminal on leafy branches; each spikelet

with a few bud-bearing bracts below it, and no true glumes be-

tween these and the first (or only) floret, at the base of which the

rachilla is jointed (fig. 11F). Perfect florets one or more, the

rachilla jointed at the base of each and produced as a slender ex-

tension beyond the uppermost to bear a sterile (often rudi-

mentary) floret at its tip (this extension usually lacking in S.

Gardens Bulletin, S.

jaculans, and in some allied species not in Malaya). Lemma of

sole floret, or of uppermost perfect floret, tightly rolled round the

palea, with a short spine-like tip; of lower florets (where present)

rather loosely enclosing the palea. Palea similar to the lemma in

texture, usually longer than its lemma, its apex with 2 points close

together, the points at the ends of more or less developed keels

with a narrow groove between them which may extend to the

base of the palea, the rachilla (or its extension) in the groove

(fig. 11A, B). Stamens usually 6 (4 in S. insulare), filaments

usually free; in some species transitions between stamens and

lodicules may occur. Lodicules usually present, the number some-

times not constant in different spikelets on the same plant, some-

times quite lacking. Ovary smooth, continued upwards into a

gradually tapering stiff style which is hollow around a central

strand of tissue, terminating in 3 short spreading stigmas, the

style at flowering usually a little longer than the palea (fig. 11C,

D). Fruit smooth, about as long as the lemma, bearing the per-

sistent stiff style at its apex (fig. 11E); seed more or less cylindric,

separable from the thin stiff pericarp (which is thicker towards

top of fruit); hilum linear, extending the whole length of the

seed; embryo small.

Field characters. Some of the Peninsula species flower con-

tinuously on leafy branches. In these cases the slender shape of

the spikelet, with its stiff style protruding from the tightly folded

lemma and palea, is distinctive. Of vegetative characters, the pre-

sence of fine appressed white (or colourless) hairs on the young

culms is universal, and only otherwise found in Gigantochloa apus

(not native and not commonly planted). Several species have

erect culms with slender drooping tips. S. gracile and S. grande

have culms not strong enough to support their own weight, and

if not supported by other plants curve even from the base and

may form a semicircle (this occurs also in Dendrocalamus pen-

dulus). The range of shape of blade and of auricles on culm-

sheaths shown by different species in this genus is considerable,

but the auricles are always bristly.

Limits of the genus. The above generic description would include

both Cephalostachyum and Teinostachyum of Gamble’s scheme.

Gamble originally placed these two genera in a different sub-tribe

from Schizostachyum, the only clear difference in the diagnoses of

the sub-tribes relating to the palea; in Dendrocalameae the palea

is said to be 2-keeled, in Melocanneae “none, or similar to the

flowering glume” (Gamble copied this from Bentham).

Vol. XVI. (1958).

In Gamble’s generic description of Cephalostachyum, the palea

is described as “2-keeled, keels close together’. But in the descrip-

tion of the species Schizostachyum tenue (belonging to Melocan-

neae) the palea is described in exactly the same words used for

Cephalostachyum, and I would agree with that description (fig.

11E). All one-flowered species of Schizostachyum examined by

me have a palea with two apical points close together. These

apices are usually elongated and 0-5—2-0 mm. long, and they are

more or less clearly produced downwards on the back of the palea

as two keels, with a more or less distinct groove between them.

I do not think it is possible to divide these species clearly into

those with 2-keeled and those: with un-keeled paleas, or to sepa-

rate them on these characters from species referred by Gamble to

Cephalostachyum. Some species of Cephalostachym (especially

the original C. capitatum and C. pallidum) are very distinct in

aspect when flowering, owing to the numerous very long-awned

bracts surrounding the dense groups of spikelets. But this is not

true of C. virgatum Kurz, and it is not clear to me why Gamble

did not refer it to Schizostachyum.

Teinostachyum, as originally defined by Munro, has several

perfect flowers in each spikelet (Bentham would only admit one

flower, and Gamble, copying Bentham’s scheme of generic sepa-

ration in his subtribal keys, had therefore a conflict between the

key on p. 77 and the diagnosis of Teinostachyum on p. 97). It

might seem reasonable to separate Teinostachyum from Schizos-

tachyum on the number of florets in a spikelet (one in Schizos-

tachyum, more than one in Teinostachyum). But I have found

that the spikelets of S. brachycladum (on specimens from several

parts of Malaya) have sometimes one flower, sometimes two;

and sometimes there is an empty lemma to represent the first

flower. Where there are two perfect flowers, the upper one has a

tightly convolute lemma and palea, with only slight keels near

the tip of the palea; in the lower flower the palea is not com-

pletely enfolded by the lemma, and has a distinct groove from tip

to base (fig. 11B). This is the condition of Teinostachyum also;

the rachilla lies in the groove of the palea of the lower flower (or

of all flowers below the last perfect one), but in the case of the

uppermost perfect flower there is no rachilla (only its thread-

like extension) and the palea is usually only slightly grooved (fig.

11A). Schizostachyum grande Ridl. has about four perfect flow-

ers, all except the apical one having fully grooved paleas, with

stout rachilla-internodes lying in the grooves.

Schizostachyum brachycladum is so closely related to S. zollin-

geri in every respect except the frequent presence of two flowers

Gardens Bulletin, S.

in spikelets of the former that it would be very unnatural to place

them in different genera. The sharp distinction between Teinos-

tachyum and Schizostachyum on number of florets therefore

breaks down, and I see no other possible separation of the com-

bined group of species of the two genera. The genus Dendrochloa

Parkinson has a spikelet-structure identical with that of S. grande

Ridl., and so it also should be included.

There is a group of species of Schizostachyum for which the

name Neohouzeaua has been proposed by Mile A. Camus. I

believe that this is a natural group of species, but the only cha-

racter which they all possess is a filament-tube instead of separate

filaments. They mostly lack lodicules, but so also does S. blumii

(type species of the genus) and some of them lack an extension

of the rachilla. I have argued elsewhere (Journ. Arn. Arb. 27:

341. 1946) that the presence of a filament-tube is not of basic

importance in the classification of bamboos, as it occurs on at

least three different evolutionary lines; and there is also evidence

that it may occur at another point in Schizostachyum (it is partial

in Dendrochloa distans Parkinson, which certainly belongs to the

same genus as S. grande Ridl.). I do not think that this character

alone is sufficient to warrant the separation of a genus; but if it

were possible, as it may be when we know them better, to sepa-

rate these species on a group of other characters, then generic dis-

tinction may be justified. The Malayan representative of this

group is S. jaculans.

DISTRIBUTION: as here construed in a broad sense, this genus has as

its centre of distribution the Burma-Siam-Indochina region. There are

certainly native species throughout Malaysia (including the Philippines

and New Guinea), and apparently a few in the islands of the Pacific.

There are a few species in southern India, Ceylon and Madagascar. In

Malaya there are several wild species, of which S. zollingeri is planted

as a village bamboo; S. brachycladum, probably not native, is also ex-

tensively planted. S. jaculans, used for making blow-pipes, has been

planted in most parts of the country and is probably wild in the north.

Uses. The thin-walled very straight culms of S. zollingeri are

the lightest bamboos of their size in Malaya, and so they are used

for making rafts. Probably S$. brachycladum is used similarly

where it occurs. These bamboos are also used as vessels in which

rice is cooked (lémang). Because they are easily split and then

flattened, they are used for making the walls (and floors ?) of

houses. But they split too readily to make useful strips for weav-

ing the side of large baskets; in the preparation of such strips the

splitting must be carefully controlled. I found that the culms of

S. grande, roughly split and twisted, were used for making the

rims of large baskets in the Cameron Highlands district. For the

side of the baskets thin strips of Gigantochloa were used, made

Vol. XVI. (1958).

by first vertical and then tangential cuts; tangential cutting of

Schizostachyum is impracticable. The siliceous surface of culms

of Schizostachyum is very hard, and probably bamboos of this

genus are best for making small implements in which sharp edges

are required. A probable Schizostachyum was reported by Rum-

phius as used in Amboina for making spears and also for flutes.

S. jaculans is used for making blow-pipes in Malaya. Fibres ob-

tained by macerating culms of S. brachycladum are used for weav-

ing in some parts of Indonesia, but I have seen no report of this

use in Malaya.

Key to Malayan species of Schizostachyum based on

spikelet-structure

Spikelets with one fertile floret (sometimes

2 in S. brachycladum)

Spikelets-groups 1-5 on the end of a

slender leafy branchlet; few spikelets

in each group a .. 1. S. gracile

Spikelet-groups more numerous, on a

straight rigid axis; many spikelets in

a group, sometimes on branched axes

Lemma 27 mm. long, its whole outer

surface hairy a .. .2, S. aciculare

Lemma not over 21 mm. long, its whole

outer surface not hairy

No lodicules; rachilla-extension us-

ually absent 26 .. 3. S. jaculans

Lodicules and rachilla-extension pre-

sent

Lemma 9-12 mm. long

Only one perfect floret; lemma

8-9 mm. long; rudimentary

floret very small .. 4. S. zollingeri

Sometimes 2 perfect florets;

lemma 12 mm. long; rudi-

mentary floret to 6 mm.

long Pa .. 5. 8. brachycladum

Lemma 15-21 mm. long

Stamens 4; lodicules 2; rachilla-

extension 13 mm.,_ rudi-

mentary floret 6 mm. long 6. S. insulare

Stamens usually 6; lodicules 3 or

more; rachilla-extension to

7 mm., rudimentary floret

under 0-5 mm. long

Spikelets with 3 or 4 fertile florets

Spikelets at several nodes on leafless ends

of branchlets ; :

Spikelets solitary, terminal on a leafy

branchlet (with one or more others

afterwards at its base ?)

Gardens Bulletin, S.

7. S. longispicu-

latum

8. S. grande

9. S. terminale

Key based on vegetative characters (excluding S. insulare)

Culm-sheath blades at 150 cm. above

ground stiffly erect, as wide as long or

nearly so

Culms to about 2 cm. diameter; leaves to

20 by 2 cm.

Culms much thicker; leaves much wider

Culm-sheath hairs dark brown; blades

much inflated; auricles to 7 mm. or

more high

Culm-sheath hairs red-brown; blades

not much inflated; auricles usually

not over 2‘5 mm. high

Culm-sheath blades much longer than wide,

erect or spreading

Culm-sheath blades erect, to 17 by 10

cm.; leaves commonly 50 by 7 cm.;

leaf-ligule to 8 mm. high

Culm-sheath blade deflexed or erect,

smaller; leaves smaller; leaf-ligule

much shorter

Culm-sheath blade about 8-10 times

as long as wide

Longest internode on a culm com-

monly 80 cm., sometimes much

more; bristles on auricles 10-12

mm. long ah

Longest internode on a culm shorter;

auricle-bristles usually shorter —

1. S. gracile

4. S. zollingeri

5. S. brachycladum

8. S. grande

3. S. jaculans

Vol. XVI. (1958).

Culm-sheath auricles 3-5 mm.

high, with free spreading ends

5 mm. long SS . 7. §. longispicu-

latum

Culm-sheath auricles not over 2

mm. high, ends not spreading

to form free lobes

Top of culm-sheath truncate .. 2. S. aciculare

Top of culm-sheath triangular 9. S. terminale

Culm-sheath blade about twice as long

as wide .. iz tid ai keris: OPEC

Culm-sheaths not known a “hES.6.-$> insulare

1. S. gracile (Munro) Holttum in Kew Bull. no. 2, 1956: 206.

Basonym: Melocanna gracilis Munro in Trans. Linn. Soc. 26:

133. 1868.

Synonym: S. tenue Gamble in Ann. R. Bot. Gard. Calc. 7: 114,

pl. 100. 1896. Ridley, Flora 5: 268.

Culms 3-4 m. long, 1:5—2:0 cm. diameter, not stiffly erect.

Culm-sheaths when young bearing very fine appressed light brown

hairs; blades of middle culm-sheaths erect, stiff, commonly 3—4-5

cm. long and 1-5—3 cm. wide, apex stiffly acuminate, upper sur-

face pale-hairy near base; blades of upper sheaths elongate and

slightly reflexed; auricles about 10 mm. in lateral extent with free

outer ends projecting beyond the width of the sheath, 2-3 mm.

high, free edge throughout bearing close curved almost smooth

bristles (often somewhat purplish) to 10 mm. long; ligule under

0-5 mm. high including a fringe of fine hairs. Leaf-blades com-

monly 10—20 cm. long, 1—2 cm. wide (on sucker-shoots to 3 cm.),

surfaces smooth and glabrous; auricles small, sometimes with a

few slender bristles. Inflorescence of 1-6 groups of spikelets at

the distal nodes of a very slender more or less fiexuous leafy

branchlet, few spikelets in each group. Pseudo-spikelets 12-14

mm. long at flowering; lemma 8—9 mm. long, glabrous, with stiff

tip; palea 10-12 mm. long, very shortly 2-pointed with a narrow

groove almost to the base; lodicules 3, fringed; ripe fruit about

as long as palea, narrowed at the top and bearing the stiff style,

total length 20 mm. (fig. 11E); seed 65 mm. long, embryo 1

mm.; rachilla-extension 6-7 mm. long, rudimentary floret hardly

0-5 mm. long.

DISTRIBUTION: southern half of Malaya, on the edge of forest, often

on river-banks in Johore (also in Malacca, Negri Sembilan and

Pahang).

RECORDED MALAY NAMES: Buloh Rappen, Buloh Akar.

Gardens Bulletin, S.

The above vegetative description is based on plants (introduced

from the wild) cultivated at the Federal Experiment Station of the

Department of Agriculture, Serdang, Selangor. These plants are in

separate clumps in the open; the old culms are not strong enough

to stand upright with the weight of their branches.

The type specimen of Melocanna gracilis Munro is Wallich no.

5032 collected at Singapore. I have seen three sheets of this col-

lection at Kew, and have no doubt that they represent the same

species as §. tenue Gamble, as typified by Ridley’s no. 5596 (K.

Berar, Pahang). Gamble confused the three closely allied species

S. gracile, S. zollingeri and S. brachycladum, and in fact it is not

always easy to separate herbarium specimens, though the living

plants are distinct. Gamble further confused the situation by using

the name chilianthum, which properly belongs to a species of an-

other genus (not in Malaya). He put the type of S. gracile, and

some specimens I would call S. zollingeri, into S. chilianthum.

‘There are some intermediate specimens, which seem to me to

match a living plant I saw near Tampin; these are tentatively re-

ferred to a variety erectum, described below. Gamble thought

that his S. tenue might be equivalent to Bambusa elegantissima

Hassk., but the latter has now been referred to the genus Chloo-

thamnus by Henrard.

Specimens: SINGAPORE, Wallich 5032 (K, type). Jonore, S. Tebrau,

Ridley 13211, 11496 (S.K). Kota Tinggi, Ridley 15413 (S,K); s.n. Dec.

1892 (S.). Segamat, Holttum S.F.N. 38302 (S). Danau, S. Sedili,

Corner s.n. 28.3.1932 (S). Maracca, Bukit Toongul, Ridley 5601

(S,K). Ayer Panas, Ridley s.n. 1891 (S.). SELANGoR, Labu River, Rid-

ley 7789 (S.) Federal Expt. Station, cult., Holttum S.F.N. 37791 (S).

PAHANG, K. Berar, Ridley 5596 (S,K). Pekan, Burkill & Haniff S.F.N.

17120 (S). TRENGGANU, Kemaman, Vaughan Stevens s.n. 1891 (S);

same origin, cult. Singapore, Ridley 6900 (S,K).

Var. erectum Holttum, var. nov.

Culmi rigidiores, erecti, apice solum nutantes; vaginae culmo-

rum dorso pilis paucis pallidis vestitae; laminae vaginarum cul-

morum longiores, auriculae parvae, haud elevatae, latitudine c. 5

mm.; spiculae plures.

TYPE: Negri Sembilan, near Tampin, Holttum S.F.N. 38408 (S.).

The type of this variety had erect slender culms, drooping at the

tip only, like a. small condition of S. zollingeri; but the blades

and auricles of the culm-sheaths were very different from S. zollin-

geri, the blades being longer and the auricles small and low. The

spikelets were more numerous in each group than is normal in

S. gracile; this is also a resemblance to S. zollingeri, but the axis

bearing the spikelets is slender and flexuous, as in S. gracile, not

rigidly straight. Among herbarium specimens of the Malacca re-

gion I have found some others which have the same kind of

Vol. XVI. (1958).

flowering branches, and place them also tentatively here. In the

Waterfall Garden, Penang, is a plant of similar habit, introduced

from an unrecorded locality in Malaya. It differs from the above

description in having culm-sheaths bearing dark hairs, with fairly

large auricles; flowers have not been seen.

Several specimens of S. gracile and its variety have ripe fruits; I

have not seen any such on S. zollingeri.

Specimens (of var. erectum): Matracca, Derry 192 (S). Vaughan

Stevens 3947 (S,K). NEGRI SEMBILAN, Pantai, Ridley 10069 (S.K).

SELANGOR, near Sepang, Ridley 7780 (S,K). cult. SInGAPorRE, Ridley

6892 (S,K), 11349 (S,K), s.n. 1903 (S).

2. S. aciculare Gamble in Ann. R. Bot. Gard. Calc. 7: 117, pl.

104. 1896. Ridley, Flora 5: 270.

A small bamboo. Culm-sheaths 12 cm. long, the top truncate,

hairs few, appressed, pale: blade reflexed, 8 cm. long, 7 mm. wide;

auricles 4—5 mm. in lateral extent, 1-5 mm. high, with many close

curved bristles to 3 mm. long; Jigule 1-5 mm. high, including a

fringe of short hairs. Leaf-blades to about 22 by 5 cm. (some-

times 22 by 3-5 cm.), glabrous or the lower surface a little hairy

near the base, petiole 5-8 mm. long; auricles shortly spreading,

bearing several bristles; ligule short, with some short slender mar-

ginal bristles when young. Inflorescences to about 30 cm. long,

at the ends of leafy branches, usually unbranched, the spikelet-

groups 2-5-5 cm. apart. Pseudo-spikelets to 5 cm. long; basal

bud-bearing bracts several, rather small, with one longer (2 cm.)

sheathing the base of the lemma, its apex not spinous; internode

between uppermost bract and lemma to 1 cm. long, its apex with

a fringe of short hairs, articulated at the base of the lemma; per-

fect flower one, with a very slender rachilla-extension nearly as

long as the lemma; Jemma c. 27 mm. long, very shortly pointed,

tightly rolled (diameter of rolled lemma 1-5 mm.), outer surface

throughout bearing short obliquely spreading hairs; palea about 2

mm. longer than lemma and almost entirely enclosed by lemma,

apical points short; lodicule 1 (seen by Gamble) or absent, nar-

row, glabrous; anthers 14 mm. long with short hairy tips.

DISTRIBUTION: known from four collections from the lowlands of

Malacca, Negri Sembilan and Selangor; also from Siam (Katok, Korat,

Kerr 8178).

RECORDED MALAY NAMES: Buloh Padi, Buloh Akar.

The long hairy lemma and the long internode below it are dis-

tinctive. The culm-sheath is described from the Kew specimen of

S.F.N. 28422. I could not find any lodicule in two spikelets dis-

sected in Singapore.

Specimens: Matacca, Sungei Hudang, Ridley 5600 (S,K). Necrr

SEMBILAN, Kuala Pedas, Ridley 10065 (S.K). Bukit Kepayang, Alvins

2167 (S, type). SELANGOR, Bukit Cheraka Forest Reserve, Klang,

Pestana S.F.N. 28422 (S,K).

Gardens Bulletin, S.

3. S. jaculans Holtt. in Kew Bull. 1953, no. 4: 494.

Synonym: S. blumii quoad Gamble (p.p.) and Ridley, not of

Nees.

Culms slender, 6—7 m. long, to 3-5 cm. diameter at the base;

longest internode commonly 80 cm., sometimes to 125 cm. long,

top of young internode covered with a little waxy powder. Culm-

sheaths to 30 cm. long, the back bearing copious medium-brown

hairs which are not closely appressed (thus easily detached), the

top truncate; blade green, at first erect, soon reflexed, attached to

a slight depression in the top of the sheath, 10-25 cm. long, 7-18

mm. wide; auricles hardly 1 mm. high, to about 17 mm. in lateral

extent, bearing slender pale bristles 8-12 mm. long; ligule hardly

2 mm. high, fringed with fine hairs to 3 mm. long (fig. 12). Leaf-

blades 12-30 cm. long, 1-2-5 cm. wide, lower surface shortly

soft-hairy; petioles of lowest leaves 3-4 mm. long, of upper leaves

5—7 mm.; auricles usually bearing a few slender bristles 4-10 mm.

long; ligule short. Flowering branches usually leafless, slender,

glabrous, bearing spikelets in dense tufts at the nodes; lower inter-

nodes 4 cm. long, upper ones 1 cm. long. Spikelets 1-flowered,

slender, glabrous, usually without any extension of the rachilla

beyond the flower (a very slender extension 5 mm. long once

seen); bracts below flower to 6 mm. long, mucronate, glabrous;

lemma c. 10 mm. long, glabrous, shortly pointed; palea c. 16 mm.

long, the two tips 0-5 mm. long, the back hardly grooved; lodi-

cules absent; anthers 6-6-5 mm. long, filaments when young

joined to form a tube 0-5—1-5 mm. long; ovary with the stiff style

at flowering 18 mm. long; stigmas 3, purple, 1-5 mm. long; fruit

unknown.

DISTRIBUTION: probably wild in the north of Malaya and in the

region further north, planted at many places throughout Malaya for

use in making blow-pipes; flowering rarely in Malaya.

RECORDED MALAY NAMES: Buloh Sumpitan, Buloh Témiang, Buloh

Kasap, Buloh Tikus.

This appears to be the species commonly used throughout the

lowlands of Malaya for making blow-pipes (see Holttum, l.c., and

remarks under Bambusa wrayi), and is doubtless the bamboo re-

ferred to by Wray in the footnote in Skeat & Blagden’s Pagan

Races of the Malay Peninsula, Vol. 1, p. 255. Two internodes are

necessary for making the inner tube of a blow-pipe; these are very

carefully joined, and then they are put into a stronger tube made

from a larger part of the same kind of bamboo. The method is

described by Skeat & Blagden, p. 281.

This species would be referred to Neohouzeaua by those who

maintain that genus (see introductory discussion on Schizosta-

chyum above). When publishing the description of the species in

Vol. XVI. (1958).

1953, I overlooked the fact that closely smmlar bamboos occur m

Java, Borneo and the Philippimes, and perhaps elsewhere m

Malaysia; I also overlooked that fact that flowermg specimens had

previously been found m Malaya, and had been identified (by

Gamble and Ridley) as S. blumii Nees. Backer’s description of the

latter species (Handb. Fl. Jav. 2: 285) differs considerably from

S. jaculans im floral characters, but agrees m shape of culm-

sheaths (though Backer says the blades are erect) and m length of

internodes. Backer gives as a synonym Bambusa longinodis

Britisk Crown Copyright. Permissien of the Centreiler, H.M.S.O.. cbtuined..

Fig. 12. Culm-sheath of Schizostachyum jaculams. A, outer surface of a

sheath, showing auricle bearing bristles with the ligule behind it.

and part of the base of the blade, x 4/3. D, immer surface of

parts shown in C, showing the short fringed ligule with the long

bristles of the auricles behind it.

POR Ped TO Pr

Gardens Bulletin, S.

Miquel, which appears to be based entirely on Arundarbor spicu-

Jorum Rumph. (Dr. van Steenis and Dr. Lanjouw kindly inform

me that no specimens named B. longinodis by Miquel can be

found in the herbaria at Leiden and Utrecht).

More recently, McClure has examined and described the type

specimen of S. blumii (Blumea 2: 86-94. 1936), and there is no

doubt that it is quite different from the species described by

Backer. It might seem reasonable therefore to use Miquel’s name,

transferred to Schizostachyum, for Backer’s species; but I think it

would be better to regard Bambusa longinodis as a nomen dubium,

because there appears to be a group of species of this alliance.

S. lima (Blanco) Merrill, in the Philippines, is one of these, and I

have seen specimens of another from North Borneo which appears

to be unnamed.

Though there is in Borneo a slender Schizostachyum with flow-

ers rather like those of S. jaculans (filament-tube present, and no

lodicules) there is no indication that this bamboo is used for

making blow-pipes. Blow-pipes are used, but they are made by

boring pieces of wood. It may be that the Bornean species of

Schizostachyum has shorter internodes than S. jaculans and is

therefore unsuitable. A specimen at Kew has a note that internodes

are 18-30 inches long (45-75 cm.); a note on another specimen

‘states that the culms are used for making spears.

Specimens: SINGAPORE, Chua Chu Kang, Ridley 6114 (S,K). Bedok,

Ridley 13323 (S,K). Seletar, Ridley 445 (S). Chan Chu Kang, Ridley

73 (S). JoHore, Buloh Kasap, Holttum S.F.N. 38307 (S). Kota Tinggi

le Doux s.n. 10.2.1948 (S). Maracca, Ayer Kroh, Ridley 10769 (K).

Vaughan Stevens 3936 (S; sterile, doubtful). SELANGOR, Kepong,

Wyatt-Smith F.D. 71511 (K,Kep., type). Bukit Lagong F.R., Kuala

Lumpur, Symington F.D. 56740 (Kep.). PAHANG, Near Bentong, A.

Rahman F.D. 4061 (Kep.). Pulau Tioman, Henderson S.F.N. 18486

(S); Burkill s.n. 1915 (S). PERAK, Road to Jor, Ridely 13852 (S,

doubtful). Cameron Highlands Road, 8th mile, Holttum S.F.N. 38419

(S); 64 mile, Le Mare s.n. 23.1.1947 (S). Changkat Jong, Holttum

S.F.N. 38433 (S).

4. S. zollingeri Steud., Syn. Pl. Gram. 332. 1854. Kurz in Journ.

As. Soc. Beng. 39 pt. 2: 88, pl. 12. 1870. Backer, Handb. FI.

Jav. 2: 283. Ridley, Flora 5: 269.

Synonym: S. chilianthum quoad Gamble in Ann. R. Bot. Gard.

‘Calc. 7: 115 (in part; not Chloothamnus chilianthus Buse; see

Henrard in Blumea 2: 60. 1936).

Culms close, stiffly erect with slender tips only drooping, to c. 15

m. tall; internodes commonly to 40 cm. long and 2-10 cm. dia-

meter. Culm-sheaths long persistent, to c. 15 cm. long, very stiff,

light brown or flushed with purple near top when young, top

rounded (not truncate), back more or less covered with appressed

‘Shining dark brown hairs; blade rigid, erect, strongly convex,

Vol. XVI. (1958).

stiffly pointed, flushed with purple when young, broadly trianglar,

commonly 7—9 cm. long and wide (lower ones wider than long),

slightly narrowed at the base, upper surface pale-hairy; auricles to

7 mm. or more high, spreading laterally beyond the width of the

top of the sheath (smaller on some plants), their edges bearing

very close curved hairy bristles; ligule to 4 mm. high, edge smooth

or short-hairy (fig. 13). Leaf-blades commonly to 30 cm. long

and 4 cm. wide, sometimes to 40 by 6:5 cm., or on small shoots

only 12 by 1 cm., base cuneate, lower surface glabrous or slightly

hairy, more or less rough to the touch, upper surface smooth,

petiole broad and c. 5 mm. long; sheath hairy when young; auri-

cles usually well developed and sometimes laterally spreading,

with many slender bristles; ligule short. Spikelets usually borne in

dense tufts at the nodes on the rigid distal part of a leafy branch-

let, flowering nodes 1—3 cm. apart. Pseudo-spikelets 15-20 mm.

long; basal bracts with stiff awn-like points 1-5 mm. long, edges

more or less fringed; perfect flower 1, with slender rachilla-exten-

sion 5—9 mm. long bearing a very small rudimentary floret; lemma

8-9 mm. long, edges slightly hairy near tip; palea about 2 mm.

longer than lemma, grooved on the back near apex only, apex 2-

pointed, edges near apex short-hairy; lodicules 3, 2-5—4 mm. long,

apex rather broadly triangular and shortly fringed; anthers 3-5-5

mm. long with short blunt tips.

DISTRIBUTION: Java, Sumatra (?); probably throughout Malaya, om

edges of forest and in clearings, from Ulu Langat in Selangor north-

wards, also often planted in villages; does not flower continuously.

RECORDED MALAY NAMES: Buloh Tulo (= telor?), Buloh Telor,

Buloh Pélang, Buloh Nipis, Buloh Dinding, Buloh Pauh (= Pa-aur?),

Buloh Kasap, Buloh Lémang, Buloh Aur.

Used for making floors and walls of houses, for rafts, and for

cooking rice; probably also for making small implements where a

sharp edge is necessary.

The name S. zollingeri was given by Steudel to Zollinger’s no.

3529 from Java, which I have not seen. I identify Peninsula plants

(from which the above description is made) with Steudel’s species:

on the evidence of Kurz and Backer. The above description agrees

with Backer’s description of Java plants, except that he gives dia-

meter of culm as 1-5-5 cm., lemma 7—8 mm. long and lodicules

1-5-2 mm. long. It is possible that the Java plants represent a

different species, in which case the Peninsula ones need a new

name. There is variation among Peninsula plants referred here by

me in (a) extent of hairiness of young culms, which are in some

cases very conspicuously glaucous when seen from a distance, in

others less so with sparse pale hairs, and (b) size of the culm-

sheath auricles, which are in some cases rather small (notably in:

Gardens Bulletin, S.

Fig. 13. Culm-sheath of Schizostachyum zollingeri. A, outer surface of

complete flattened sheath. B, inner surface of blade and top of

sheath, to same scale as A. C, inner view of part of top of

sheath, showing ligule, auricle and part of blade, « 2. D, outer

view of top of sheath and an auricle.

Vol. XVI. (1958).

plants seen in forest clearings at Ulu Langat). The peculiar shape

of the blade of the culm-sheath, the rounded top of the sheath,

and the very dark colour of the hairs on the sheath, are very dis-

tinctive, and are well described by both Kurz and Backer.

Specimens: SINGAPORE, Bukit Timah Road, Ridley 6111 (K), 6116

(S,K), 10939 (S,K). Yio Chu Kang, Ridley 11293 (K), 11294 (S,K).

Ulu Pandan, Holttum s.n. 10.10.1948 (S). Bukit Panjang, Ridley

11850 (S,K). JoHorE, Bukit Muar, Fielding 4420 (S,K). Serom, Rid-

ley 10990 (S,K). Road to Castlewood, Ridley s.n. 1904 (S). Ulu

Segaun, G. Panti, Corner S.F.N. 30679 (S,K, doubtful). NeGri S—Em-

BILAN, Jelebu, Md. Tahir F.D. 9594 (S,K). Pertang, Hussein F.D.

9599 (S,K). 151 mile Ampangan Road, Buyong F.D. 11004 (K).

Tampin-Seremban Road, Holttum S.F.N. 38409 (S). Pertang, Pantai

and Ulu Klawang, per A. Arber (K). G. Tampin Reserve, Kinsey s.n.

(K). No locality, Alvins 1089 (S). SELANGOR, Langat Valley, Holttum

S.F.N. 38411 (S). Ginting Simpah Road, Holttum S.F.N. 38414 (S).

PAHANG, near Pekan, Burkill & Haniff S.F.N. 17253 (S). Perak, Grik,

Hamid F.D. 6422 (S,K), 8251 (S,K). Semat, Jenal F.D. 8323 (Kep.)

S. Kulim, Burkill & Haniff S.F.N. 13811 (S,K). Telok Anson, Pestana

S.F.N. 28431 (S).

5. S. brachycladum Kurz in Journ. As. Soc. Beng. 39 pt. ii: 89,

pl. VI fig. 2. 1870. Backer, Handb. Fl. Jav. 2: 282.

Habit of S. zollingeri, but with less drooping culm-tips; culms to

8 cm. diameter, green, or yellow with a few narrow green longi-

tudinal streaks. Culm-sheaths long persistent, to c. 18 cm. long,

rigid, densely covered with appressed fine red-brown hairs, junc-

tion of top of sheath with blade horizontal; blade erect, slightly

convex, triangular with stiffly acuminate apex, on middle nodes to

c. 8 cm. long and 10 cm. wide (base extending to the full width

of the top of the sheath); auricles rather small, usually 2-5 mm.

high (but see var. auriculatum below), crisped, bearing crisped

bristles 4-5 mm. long; ligule to 2 mm. high, edge smooth (fig. 14).

Leaf-blades to about 40 by 7 cm., lower surface softly hairy (hairs

c. 1 mm. long); auricles usually with several bristles (lacking on

upper leaves); ligule low, entire. Flowering branches as in S.

zollingeri. Spikelets 1- or 2-flowered, with slender rachilla-exten-

sion bearing rudimentary floret to 6 mm. long (shorter if 2 perfect

florets present); rachilla-internode between the 2 perfect florets 4

mm. long; lemma 10-12 mm. long including a stiff point 1 mm.

long, edges fringed near apex; palea 11-14 mm. long, tip very

shortly 2-pointed; lodicules 3, translucent or purplish, nearly

equal, to 5 mm. long, apical halves shortly hairy on and near

edges; anthers purple, 6 mm. long; ovary and style at flowering c.

15 mm. long (fig. 11, A-C).

DISTRIBUTION: Java and Eastern Malaysia; in Malaya often planted

as a village bamboo (green culms) or as an ornamental (yellow

culms), some plants at least flowering continuously. A specimen of

Ridley’s, dated 1896, of the yellow variety, has a note “cult. Singapore,

said to have come from Borneo”.

Gardens Bulletin, S.

‘g[olIne Jo pue ‘apeyq jo sdejins JduUT ‘oyNSI] SUIMOYS MIA

iouut ‘q <. xX ‘apriq pue yiesys jo syed yuooulpe YM s]oLINe oO JO MOIA IBINO ‘5D “YJeOUS BH iat pue bead jo

MOIA JouUl ‘g ‘¢/[ X ‘GyeoYs pous}jey sjo[dWIOD Jo sdNjIns JojnO “Y ‘wWinpH]IXYIDIG WinKkYIDISOZ1YIF JO YYeoys-WIND “pl “sIy

q\virtl hwy

Vol. XVI. (1958).

RECORDED MALAY NAMES: Buloh Nipis, Buloh Lémang, Buloh Padi,

Buloh Urat Rusa, Buloh Pélang.

The specimens in Gamble’s herbarium which I would place

under this species were mostly referred by him to S. zollingert;

but it is clear from his notes that he was not satisfied as to the dis-

tinctions between specimens labelled by him S. zollingeri and S.

chilianthum. It is in fact difficult to name flowering material of

these plants, though vegetatively they are very distinct. Gamble

did not mention having seen spikelets with two perfect flowers,

though such occur on some specimens from his herbarium. Backer

also has no reference to two-flowered spikelets. I have found two

flowers in specimens from Perak and Kedah, as well as in yellow-

culm plants cultivated in Singapore. Of other differences from S.

zollingeri, 1 believe that the longer lemma of S. brachycladum is

generally a good distinction, and the softly hairy leaves. Leaves of

some specimens of S. zollingeri are however slightly hairy.

Specimens: SINGAPORE, Botanic Gardens, Ridley 8086 (S); Burkill

sm. 1.5.1916 (S,K); Md. Nur s.n. 17.8.1918, 27.3.1922 (S); Pestana

s.n. 2.6.1933, 28.6.1933, 28.8.1933 (S). Pasir Panjang, Ridley 10381

(S,K). St. John’s Island, Ridley 11360 (S,K). Nassim Road, Ridley

12560 (S,K). Blakang Mati, Ridley 12561 (S,K). Mt Echo, cult.,

Burkill s.n. 24.7.1915 (S,K). JoHorE, Road to Reservoir, Ridley 9177

(S,K). PERAK, Krian, cult., Ridley 9385 (S,K). Ipoh, Ridley 11869

(S,K). Semat, Jenal F.D. 8322 (Kep., K). Taiping, Dolman F.D.

10238 (K,Kep.). Near Taiping, Holttum s.n. 28.10.1946 (S). PAHANG,

Kuala Lipis, Marshall F.D. 21377, 21378, 21379, 21382 (Kep., sterile).

PENANG, Waterfall Gardens, Md. Nur s.n. 26.9.1918 (S); Pestana

s.n. (S). Government Hill, 1,000 ft., Curtis 3792 (S).

Var. auriculatum Holttum, var. nov.

Auriculae vaginarum culmorum majores, eis S. zollingeri simi-

les. Singapore, Pasir Panjang, 63 mile, Kiah, s.n., 11.12.1934

(K, Kep. S).

The type of this variety appears to agree in all characters with

normal S. brachycladum except for the auricles of the culm-

sheaths, which are as big as those of S. zollingeri. I have not seen

another plant with this character. The type plant is certainly not

native in Singapore, and was probably brought from somewhere

in Indonesia. Ridley apparently collected specimens from the same

plant in 1899 (no. 10381).

6. S. insulare Rid]. in Journ. Str. Br. R. As. Soc. 61: 64. 1912.

Flora 5: 270.

Vegetatively indistinguishable from S. longispiculatum; rachilla-

extension 13 mm. long bearing a rudimentary floret 6 mm. long;

stamens usually 4, with 2 lodicules.

DISTRIBUTION: only known from the type specimen, from Pulau

Rawei in the Butang group, north of Langkawi (Ridley no. 15931; S,

K.) Ridley stated that the filaments of the stamens were united to

Gardens Bulletin, S.

form a tube, but in a spikelet examined by me the filaments of fully

expanded stamens appeared to be free. The long rachilla-extension

and large rudimentary floret are unlike anything in the various speci-

mens of S. longispiculatum examined by me, so that I think S. insu-

lare may reasonably rank as a distinct species. It will be interesting to

see whether specimens from the mainland north of Malaya are simi-

lar, or whether S. insulare is confined to the islands.

7. S. longispiculatum Kurz in Journ. As. Soc. Beng. 39 pt. ii: 89,

pl. VI fig. 1. 1870. Ridley, Flora 5: 270.

Synonyms: S. latifolium Gamble in Ann. R. Bot. Gard. Calc. 7:

117. 1896. Ridl. Flora 5: 270.

Ochlandra ridleyi Gamble l.c. 127. 1896. Ridley,

Flora 5: 272:

Schizostachyum ridleyi Holtt. in Gard. Bull. Singap.

11: 296. 1947.

Culms erect with slender drooping tops, commonly 1:5—2:5 cm.

diameter, 3—5 m. tall, longest internodes 35-55 cm. long, walls

3-4 mm. thick; surface covered with white hairs when young, with

a little waxy powder just below the nodes. Culm-sheaths 12-14

cm. long, thin, green turning to papery light brown when old,

hairs pale, sparse and appressed; blade green or pinkish when

young, spreading or reflexed, on middle sheaths 6—9 cm. long and

12-17 mm. wide, abruptly narrowed at base to 10-12 mm.; auri-

cles on a base c. 10 mm. long, 3—5 mm. high, extending laterally

as narrow free lobes 5 mm. beyond the base of attachment, edge

bearing bristles S—10 mm. long; ligule 1-1-5 mm. high, very shortly

fringed (fig. 3). Leaf-blades commonly about 25 by 3 cm., largest

seen 40 by 7 cm., lower surface short-hairy when young, petiole

4-8 mm. long; projecting auricles often present like those on

culm-sheaths but smaller, bearing bristles 4-7 mm. long; ligule

short. Spikelets borne in groups at the distal nodes of a branch

which is leafy at the base; flowering part of the branch 10—25 cm.

long, sometimes with short lateral branches near the base. Pseudo-

spikelets 22-30 mm. long, slender and terete; basal bracts with

spine-like tips, edges slightly fringed; lemma 18-21 mm. long in-

cluding spine-like apex of 25-3 mm., surface near apex hairy

(hairs spreading); palea about 2 mm. longer than lemma, apex

with 2 close slender points 1-5-2 mm. long, back hardly grooved;

rachilla-extension very slender, usually 4-7 mm. long, bearing a

rudimentary floret less than 0-5 mm. long; anthers 9-10 mm.

long, apex very shortly apiculate, the tip sometimes with a few very

short hairs, or in some cases the apiculus so reduced that it is

overtopped by the tips of the pollen-sacs, the apex thus 2-lobed;

one or more stamens sometimes more or less intermediate in

Vol. XVI. (1958).

structure between a stamen and a lodicule; lodicules very vari-

able in spikelets on the same branch, 3-10 in number, unequal,

commonly 5—7 mm. long, in two observed cases a single lodicule

18 mm. long.

DISTRIBUTION: Malaya, Borneo, Western Java (?); in Malaya not

uncommon, especially on the edges of forest by rivers, found in all

parts of the country.

Gamble noted that his S. latifolium differed from S. longispicu-

latum in the presence of lodicules (said to be lacking in S. longis-

piculatum) and in the apiculate anthers. I have found great

variation in the lodicules (as noted in the description) and the

occasional complete absence of lodicules would not be surprising.

As regards the apices of the anthers, I find much variation also in

the development of the apiculus (a prolongation of the connec-

tive); when it is small, the tips of the pollen-sacs overtop it, and

the anther could then be called bilobed. In his original description

of Ochlandra ridleyi, Gamble noted that three stamens shorter

than the rest had a slender hairy apiculus, the others a short blunt

one. The specimen which Gamble examined had an unusually

large number of lodicules, which is the only reason for regarding

it as Ochlandra. In every other respect it is indistinguishable from

the type of his Schizostachyum latifolium.

Specimens: SINGAPORE, Bukit Mandai, Ridley 4620 (S,K, type of

O. ridleyi). Chua Chu Kang Road, Holttum S.F.N. 38406 (S).

JoHORE, Bukit Muar, Fielding s.n. 1892 (S). Mawai Road, Kota

Tinggi, Holttum S.F.N. 38404 (S). Maracca, Tebong, Ridley s.n. Jan.

1917 (K). NeGRI SEMBILAN, Johol Woods, Ridley s.n. 17.1.1917 (K)-.

PAHANG, Tanjong Antan, Pahang River, Ridley 5598 (S,K). Kota

Glanggi, Ridley 5602 (S,K). Kuala Lipis, Machado 11592, 11593

(S,K). Kuala Lipis, Burkill & Haniff S.F.N. 15656 (S). Bentong,

river bank, Burkill & Haniff S.F.N. 16423 (S,Kep.). Sungei Cheka,

Holttum S.F.N. 24784 (S,Kep.). Kuala Tahan, Holttum s.n. 27.9.1954

(S,K). SELANGOR, Ginting Bidai, Ridley 7788 (S,K). Batu Tiga, Rid-

ley 11936 (S). Perak, Lumut, Ridley 8389 (S,K), 10322 (S,K).

Tanjong Malim, Ridley 11863 (S,K). Tapah, Ridley 14035 (S,K).

Cameron Highlands Road, 700 ft., Holttum S.F.N. 38427 (S).

KELANTAN, Pehi River, Ridley s.n. 3.2.1917 (K). Kuala Krai, Haniff

& Nur S.F.N. 10095 (S,K,Kep.).

8. S. grande Rid]. in Journ. Str. Br. R. As. Soc. 82: 204. 1920.

Flora 5: 271.

Culms to at least 20 m. long, to about 12 cm. diameter, erect

only when young, usually curved in almost a semicircle and often

drooping to the ground when not supported by trees; internodes

to 90 cm. long, covered when young with copious white appressed

hairs. Culm-sheaths to 35 cm. long, when young pale yellow grad-

ing to pale green, covered on the back with appressed almost

white hairs, line of junction with blade slightly arched, auricles

very small with rather short slender bristles (5 mm. long); blade

Gardens Bulletin, S.

early deciduous, narrowly triangular (broadly triangular on lower

sheaths), erect, dark brown when young, to at least 27 cm. long

and 10 cm. wide, stiff, edges inrolled towards apex, base very

slightly narrowed, outer surface smooth, upper surface rather

densely covered with appressed pale hairs; ligule 4-8 mm. tall,

irregularly toothed. Leaf-blades to 60 cm. long and 10 cm. wide,

commonly 50 by 7 cm. (smaller leaves 30 by 4-5 cm.), apex

acuminate, base somewhat asymmetric, surfaces glabrous, petiole

very stout, 10 mm. long; sheaths often appressed-hairy when

young, the top ascending above the attachment of the petiole on

either side and joined across by the ligule, auricles small, no bris-

tles seen; ligule to 8 mm. tall above base of leaf-stalk, edge some-

times bearing slender bristles near each end. Spikelets in dense

groups at the nodes of the distal part of a leafy branchlet, or on

leafless branched branches direct from the main culm; mature

spikelets c. 3 cm. long, consisting of 3 (or 4?) florets and a rudi-

mentary terminal one, the uppermost perfect floret with tightly

rolled lemma and palea, lower florets with loose lemma and palea;

rachilla-internodes 7 mm. long, slender, widened to the apex,

glabrous; lemma of lower florets 10-15 mm. long with a short

stiff tip, glabrous; palea of lower florets 12-17 mm. long, glabrous,

very shortly 2-pointed, with a narrow groove down the back;

anthers 8 mm. long, not apiculate; lodicules 3, somewhat variable,

3-5-5 mm. long, narrowly elliptic, short-pointed, shortly fringed;

fruit 10-12 mm. long, cylindric, abruptly narrowed into a stiff

curved beak 15 mm. long.

DISTRIBUTION: one of the commonest gregarious bamboos of open

places in the foot-hills of the Main Range, very abundant at Ginting

Simpah (2,000 ft.), at the Gap (3,000 ft.) and near the road to

Cameron Highlands; found also on Gunong Raya in Langkawi islands

at 2,600 ft. (on granite), and in the lowlands of Pahang (near Ben-

tong and Kuala Lipis), Upper Perak (Grik) and Kelantan.

RECORDED MALAY NAMES: Buloh Séméliang (from several localities),

Buloh Séminyeh.

The spikelets of S. grande have the same structure as those of

Dendrochloa distans Parkinson from Burma (Indian Forester 59:

707. 1933), and I would include the latter species in Schizosta-

chyum. Parkinson’s species has much larger spikelets than those

of S. grande (5—7 much larger florets, and rachilla-internodes 2

cm. long). Parkinson stated that the filaments are partly united

(3 together, 2 together and 1 free), but I do not regard this as a

difference of generic rank. He also stated that the pericarp is

adherent to the seed, and therefore thought his genus should be

near Bambusa. I have examined dried fruits of the type specimen,

Vol. XVI. (1958).

and find that when soaked the thin pericarp can be detached from

the seed, as in Schizostachyum; the structure of the fruit is exactly

like that of Schizostachyum gracile.

Specimens: SELANGOR, Ginting Bidai, Ridley 7787 (K,S). Pahang

Track, 15th mile, Ridley 8457 (S,K); Machado 11591 (S,K). Semang-

kok Pass, Ridley 12043 (Type, S,K); Ridley s.n. 21.1.1921 (K). Near

Gap, Burkill F.D. 7774 (S); Strouts F.D. 10458 (Kep.); Symington

F.D. 20162 (Kep.,S); Curtis 3745 (S,K). Ginting Simpah, Holttum

S.F.N. 38412 (S). PAHANG, Kuala Lipis, Somerville F.D. 10489 (K);

Marshall F.D. 21380, 21384 (Kep.). Raub Road, Best S.F.N. 14120

(S). Bentong, Burkill & Haniff S.F.N. 16490, 16598 (S). Chigar Perah,

Henderson S.F.N. 22585 (S). Gunong Tahan, Wray & Robinson

5550 (S). PERAK, Grik, Hamid F.D. 6423 (Kep.,S), F.D. 8257 (S.K),

F.D. 8896 (Kep.). KELANTAN, Ulu Nenggiri, Symington F.D. 51656

(Kep.). Kelumpur, Haniff & Nur S.F.N. 10389 (S,K,Kep.). KEpDanH,

Gunong Raya, Langkawi, Corner & Nauen S.F.N. 37993 (S,Kep.).

9. S. terminale Holtt. in Gard. Bull. Singap. 15: 274. 1956.

Culm in type specimen 1 cm. diameter. Culm-sheaths 13 cm.

long, thin, with rather sparse pale appressed hairs, top of sheath

at junction with blade almost triangular, (with rounded apex),

junction of sheath and blade not sharp; auricles continuing down-

wards the sides of the triangle, about 5 mm. in lateral extent and

2 mm. high, bearing close curved pale hairs 5 mm. long; blade

erect, narrowly triangular, about 8 cm. long and hardly 1 cm.

wide at base, edges inflexed; ligule short. Leaf-blades 16—20 cm.

long, 2:°7-4-2 cm. wide, stalks of upper ones 3—4 mm. long, sur-

faces glabrous; ligule short; auricles sometimes with bristles to 5

mm. long. Spikelets terminal on leafy branches, slender, terete,

4-5 cm. long, with about 4 perfect florets and 1 or 2 imperfect

ones above them; basal rachilla-internode 4 mm. long, other inter-

nodes to 7 mm. long; Jemma smooth, 15 mm. long including a

slender tip 1 mm. long; palea a little longer, its keels well deve-

loped, very shortly and stiffly hairy, ending in free tips 0-5 mm.

long; lodicules 3, 5 mm. long, one 2 mm. wide and symmetrical.

the others a little narrower and asymmetric, translucent, with

many veins in the basal 2/3, apical part minutely punctate, edges

not fringed; stamens not seen; style at flowering a little longer than

the palea; immature fruit 10 mm. long, bearing the persistent

style, the whole 25 mm. long.

DISTRIBUTION: only known from the type specimen, collected at

Inchong Estate, on the edge of the Sungei Krian in South Kedah

(Nauen, S.F.N. 35831, S,K).

The spikelet-structure of this species resembles closely that of

S. grande Ridl., the only clear difference being the ciliate keels of

the paleas. In the spikelet examined by me the apical part was not

fully developed; there appeared to be two imperfect florets sepa-

rated by a short rachilla-internode, but further growth might result

in the development of the penultimate floret. Vegetatively the

Gardens Bulletin, S.

species is very much smaller than S. grande. The most striking

difference however, if it proves constant, is the production of

spikelets singly at the apices of leafy branches. It is possible that

other spikelets may be produced later in the axils of bracts at the

bases of the first ones, but I see no evidence of this in the type

collection.

BAMBUSA

Bambusa Schreber, Gen. Plant. ed. 8, 1: 236. 1789.

Type species: Bambos arundinacea Retz., Obs. Bot. 5: 24.

1789. McClure in Blumea Suppl. 3: 90-109. 1946. Holttum in

Taxon 5: 26-28, 65-67. 1956.

Culms erect, or in some mountain species climbing or needing

support from other plants, stout or slender, closely or fairly closely

tufted, in a few species slightly zig-zag; culm-sheaths often with

well-developed bristly auricles. Spikelets in groups of few to many

(of unequal age) at the nodes of flowering branchlets which may

bear leaves in their basal part; rachilla jointed at the base of each

fertile lemma, the internodes commonly 2 mm. or more long; basal

bracts and glumes several, one or more with axillary buds, the

upper 1-3 without buds; flowers 2 to many, uppermost 1 or 2

usually imperfect; lemmas broad and many-veined; paleas 2-

keeled, of firmer texture than in Gigantochloa and with broader

inflexed edges; lodicules 2 or 3, one smaller than the others,

usually fringed; stamens 6, filaments usually free (forming a tube

in B. heterostachya), anthers sometimes apiculate or with a tuft

of hairs at the top; ovary hairy at the apex, with a long or short

style and 1-3 slender hairy stigmas; fruit more or less cylindric

with a hairy top, slightly grooved on the side towards the palea

(groove is on line of the hilum), pericarp somewhat fleshy and

separable from the seed at the top, otherwise thin but at least

sometimes separable from the seed in a fresh condition, embryo

small, basal, its position seen through the thin pericarp.

This is a genus of many species, widely distributed in tropical

and subtropical Asia, with a few species native in Malaysia. It is

nearly allied to the tropical American genus Guadua. In Malaya

two exotic species are commonly planted and a few others less

commonly, there are two native lowland species, and on the

mountains four or five slender or climbing native species are also

known.

The characteristic feature of the spikelets is the jointed rachilla,

with distinct and often long internodes, the lemmas all being of

equal length (except the rudimentary one at the top of the spike-

let); in these characters Bambusa differs from Gigantochloa, which

Vol. XVI. (1958).

has very short rachilla-internodes and the upper lemmas progres-

sively longer. Some of the mountain species of Bambusa have

spikelets with a reduced number of flowers; in B. klossii there is

sometimes only one perfect flower, and always one in B. wrayi.

It is possible that the mountain species of Malaya (and with them

B. cornuta Munro of Java) could be separated as a distinct genus;

but at present I do not see a clear separation, and I am not sure

that B. wrayi and B. pauciflora are closely related.

In B. ridleyi two of the very large lodicules have a very much

swollen fleshy base; whether this is general in the genus, I do not

Know, as it is a character not clearly shown by dried specimens.

The ovary always has a rather fleshy hairy top, bearing the style

and stigma; this top is sometimes wider than the glabrous cylin-

dric ovary which bears it, and the latter is sometimes called a

stalk. The style varies in length. The fruit structure, seen by me in

fresh material of B. tulda, does not differ significantly from that

of Gigantochloa.

In proposing the generic name Bambusa, Schreber gave a gene-

ric description only, and cited no species (not even Arundo bam-

bos Linn.). In the addenda to his second volume he referred to

Bambos Retzius as a synonym, again without mentioning a species,

but as Retzius did describe the species Bambos arundinacea (and

no other), that must be the type species of the genus. The names

Bambusa and Bambos were published in the same year, and I have

seen no evidence as to which was in fact earlier; it is customary to

accept Bambusa as the correct name.

When describing the new genus Bambos, which he saw was

something quite different from Arundo, Retzius quoted the name

“Arundo bambos auctorum” as a synonym. It might be argued

that this meant that he regarded his species as synonymous with

Arundo bambos Linn., and therefore the meaning of this latter

Name must be considered. Linnaeus gave no description in 1753,

but quoted several former botanical works, two of them his own.

The only specimen which can be associated with any of these

earlier descriptions is almost certainly B. vulgaris; but it is prob-

able that another of the descriptions cited refers to the bamboo

here called B. arundinacea. I have therefore argued that the name

Arundo bambos L. should be regarded as a nomen confusum and

ignored. Fortunately the type specimen of Bambos arundinacea

Retz. has lately been discovered, so that the type species of the

genus Bambusa can be clearly established. For further discussion,

see the note on B. arundinacea in the following account of the

species, and the publications there cited.

Gardens Bulletin, S.

Field Characters. The species here included in Bambusa are

so diverse vegetatively that it is impossible to give any vegetative

character as distinctive of the genus. The really common species

(B. vulgaris, in every village and often on river banks, and B.

glaucescens in hedges throughout Malaya) are individually quite

distinctive; notes on habit are given under each in the account

which follows. The thorny bamboos are also at once recognizable

from their copious spreading low thorny branches; the distinctions

between the two species are given in the key. The common moun-

tain species is a slender, more or less erect, tufted bamboo of high

ridges (B. magica). The remaining three (or four) mountain

species have slender culms not strong enough to support them-

selves, and seem to be local. Of these, B. wrayi, used for blow-

pipes because it has slender internodes up to 200 cm. or more

long, appears very different, both vegetatively and in spikelet

structure, from B. pauciflora, B. montana and B. klossii; these iast

three are certainly closely allied and perhaps show relationships to

Dinochloa.

In addition to the species here described, B. tulda has long been

in cultivation in the Botanic Gardens, Singapore. There were for-

merly two old plants of it. One flowered about 1940 and died,

leaving a few seedlings in the adjacent area. The other flowered a

little on leafy branches of all its culms over a period of at least

six years, gradually dwindling in size, and finally died also.

Uses. The larger species have culms which are strong and useful

for structural purposes. Hildebrand states that those of B. vulgaris

are much inferior in durability to those of B. arundinacea

but as B. vulgaris is so abundant in Malaya, being in fact the

common large bamboo in villages everywhere, I think its culms

are often used. As judged by a section as seen with a hand-lens,

the structure of the walls in B. vulgaris is much like that of a

Gigantochloa, but I believe that B. vulgaris does not split so well

for basket-making as Gigantochloa. B. heterostachya may turn

out to be a not uncommon village bamboo in the Malacca region;

a collector has noted that it is good for baskets, and (if I am right

in identifying with it a bamboo seen near Teluk Anson) it is some-

times used for splitting into strips for other purposes.

The young shoots of B. vulgaris are said to be of good edible

quality; in fact, Hildebrand says that this is their chief use in Java.

The longest internodes of B. wrayi, a very local species, are used

for making blow-pipes by those forest peoples who have access to

them. B. magica has been credited with magic properties (being

called Buloh Périndu), but Burkill suggests that other bamboos

may have been so used also.

‘

;

Vol. XVI. (1958).

Key to the Malayan species of Bambusa (native and cultivated)

Culms bearing many thorny branches at ihe

lower nodes

Culm-sheaths with a bristle-edged auricle

on each side of base of blade; ligule

mee peered

edge a

ae gta-

dually decurrent each side along top

of sheath, decurrent part crisped and

not auricle-shaped; ligule 2 mm. tall

with frmge of fine hairs 3 mm. long

Culms not bearmg thorny branches from

lower nodes

Stout tall bamboos (to at least 6 cm. dia-

meter); culm-sheaths with broad

erect blades and distinct bristly auri-

cles

Culms plam green, or pale yellowish

_ With a few narrow green streaks;

culm-sheath ligule to 3 mm. tall,

not bristly on edge

Culm-sheath ligule 1 mm. tall; leaves

glaucous and short-hairy beneath

Culm-sheath ligule 3 mm. tall; leaves

glabrous beneath .

ee Se ee

dark and light green; culm-sheath

ligule to 6 mm. tall bearing a

fringe of bristles ¥

Slender bamboos (usually ‘not over 25

cm. diameter); culm-sheaths other-

wise

Cultivated bamboos

Leaves very finely hairy beneath, not

variegated

Leaves pale blue-green beneath;

common hedge bamboo with

culms not over 2 cm. diameter

1. B. blameana

2. B. arundinacza

3. B. burmanica

6. B. glaucescens

Leaves not pale blue-green be-

neath; often grown as a pot

plant, with dwarfed culms

and short swollen internodes;

if in open, culms- normal, to 4

cm. diameter

Leaves glabrous beneath, variegated

Wild bamboos, in forest only

Leaf-auricles 4-10 mm. tall, bearing

bristles to 20 mm. long; lowland

forest

Leaf-auricles and their bristles much

shorter; mountain plants

Longest internodes of culm to 200

cm. long

Longest internodes much shorter

Erect tufted bamboo of moun-

tain ridges; lateral branches

in dense tufts; spikelets

with several perfect flowers

Slender bamboos, not stiffly

erect; upper parts of culms

climbing and trailing, bran-

ches few; spikelets with 1 or

2 perfect flowers

Spikelets 10-12 mm. long, al-

ways with 2 perfect flow-

ers, - keels “of --paleas

smooth; below each node

a band of appressed white

hairs

Spikelets 7-9 mm. long,

sometimes with only 1

perfect flower; keels of

palea fringed; hairs below

node scattered, not form-

ing a continuous band

Fringe on palea very short;

fringe on __lodicules

hardly 0-5 mm.; smal-

ler leaves hairy

Gardens Bulletin, S.

7. B. ventricosa

8. B.sp.

9. B. ridleyi

.. 10. B. wragi

11. B. magica

. 12. B. pauciflora

. 13. B. montana

Vol. XVI. (1958).

Fringe on palea_ longer;

fringe on lodicules

longer; smaller leaves

not hairy .. 14. B. klossii

1. B. blumeana Schult., Syst. Veg. 7: 1343. 1830. Gamble in

Ann. R. Bot. Gard. Calc. 7: 50, pl. 47. Ridl. Flora 5: 256.

Synonym: B. spinosa Bl. ex Nees in Bot. Zeit. 1825: 580.

Backer, Handb. Fl. Jav. 2: 267. (Not B. spinosa Roxb. ex Buch.

Ham. 1822).

Culms tall, to about 10 cm. diameter, growing close together;

internodes glabrous when young; lateral branches borne at nearly

all nodes to the base, the lower branches spreading horizontally

and bearing stout curved spines in groups of 3 (the middle one

longer than the others and sometimes branched again), upper

branches branched at the base, obliquely ascending. Culm-sheaths

to 30 cm. long, the back covered with loose dark brown hairs;

blade of middle sheaths more or less reflexed, more than half as

long as sheath, narrowly ovate (quite narrow on upper sheaths)

and stiffly acute, convex on the back when old, back at first den-

sely hairy but glabrous when old, upper surface with more or less

persistent dark hairs in longitudinal rows, base constricted and

narrowly joined laterally to comparatively small stiff auricles bear-

ing stiff curved bristles on their edges; ligule stiff, to 5 mm. tall

(individed part), the middle tallest and shortly fringed, the outer

parts with stiff bristles to 12 mm. long (fig. 15). Leaf-blades to

about 20 by 2 cm. (on upper branches more usually 15 by 1:5

cm.), base truncate or very broadly cuneate above the short stalk,

lower surface not glaucous, sometimes short-hairy towards the

base; auricles small, sometimes bristly; ligule short, sometimes

with stiff bristles towards the ends. Spikelet-groups 1—5 cm. or

more apart; spikelets to 4-5 cm. long, the hairy rachilla zig-zag

and distinctly visible between the florets, its internodes 4 mm.

long; florets 5-12 (Backer); lemmas 6-8 mm. long, glabrous;

paleas about same length as lemmas, keels strongly fringed, 3-5

veins between the keels; anthers 4—5 mm. long; style short,

stigmas 3.

DISTRIBUTION: native in Java and eastern Malaysia (planted in Indo-

china?); sometimes planted in Malaya.

RECORDED MALAY NAME: Buloh Duri.

Flowering of this species is apparently rare in Malaya. A plant

flowered in Singapore in 1933, the only occasion on which I have

seen flowers. A large plant in the Waterfall Garden, Penang, was

never seen to flower in the period 1922-1940. There is no evid-

ence that this species ever dies after flowering, as B. arundinacea

‘gIn81] Burmoys ‘yyeoys Jo do} jo yed Jo MorA JouUT ‘q °7 X ‘apeyq pue yIeoYs Jo sjied

yusoelpe YM S[oLINe oUO JO MOIA JOINO ‘q ‘UONIsod yeInyeu UT ‘epeyq YUM “yIeoys Jo do} “D “yyeoys Jo do} pue

apryq Jo covjins Jouur ‘g “¢/] X “yIeoys pousyey ojo[duI09 Jo aovjns Joyno “y “pUDaWN]G DSNquipg JO YWeoYs-WIND “ST “Sly

Ee eae eT ee LD See obade ©

OY mn,

Gardens Bulletin, S-

Vol. XVI. (1958).

does. There are plants at the Forest Research Institute, Kepong.

Though very similar to B. arundinacea m its thorny branches,

the present species can be easily distinguished by its culm-sheaths,

as indicated m the key. Hildebrand says that B. blammeana does not

grow so large as B. arundinacea, nor are its culms so durable.

They are however superior to those of B. vulgaris.

McClure has described B. dissimulator from Honam Island.

Kwangtung, a species very similar to B. blumeana m culm-sheaths

and spikelets (Lmgnan Sci. Journ. 19: 413: 1940).

Specimens: SINGAPORE, Botanic Gardens, Ridley 3946 (S.K); Ridley

sn. 1894 (K); Pestana sm May 1933 (S). Pasanc, Pekan, Ridley

sn. 1889 (S).

2. B. arundimacea (Retz.) Willd, Sp. PL 2: 245. 1799. Gamble

in Ann. R. Bot. Gard. Calc. 7: 51, p. 48.

Basonym: Bambos arundinacea Retz., Obs. Bot. 5: 24. 1789.

Roxb., Pl. Corom. 1: 56, t. 79. 1798. Holtt. m Taxon 5: 65—67.

1956.

Synonym: Bambusa bambes auctt., supposedly based on Arundo

bambos L., a nomen confusum (Holtt. m Taxon 5: 26-28. 1956).

; Culms to 25 m. tall, to 12 cm. diameter, with branches from all

‘

1S)

oy} ysUI

Joyjour

‘9poLINne jo MdtA JOINO ‘5 ‘7 X ‘opetq JO seq pue opine Ue jo soejiNs JOU} Airey-yiep

ee o[NSI] MOT Oy} SUIMOYs “YJVOYs JO d

0] Jo ed JO MOIA JOUUT “G “F < ‘aseq sj ye posodiodns yyeoys

JO MOIA JOUUT YIM ‘YyeoYs pousHey a1a[durlos & Jo soejins Joyno “Yy “vadDUIpunsD DsNnquing JO yresys-WIND ‘QI “S14

‘f/f

Vol. XVI. (1958).

Bi,

“¢

ESN

SAN

2~ . — — im,

——s

Fig. 17. Bambusa arundinacea, top of culm-sheath and blade in natural

position, x 2/3.

DISTRIBUTION: almost throughout India, and in Burma; not com-

monly planted in Malaya but much planted in East Java; grows well

in Singapore. (The common thorny bamboo of Kwangtung is a nearly

related but apparently distinct species, B. sinospinosa McClure, Ling-

nan Sci. Journ. 19:411.1940).

As grown in Singapore, this species has larger straighter culms

than B. blumeana, more open branching, and narrower smoother

leaves; the most distinctive features are shown by the culm-sheaths.

A large clump in Singapore has not been known to flower. In

1937-38 several large clumps in the Waterfall Garden and at the

Residency, Penang, began to flower; they flowered on every new

culm for the next three years, without producing any leaves, and

finally all died. In India, plants of this species flower when about

30 years old, the flowering in any one district being gregarious;

the grains are sometimes collected as food. Backer reports that

flowering in Java is rare.

Hildebrand states that the culms are very durable, equalling

Gigantochloa apus in this, and larger. Further planting experi-

ments in Malaya are desirable, if possible with material from Java;

the species is reported as variable in India, and introductions

thence might not yield the variety best suited to Malaya. It is

doubtful whether B. arundinacea occurs in Malay villages.

The name Bambusa bambos, supposedly based on Arundo

bambos L., has been used for this species in recent years by vari-

ous authors. It is almost certain however that Linnaeus based his

Species in part on a specimen of the species later called B. vulgaris.

I have discussed this matter elsewhere (Taxon 5: 26-28) and

have shown that Arundo bambos L. should be regarded as a no-

men conjusum and therefore ignored. The true identity of Bambos

Gardens Bulletin, S.

arundinacea Retz. has also been in doubt, but has been decided

by the recent discovery of the type of the species in the Retzius

Herbarium at Lund. This shows that Retzius had specimens of

both the bamboo above described and B. vulgaris, and it is evi-

dent that he described the stamens at least from the latter. In such

cases it is necessary to select one of the two specimens as the type

of the species, and I have proposed that the specimen representing

the thorny bamboo should be the type, thus permitting the con-

tinued use of the name in the sense in which it was used from the

time of Roxburgh onwards until quite recent years. Roxburgh also

described a second thorny bamboo which he called B. spinosa

Roxb. (not to be confused with B. spinosa Bl.), but Gamble united

the two under the earlier name B. arundinacea.

Specimens: PENANG, Residency garden, Abdul Kadir S.F.N. 36177

(S,Kep.); Cheang Kok Choy. S.F.N. 37940 (S). Waterfall Garden,

Bamboo no. 7 and 11, Pestana s.n. (S). SINGAPORE, Botanic Gardens,

several collections, Pestana s.n. (S).

3. B. burmanica Gamble in Ann. R. Bot. Gard. Calc. 7: 35, pl.

33. 1896.

Culms fairly large, exposed parts of young internodes bearing

numerous spreading stiff short pale hairs and some appressed dark

hairs, also a little white waxy powder below the nodes. Culm-

sheaths green when young, the back covered with dark hairs to-

wards the base, top much upcurved at junction with blade; blade

erect, rather broadly triangular except on uppermost sheaths, base

almost cordate, apex acuminate, green turning light brown when

old, upper surface hairy at the base only; auricles spreading much

beyond sides of the sheath, 15 mm. long horizontally and 7 mm.

high, edges with curved slender pale bristles 10 mm. long; ligule

1 mm. tall, dark, edge slightly toothed. Leaf-blades to 20 cm. long

and 2-5 cm. wide, lower surface blue-green and finely short-velvet-

hairy, stalk to 4 mm. long; auricles small, lower ones sometimes

produced laterally, none seen bearing bristles; ligule hardly 1 mm.

tall. Spikelets (fide Gamble) with 5 or 6 perfect florets; ovary

with swollen hairy top, style very short, stigmas 3.

DISTRIBUTION: Burma, northern Malaya.

RECORDED MALAY NAME: Buloh Aoh Bukit.

The above description was made from a plant found by me in

a bamboo thicket near the road to Sintok in northern Kedah

(specimen in H.S.). The plant was not flowering. The culm-

sheaths match those of B. burmanica very closely, and also the

glaucous hairy lower surface of the leaves.

Vol. XVI. (1958).

4. B. vulgaris Schrad. m Wendl, Collect. Pi. 2: 26, t. 47. 1810-

Gamble m Ann. R. Bot. Gard. Calc. 7: 43, pl. 40. Backer.

Handb. FL Jav. 2: 273. Ridley, Flora 5: 256. McClure m

Blumea Suppl. [il: 108.

Culms not closely tufted, 10-20 m. tall, slightly zig-zag. at the

base 5—10 cm. diameter, walls 7—15 mm. thick, nodes promment;

internodes 20—45 cm. long, when young with appressed dark hairs

and also scales of wax, when old smooth and shmmg. Lower

culm-sheaths short, upper ones on well-grown culms 30 cm. or

more long, back densely covered with appressed nearly black

loose hairs, top slightly rounded at junction with blade; blade yel-

lowish when young, erect or bent back, rarely quite reflexed even

near base om upper surface most persistent, rather broadly trian-

gular (middle sheaths) and stiffly pomted, edges crisped towards

the more or less cordate base which is connected laterally with 2

pair of auricles 1 cm. or more high, their edges bearmg pale stout

curved bristles 7 mm. or more long; ligule 3 mm. tall or rather

more, edge slightly crisped and very shortly hairy (fig. 18). Leaf-

blades 9-30 cm. long, 10-40 mm. wide, glabrous, stalk short:

auricles small and firm, sometimes with a few bristles; ligule 1—2

mm. hich. Spikelet-groups 2—6 cm. apart, few m a group on the

smaller branchlets; mature spikelets 20-35 mm. long, laterally

flattened, 4-5 mm. wide; florets 5-10, lower rachilla-internodes

short, upper ones 1-5-3 mm. long; glumes 1 or 2, stiffly pointed,

short-hairy towards the apex; lemmas 8—10 mm. long, short-hairy

on the edges near the stiffly pointed apex; paleas a little shorter

than lemmas, keels hairy; lodicules 3, long-hairy, 2—2-5 mm. long;

anthers 6 mm. long, with a small tuft of short hairs at the tip;

style slender, 3-7 mm. long; with 3 short stigmas.

DISTRIBUTION: cultivated throughout the tropics and probably the

commonest village bamboo in all parts of Malaya. often abundant on

river banks; origin unknown.

RECORDED Matay NAMES: Buloh Minyak, Buloh Mimyak Azo. Bu-

loh Aazo Beting, Buloh Pau.

Var. striata

Basonym: B. striata Lodd. ex Lindl. Penny Cyclop. 3: 357.

1835. Hook. fil. n Bot. Mag. t. 6079. 1874.

Culms pale yellow with a varying number of narrow longitudinal

green streaks, otherwise as the typical green form but possibly

less robust. Occasional in Malay villages.

Sep Matay NAMES: Buloh Gading, Buloh Azo Gading (ivory

‘z7/¢ XK ‘apetq jo yred pue sporine ‘ansq] Zurmoys ‘yJeays JO do} Jo yied jo AMOTA

apeq jo syed jusoef[pe pue spoline ou0 JO MOIA Jano ‘gq ‘pesodiedns yyeoys Joyjoue JO

‘g0vJIns Jono “pousyey ‘yjeoys o}o,dwI05 ‘Vy ‘slupsjna psnquivg jo yieoys-wynD “gt ‘SI

youu ‘OQ ‘qyeoys pue

do} pue opeyq jo covjins Jouut quia

i) Dido

Ye? ee bi

a, or

Gardens Bulletin, S.

mal.

Prt

o T

“aay

Vol. XVI. (1958).

Hildebrand states that in Java there is also a variety with green

culms having yellow streaks of varying width; I have not seen this

in Malaya, and it appears not to have a distinctive varietal name.

Uses. Hildebrand states that in Java B. vulgaris is chiefly valued

for its edible shoots, which are of good quality. He states that the

culms are not much valued, being less durable than those of the

thorny bamboos and of the best Gigantochloas. But as this species

is so abundant in Malaya, it is probably used a good deal for

structural purposes, especially of a temporary nature. The slightly

sinuous shape of the culms may be a disadvantage for some pur-

poses; this character and the fact that the culms do not grow very

close together (as do those of most other village bamboos) is

often sufficient to distinguish plants of this species when the very

characteristic young culm-sheaths are not showing.

Hildebrand also notes that pieces of culms of B. vulgaris root

very easily, and often clumps of this species on river banks have

grown from posts set in the banks for mooring boats or other

such purpose. This may in part explain the abundance of large

clumps of B. vulgaris on river banks in Malaya. Flowering in

Malaya is not common; no fruits have been recorded.

Specimens: SINGAPORE, Colonial Secretary’s garden, Ridley 5599 (S,K).

Botanic Gardens, Ridley 6115 (S,K), 6545 (S,K), 6679 (S,K), 8079

(S,K), 8565 (S), 8899 (S,K), 8949 (S); Md. Nur & Pestana, several

Sheets from different plants (S). Perak, Grik, Hamid F.D. 6421

(S,Kep.), 6426 (S,Kep.), 6428 (S,Kep.), 7923 (K) 8255 (SK).

Blanja, Wray 141 (S,K). Taiping, Wray 5625 (S).

5. B. heterostachya (Munro) Holtt. in Journ. Arn. Arb. 27: 341.

1946.

Basonym: Gigantochloa heterostachya Munro in Trans. Linn.

Soc. 26: 125. 1868. Gamble in Ann. R. Bot.

Gard. Calc. 7: 66, pl. 57. Ridley, Flora 5: 262.

Synonyms: Gigantochloa latispiculata Gamble, l.c. 67, pl. 59.

Ridley, l.c. 263.

Bambusa latispiculata Holttum Lc. 341.

Culms commonly 6-10 m. tall (to 16 m. ?) and to 6 cm.

diameter, not closely tufted, not branched at the lower nodes,

medium green copiously and irregularly longitudinally streaked

with paler green, edges of streak not sharply defined; walls of

medium thickness. Culm-sheaths commonly 12 cm. long, black-

hairy on the back; blade erect, on middle sheaths broadly tri-

angular, commonly 7 cm. long and 5 cm. wide, apex stiff and

Narrowly acuminate, base slightly cordate, edges towards base

bearing short bristles; auricles 10 mm. high, 2-5 cm. in lateral

extent, edges slightly crisped, bearing many fine curved bristles

to 15 mm. long; ligule 6 mm. tall (not including bristles) in the

Gardens Bulletin, S.

middle, edge incised so as to be bristly throughout, bristles 3 mm.

or more long. Leaf-blades commonly 15-20 cm. long and 2-2-5

cm. wide (to 35 by 3-5 cm.), lower surface glabrous, stalk

short; auricles small, glabrous; ligule 1-2 mm. high, edge some-

times bearing a few bristles. Spikelets few in each group (usually

only one mature at a time), at maturity 4 cm. long, flat, 8-10

mm. wide, with quite smooth glumes and lemmas; rachilla-

internodes 2 mm. long, flat; lemmas to 20 mm. long, stiffly

pointed, to about 10 in number, the uppermost one or two con-

taining imperfect flowers; paleas shorter than lemmas, 2 mm.

wide between the fringed keels, about 7 veins between the keels

with distinct cross-veins; lodicules 3, 3 mm. long, with a fringe

of fine hairs over 1 mm. long; filaments of stamens joined to

form a tube which appears to disintegrate on extension; anthers

purple, 8 mm. long, with a few stiff hairs at the tip; ovary shortly

ellipsoid, the hairy upper part narrowed to a hairy style which

bears 1-3 finely hairy stigmas.

DISTRIBUTION: only known from villages in Singapore, Johore, Negri

Sembilan, Malacca and probably Lower Perak.

RECORDED MALAY NAMES: Buloh Tilan (Télang ?), B. Tilan Minyak,

B. Péring, B. Péngahit (?).

The type specimens of G. heterostachya Munro were collected

in Malacca by Griffith. The name given to this species referred

to the fact that in addition to the fully developed broad flowering

spikelets there were also slender spikelets which Munro stated

to be sterile. I have dissected similar slender spikelets on other

specimens and find them to be immature, not sterile, though it is

possible that some sterile spikelets may be formed. I can see no

other distinction between Griffith’s type specimens at Kew and

those upon which Gamble established a new species G. lati-

spiculata. Several spikelets on Griffith’s specimens are broad and

flat with many lemmas; Gamble’s figure is not representative of

these large spikelets.

Munro and Gamble included this species in Gigantochloa,

because of the stamen-tube. But the spikelet-structure, with its

jointed rachilla and many equal lemmas, is Bambusa-like, and the

stamen-tube is much more delicate when extended than that of

Gigantochloa. Bambusa-like lodicules are also present. The culm-

sheaths are very like those of Bambusa vulgaris, though smaller,

but are easily distinguished by their ligule with its toothed or

lacerate edge (as in B. blumeana and in some species of Gigan-

tochloa). The streaky culms are straighter and closer than those

of B. vulgaris; the streakiness is another Gigantochloa-like

character. | a |

Vol. XVI. (1958).

It is possible that this bamboo is not uncommon in villages in

the southern part of Malaya. I have certainly never seen it wild,

nor have I seen it in villages in the north. Alvins stated that his

original plant of G. latispiculata (collected 1886) was 50 feet

tall and used for making baskets. I should judge that for this

purpose this species would be similar in quality to the medium-

sized Gigantochloas. Alvins gave the name Buloh Tilan Minyak;

and the name B. Tilan was also given by a Malay Forest Guard

who collected the species in Negri Sembilan fifty years later.

Burkill (Dictionary, p. 1069) alters tilan to télang, which means

“marked with long streaky patches” (Wilkinson) and he may

be right in so doing, as the meaning is appropriate; the form

tilan may be local in Malacca and Negri Sembilan. Ridley had

the species flowering in Singapore in 1896 (no. 8122); he reported

the culms as 20—25 feet tall, “light green, white stripes”, and gave

the name B. rumphii var. striata. Another specimen collected in

1903 was labelled B. vulgaris. I found a flowering plant in Upper

Serangoon in 1946, and plants without flowers on Pulau Samalun

(S.W. Singapore) and near Kota Tinggi in Johore (plant brought

to the University of Malaya for cultivation, not yet flowering).

The specimen found by me near Telok Anson was in a patch

of forest not far from open country; it was not well grown, the

stems rather slender and the bristles on culm-sheath auricles

very fine, but the ligule and the shape of the auricles point to

this species. A Malay gave the name Buloh Péngait (from jahit,

to sew) and said that strips of the culm were used to attach

coconuts together, for transport (floating) by river. The same

kind of strips could be used for basket-work.

Specimens: SINGAPORE, Cult. Hort. Bot. Ridley 6680 (S,K), 8122

(S), s.n. 28.9.1903 (K). Upper Serangoon Road, Holttum s.n. Sept.

1946 (S,Kep.), Ahmad bin Hassan s.n. 27.9.1946 (S). Pulau Samalun,

Sinclair S.F.N. 38601 (S). Maracca, Ayer Panas, Griffith 6731 (K,

type). Bukit Bruang, Alvins s.n. 16.1.1886 (S, type of G. latispiculata).

No locality, Cantley’s Collector, 1717 (S). NEGRI SEMBILAN, Pera-

dong, F.Gd. Hussein F.D. 9597 (S), 9598 (S,K). Perak, Changkat

Jong, near Telok Anson, Holttum S.F.N. 38432 (S).

6. B. glaucescens (Willd.) Sieb. ex Munro in Trans. Linn. Soc.

26: 89. 1868. Holtt. in Kew Bull. no. 2, 1956: 207-211.

Basonym: Ludolfia glaucescens Willd. in Ges. Naturf. Freunde

Berl. Mag. 2: 320. 1808.

Synonyms: Arundinaria glaucescens Beauv., Essai Agrost. 144,

152. 1812. Ruprecht in Act. Acad. Caes. Petrop.

see, VI Sci nat. 5, pt. 2: 23, t. 1, f. 3. 1839.

Bambusa nana Roxb., Fl. Ind. 2nd Ed. 2: 199.

cas (nom. subnud.). Munro l.c. 89. Ridl., Flora

7258.

Gardens Bulletin, S.

Bambusa multiplex sensu Backer, Handb. Fl. Jav

2: 269 (not Arundo multiplex Lour. 1790).

Culms to about 5 m. tall and to 2 cm. diameter; longest

internodes 30-50 cm. long, covered with white waxy powder

when young, green and smooth when old; branches in rather

dense short tufts from the nodes. Culm-sheaths (without the

blade) commonly 12-15 cm. long, back smooth and hairless,

green or light red-brown when young, soon turning light buff,

apex rounded; blade always erect and pressed to the culm, nar-

rowly triangular, acuminate, the base of the triangle extending

to the full width of the rounded top of the sheath and there

often slightly auricled with a few stiff bristles on the auricles,

lower (outer) surface at first sparsely hairy, soon glabrous; ligule

to 1-5 mm. tall, edge irregularly and shortly toothed (fig. 19).

Leaf-blades about 5-12 cm. long and 6—15 mm. wide, base rather

broadly cuneate above a short stalk, lower surface glaucous and

covered with close very short velvety hairs; sheaths usually with

a small fleshy crescent-shaped auricle on one or both sides, the

auricles bearing stiff erect bristles when young; ligule low, minutely

toothed. Spikelet-groups in Malayan plants usually on the ends

of leafy branchlets, or sometimes on leafless branches; spikelets

3—4 cm. or more long when fully grown, with racilla-internodes

to 5 mm. long; lemmas 10-16 mm. long, not apiculate, glabrous,

the veins slightly raised; paleas a little shorter than lemmas, keels

very shortly fringed (usually near apex only), veins between keels :

7-8; lodicules 3-4 mm. long, narrowed to the apex, not fringed;

anthers 6 mm. long, without apical hairs; stigmas 3, arising directly _

from the swollen hairy apex of the ovary. |

DISTRIBUTION: native in China and Japan; commonly planted for

hedges in Malaya and occasionally flowering if allowed to grow un-

checked by clipping.

RECORDED MALAY NAMES: Buloh Pagar, Buloh China.

This species makes excellent hedges if well tended; but in the

neighbourhood of secondary forest, tree-seeds, brought especially

by birds, often grow in the hedge, and the trees can smother the

bamboo entirely if not watched and pulled out from time to time.

The name B. multiplex, used by Backer, is based on Arundo

multiplex Loureiro, a species which lacks a type specimen and

was only identified by Merrill with B. nana Roxb. on the evidence

of its vernacular name and its use for hedges (see Merrill’s —

Commentary on Loureiro, p. 83). Merrill states that specimens

of B. nana agree with Loureiro’s description; but Loureiro

described the leaves of his species as “6 poll . longa.. glabra...

coloris fusco-viridis”, none of which statements agree with B.

Vol. XVI. (1958).

Fig. 19. mare an oil alee glaucescens. A, complete sheath, flat-

et surface of blade and top

er urface of part of ee of sheath, cea ng

ligu #2 as ade). aie icle aid par te base of blade, x 4. D,

outer surface of same parts Apis re. .

Gardens Bulletin, S.

glaucescens (B. nana). The distinguishing features of B. glauces-

cens are the softly hairy, pale glaucous, lower surfaces of the

small leaves, and the culm-sheaths as above described (not men-

tioned by Loureiro). pie

B. glaucescens was introduced into cultivation in greenhouses

in Europe before 1800, and named from a plant which flowered

at Berlin in 1808; there is plenty of evidence that it was well

known by this name in the early years of the 19th century.

Similarly it was well known in India by Roxburgh’s name B. nana

from about the same time, having been introduced to Calcutta

from China in 1794 (Roxb., Hort. Beng.). But Roxburgh pub-

lished no description (only the statement that it was a shrubby

thornless bamboo used for hedges), and the first description

under the name B. nana is that of Munro (1868). For a full

discussion, see Holttum in Kew Bulletin no. 2, 1956.

Gamble identified Wray’s original specimen of B. magica as

B. nana, for which reason he thought this species to be native

in Malaya, but it is certainly introduced.

Specimens: SINGAPORE, Holland Road, Ridley s.n. 1898 (S). NEGRI

SEMBILAN, Seremban, Alvins 1983 (S). PERAK, Taiping, Mahmood

F.D. 9989 (S). PENANG, Government Hill, Ridley s.n. Feb. 1892 (S);

Nauen S.F.N. 38097 (S). Between Ayer Itam & Penara Bukit, Burkill

S.F.N. 1465 (S,K). No locality, Curtis 1720 (S,K).

7. B. ventricosa McClure in Lingnan Sci. Journ. 17: 57. 1938.

Culm-internodes of dwarfed pot-grown plants swollen, nearly

as wide as long, to about 2:5 cm. diameter; internodes of well-

grown plants in open ground much longer and of even thickness,

to 4 cm. diameter, the culms to 8 m. long, successive internodes

forming a slightly zig-zag pattern; long slender horizontal branches

frequent from lower nodes of such culms. Culm-sheaths some-

what flushed with purple when young, quite hairless on the back,

on the largest culms to 20 cm. long, top rounded, with a crisped

auricle to 4 mm. high on each side of the base of the blade,

edge of auricle bearing crisped bristles 4-5 mm. long; blade erect,

smooth on the back and rough-hairy on the upper surface, on

upper sheaths narrowly triangular with a slightly narrowed base

and edges stiffly inflexed towards the pointed apex, edges bristly

near the base, line of junction between sheath and blade not

distinct externally in the middle part; ligule to 5 mm. high, lower

towards its outer ends, edge sinuous and minutely hairy. Leaf-

blades commonly to 15 by 2 cm., lower ones with broad truncate

base, stalks short, lower surface very shortly soft-hairy (velvety

= er *

Vol. XVI. (1958).

to the touch), colour as upper surface; sheath-auricles usually

with fine bristles when young; ligule short.

DISTRIBUTION: probably native im China; sometimes cultivated as 2

pot plant in Malaya, and m the open ground at the Botamic Gardens,

Singapore.

I obtained a potted plant of this species m 1936; the culms

were not much more than 100 cm. tall, and the mternodes much

swollen, as above described. This plant was planted out m open

ground in the Botanic Gardens, Singapore, and im about three

years grew to quite a large clump, with tall culms having imternodes

of the cylindrical form usual in bamboos. Occasional more or less

dwarfed culms were produced among the tall ones. The characters

of the dwarfed and normal plant, and of culm-sheaths and leaves,

agree closely with McClure’s description, except that the Smga-

pore plant im the open ground grew to a much larger size than

reported by McClure in Canton. As noted by McClure, this

species is certainly a near ally of B. glawcescens, and it would

make an excellent tall hedge, either trimmed or allowed to grow

naturally.

8. Bambusa sp.

Culms slender, forming a close compact clump, to about 250

em. tall. Culm-sheaths black-hairy on the back, with small bristly

auricles on each side of the blade; blade triangular, erect when

young and somewhat spreading when old; ligule 2 mm. high with

imregularly waved and minutely hairy edge. Leaves commonly to

12 cm. long and 1 cm. wide, with several longitudinal white stripes

of unequal width, both surfaces glabrous.

Cultivated occasionally in Singapore (first seen by me near

Upper Serangoon Road). I have not been able to identify this

species, but it appears to be a Bambusa related to B. glaucescens,

probably of Chinese origin.

9. B. ridleyi Gamble in Ann. R. Bot. Gard. Cale. 7: 34, pl. 32.

1896. Ridley, Flora 5: 257, fig. 225.

Culms slender, to 5 m. tall, little branched; internodes to 45

cm. long (Ridley), when young bearing spreading hairs 05 mm.

long. Culm-sheaths about 9 cm. long, hairs pale, sparse; blade

erect, 5—7 cm. long, 2 cm. wide, base cordate, apex acuminate;

auricles spreading, ear-shaped, to 10 mm. long, bearing many

slender bristles 10 mm. or more long; ligule low, entire (fig. 204).

Leaf-blades 20-30 cm. long and 2-5-5 cm. wide, base rather

narrowly cuneate with stalk 2-3 mm. long, surfaces glabrous:

auricles large, erect, thin, 4-10 mm. tall, with marginal bristles

to 20 mm. long (fig. 208), the two auricles connected acress

Gardens Bulletin, S.

the front of the blade by the short stiff entire ligule (sometimes

only one auricle developed). Spikelets on slender branches which

may be leafy at the base, at nodes 1-15 cm. apart, few spikelets

at each node; spikelets 4-5 cm. long, with 3-5 perfect florets

and 1 or 2 imperfect ones; rachilla-internodes to 7 mm. long;

lemmas 18-22 mm. long, glabrous, many-veined; paleas shorter

than lemmas, keels slightly fringed near apex only, prolonged at

apex into two short points, veins between keels 10 or more;

lodicules 3, two 8-9 mm. long, asymmetric, fringed, the lower

A B

Fig. 20. Bambusa ridleyi. A, top of culm-sheath, with blade and auricle in

2 natural position, on a slender culm, x 5/3. B, base of a leaf-

blade and its auricle, enlarged.

]

Vol. XVI. (1958).

1/3 very much swollen, the third lodicule shorter, sometimes only

3 mm. long; anthers 8 mm. long, apiculate; ovary 2 mm. long,

style 4 mm.; stigmas 3, 10 mm. long (fig. 8, 9).

DISTRIBUTION: Singapore, Pahang.

RECORDED MALAY NAME: Buloh Akar (for the Pahang plant).

This small bamboo is distinct in its large leaf-auricles, long

lemmas and 2-pointed paleas. It was first discovered by Ridley

in the Bukit Timah forest, Singapore, where at least one plant

persists. Ridley brought a plant to the Botanic Gardens, where it

flowered in 1933, but later I could not find it. The only specimen

from outside Singapore was collected by Symington in Pahang.

This was sterile, and Symington records it as a “climbing bamboo,

rampant over a small area”, which would not agree with the

plants I have seen in Singapore, so that the Pahang plant may

represent another species, though I have never seen any other

bamboo with such leaf-auricles. Or it may be that the plants I

have seen in Singapore, in the shade of forest, are suppressed from

vigorous growth by lack of light.

Specimens: SINGAPORE, Bukit Timah, Ridley 1693 (S); s.n. 12.8.1889

(S); s.n. 1902 (S,K); s.n. 1903 (S); s.n. 1911 (K). Botanic Gardens,

Pestana S.F.N. 28316 (S.K). PAHANG, Kemasul Forest Reserve,

Symington F.D. 49810 (Kep.).

10. B. wrayi Stapf in Kew Bull. 1893: 14. Gamble in Ann. R.

Bot. Gard. Calc. 7: 49, pl. 46. Ridley, Flora 5: 259.

Culms 12-18 m. tall, about 2-5 cm. diameter (Wray), the top

drooping almost to the ground; internodes yellow, shining, longest

sometimes 200 cm. long. Culm-sheaths not known. Leaf-blades

“smooth, 15-25 cm. long, 2-3 cm. wide, base rounded above a

short stalk; auricles small; auricles and ligule when young bearing

erect bristles to 12 mm. long, those of ligule longest. Spikelet-

groups at the nodes on the ends of short or long leafy branches;

pseudo-spikelets consisting of 3 or 4 basal glume-like bracts of

increasing size (with axillary buds) separated by internodes of

1 mm., one empty glume 10 mm. long separated from the perfect

floret by a very short internode (the rachilla articulated below

the glume and below the lemma), one perfect floret, and one

small imperfect floret borne on a rachilla-internode 6 mm. long:

lemma of perfect floret glabrous, 10-11 mm. long, with about

9 veins; palea slightly shorter, or as long as lemma, 2-keeled,

with 2 veins between the keels and 3 veins between each keel

and edge, keels bearing a few short hairs; lodicules 3-5 mm. long,

narrowly elliptic, very shortly fringed; anthers at least 4 mm. long

Gardens Bulletin, S.

(immature), not apiculate; ovary narrowed gradually at the apex,

which bears 3 long slender plumed stigmas; imperfect floret con-

sisting of a lemma enclosing a small palea only.

DISTRIBUTION: only known from a restricted area on the Main

Range near the sources of the Selama and Plus rivers, at 4,500—6,000

ft. altitude.

RECORDED MALAY NAME: Buloh Bersumpitan (Blowpipe bamboo).

Wray saw blow-pipes made from single internodes of this

bamboo seven feet long, and described their construction in Perak

Museum Notes 1, no. 3, p. 54 (1894). When he visited Gunong

Inas he saw living plants in flower, and sent specimens to Kew

(no. 4166), which were described by Stapf. The only other

known specimens of this species were found by R. H. Yapp,

also on G. Inas (s.n., 20. 12. 1899, at Kew). Wray noted however

that the use of this bamboo for blowpipies is confined to the

mountainous region near G. Inas; elsewhere, and in the lowlands,

another bamboo was used, a species not then named, having

internodes to about four feet long. The only such bamboo known

to me is Schizostachyum jaculans, which see for further details.

Stapf discussed the peculiar structure of the spikelets in this

species, comparing it with Nastus, in which genus also a spikelet

contains only one perfect flower and one imperfect. But in Nastus

the several glumes, of increasing size, are all empty (without

axillary buds) whereas in B. wrayi they have axillary buds, except

the highest, which I have found empty in all spikelets examined

by me, though this is not mentioned by Stapf. Along with this

difference is another, which Stapf did not mention. In Nastus the

spikelet is a separate entity, its short stalk in the axil of a

tiny bract very different from the glumes; in Bambusa (including

B. wrayi) there is no sharp difference between the lower bracts

on the spikelet-axis and the bract in the axil of which that axis

arises. I suggest that the whole inflorescence-pattern in Nastus is

more specialized than in Bambusa; it is very much like the pattern

found in Arundinaria. The ovary in Nastus is quite similar to

that in Bambusa, for which reason I believe the genera are related.

(A paper on the genus Nastus, in which the subject is discussed

more fully, is published in the Kew Bulletin no. 4, 1955, p. 591).

Bambusa wrayi is one of a group of species in Malaya and

neighbouring countries, all characterized by a reduced number of

florets in the spikelet. These species are not necessarily closely

related; they may have arisen by an evolutionary process from

different species of Bambusa with many florets. The species in

question are: B. pauciflora, B. klossii and B. montana in Malaya; —

B. griffithiana Munro in Burma and B. cornuta Munro in Java

(see Backer, Handb. Fl. Jav. 2: 265). The three Malayan species

Vol. XVI. (1958).

just mentioned are certainly closely allied, but they do not seem

very close to B. wrayi; though they are slender climbing species,

they do not have very long internodes.

11. B. magica Ridl. in Journ. Str. Br. R. As. Soc. 44: 208. 1904.

Flora 5: 258.

Synonym: B. elegans Ridl., lc. p. 209. Flora 5: 258.

Culms slender, to 5 m. tall; internodes to more than 60 cm.

long; branches at nodes very densely clustered, slender. Culm-

sheaths to 25 cm. long, the back rather sparsely covered with

coarse brown hairs, narrowed to a rounded apex; blade at first

erect, then quite deflexed, to 10 cm. long (or more on higher

sheaths) and about 3 mm. wide, base abruptly widened to a

narrow margin along the top of the sheath, ending in a small

bristly auricle on each side (bristles often broken off); ligule 1-2

mm. high, with a fringe of stiff pale bristles up to 6 mm. long.

Leaf-blades 8-16 cm. long, 8-15 mm. wide, thin, glabrous,

glaucous beneath, stalk very short; auricles small and sometimes

bristly; ends of ligule sometimes bearing small bristles. Spikelet-

groups usually at nodes on the ends of leafy branches, the nodes

about 1 cm. apart, the primary pseudo-spikelet at each node in

the axil of a deciduous sheath 2-3 cm. long. Pseudo-spikelets

2-2-5 cm. long, secondary spikelets in the axils of the basal 1-2

bracts of the primary ones; following these bracts is usually one

empty glume and about 4 perfect florets with a terminal imperfect

one; rachilla-internodes flat, to 5 mm. long, with very short

hairs on their edges; lemmas 10-15 mm. long, apiculate, edges

shortly hairy; paleas shorter than lemmas, keeled towards apex,

keels shortly fringed; lodicules 4-5 mm. long, acuminate, slightly

hairy; anthers 4-5 mm. long.

DISTRIBUTION: a common bamboo on the higher and more expcsed

ridges of the Main Range, not known outside Malaya.

RECORDED MALAY NAME: Buloh Périndu.

Ridley described B. magica and B. elegans in the same publica-

tion; the type specimen of the former was from Gunong Berumban,

of the latter from Ulu Semangkok (near The Gap). The specimen

named B. elegans was imperfect, and Ridley did not realize its

identity with the other. The species is very abundant on the higher

ridges around both Fraser’s Hill and Cameron Highlands, and I

think there is only one such species. The above description is made

from my own collections from Cameron Highlands, and differs

im various respects from Ridley’s; but I believe it agrees with

Ridley’s specimens. The habit of the plants is not unlike that of

B. glaucescens, but the culm-sheath blades are very different, the

Gardens Bulletin, S.

leaves are glabrous, and the spikelets much shorter. Gamble

identified Wray’s original specimen of B. magica with B. glauces-

cens (see B. nana in Gamble’s Monograph).

Wray stated that this species is the Buloh Périndu (yearning

‘bamboo) of Malays, “credited with mystic properties, and much

in request by love-lorn swains whose mistresses are cold and

unresponsive” (Journ. Str. Br. R. As. Soc. 21: 159. 1890). But,

as Burkill points out (Dictionary, p. 299) this name and magic

properties may also be ascribed to other bamboos.

Specimens: PAHANG, G. Berumban, Wray 1560 (K,S,type); Ridley

13853 (S,K); Holttum S.F.N. 23403 (S,K). Cameron Highlands

Boundary Path, 5,000 ft., Holttum S.F.N. 31390 (S,Kep.); G. Brin-

chang 5,500—6,000 ft., Holttum s.n. 4.8.1946 (S); Rhododendron Hill,

5,100 ft., Henderson S.F.N. 17887 (S); Ulu Terla, Ja’amat F.D.

27612 (S,Kep.); Ulu Telom, Dolman F.D. 27251; no _ locality,

Symington F.D. 20992 (S,Kep.), Ja’amat F.D. 27030 (S,K,Kep.).

SELANGOR, Ulu Semangkok, Ridley 12114 (S,K, type of B. elegans);

top of hill, Burn-Murdoch 11926 (S,K).

12. B. pauciflora Ridl., Flora Malay Penin. 5: 259. 1925.

Culms slender, supported on trees or trailing and branching on

the ground in the forest, slender, diameter commonly under 2 cm.,

walls 3—4 mm. thick; internodes to c. 30 cm. long, a length of

1-1-5 cm. just below each node densely covered with shining

white fine appressed hairs. Culm-sheaths c. 15 cm. long, pinkish

or pinkish-brown when young, hairs on back sparse, pale brown,

fine and closely appressed, shining; blade reflexed, green and

leaf-like, one seen 8 cm. long and 1-8 cm. wide, narrowed abruptly

at base (blades of lowest sheaths probably shorter and not green);

auricles very low, extending laterally 7 mm. and slightly up-curved

at the outer ends, firm, with a few bristles near the outer ends

only; ligule consisting of an entire part hardly 1 mm. high, bearing

a fringe of stiff bristles 10 mm. long. Leaf-blades to 20 by 2:8 cm.,

stalk short, lower surface bearing scattered hairs or nearly gla-

brous; auricles small, sometimes with a few bristles; ligule short.

Spikelets borne in small groups on pendulous very slender

branches, the groups 5-50 mm. apart;spikelets 10-12 mm. long,

narrowly ellipsoid; basal bracts and glumes 3 or 4, of increasing

size; florets 3 (always?), 2 perfect and 1 imperfect; rachilla-

internodes 2 mm. long; lemmas 8 mm. long, smooth and glabrous,

broad, many-veined, very shortly apiculate; palea as long as lemma

but much narrower, 2-keeled from base to apex, keels not fringed,

veins between keels 3, and 3 between each keel and edge; lodicules

2 mm. long, blade almost circular on a narrowly oblong base,

Vol. XVI. (1958).

upper margin with fine spreading hairs to 1 mm. long; anthers

3-5 mm. long, not apiculate.

DISTRIBUTION: only known from Fraser’s Hill, on the quartzite

ridge.

RECORDED MALAY NAME: Buloh Padi (grass-like bamboo).

The original collection was made by Mohamed Nur (S.F.N.

11234), and no other flowering specimens have been found. The

above description of the culms and culm-sheaths was made by me

from specimens near the path to Pine-Tree Hill from Fraser’s Hiil

in 1953; the densely felted pale shining hairs below each node

of the culm and its branches are very distinctive and are well

shown by the type collection, so that I feel certain the plants

seen by me belong to this species. A herbarium specimen, without

flowers, was made at the same locality by Wyatt-Smith (F.D.

76177, Kep.). The culm-sheath auricles are very Gigantochloa-

like.

This species is closely related to B. montana and B. klossii, but

has larger spikelets, a different culm-sheath ligule, and the very

distinctive zone of hairs below each node. I do not doubt that

it is a quite distinct species. So far as observed, it occurs only

on the quartzite ridge at Fraser’s Hill, on which also are some

other peculiar plants not found on the granite (e.g. the ferns

Hypolepis bivalvis and Cyathea kingii, and Dacrydium falciforme ).

It may occur on the higher ridges on the granite of the Main

Range, where the soils are much leached and peat occurs, but I

have not noticed it. It is perhaps significant that B. klossii occurs

in the poor soil derived from sandstone on the upper part of

Kedah Peak, and B. montana only very locally on the higher

parts of Penang Hill, which is granite but yields a very sandy soil.

13. B. montana (Ridley) Holtt. in Kew Bull. no. 2, 1956: 206.

Basonym: Dinochloa montana Ridl. in Journ. Str. Br. R. As.

Soc. 44: 210. 1905. Flora 5: 267.

A thicket-forming bamboo with long scrambling culms (Ridley);

culm-sheaths not known. Leaf-blades to about 20 cm. long and

3 cm. wide, base slightly unequally cuneate above a stalk 3-5

mm. long, larger leaves glabrous beneath, auricles usually small

and not bristly, ligule short; small leaves on flowering branches

5 by 1 cm. to 10 by 1-5 cm. may have rather copious fine hairs

0-5 mm. long on lower surface, and slender bristles to 7 mm.

long on auricles (specimen in Herb. Kew.), but some such leaves

have very few hairs and glabrous auricles (specimen in Herb.

B.M.). Flowering branches slender, sometimes leafy at the base,

bearing spikelets in dense groups at nodes 5 mm. to 5 cm. apart,

Gardens Bulletin, S.

the smaller internodes hairy along one side, hairs curved and

reflexed, not closely appressed. Spikelets c. 9 mm. long, with 2

empty glumes above the basal bud-bearing bracts, axis jointed

above the upper glume; florets 3, the first perfect, the second

sometimes imperfect (?) and the third rudimentary, the rachilla

jointed below each floret, rachilla-internode between first and

second florets 2 mm. long (fig. 21); lemma of first floret 7 mm.

long, shortly apiculate, glabrous, many-veined; palea minutely

fringed on the keels; lodicules 3, unequal, largest 2-2 mm. long,

bearing a close fringe of hairs 0-5 mm. long; anthers 6, purple,

4.3 mm. long, not apiculate; ovary and stigmas not seen; fruit

“oblong, beaked, narrowed at base, beak hairy” (Ridley).

DISTRIBUTION: only known from Penang Hill, from three collections

made by Ridley and Curtis (apparently all from the same plant) in

December 1895, April 1896 and March 1899, nos. 7064, 7265, 10171.

A Cee

Fig. 21. Bambusa montana. A, outer view of complete spikelet, x 5; sta-

mens are protruding from lower floret. B, same with bracts and

glumes removed; upper floret seen attached to rachilla on left

(this floret is still immature and its lemma enfolds a third rudi-

mentary floret).

The above description is prepared from the sheets of no. 10171

at Singapore, Kew and the British Museum, the only sheet having

mature spikelets being that at the British Museum. Ridley reported

that he saw fruits, and six lodicules, but no fruits remain on any

of the specimens. The six lodicules were doubtless from the two

florets, only one of which would produce a fruit. The second floret

in the spikelets I dissected at the British Museum was immature.

The first specimen collected by Ridley and Curtis (7064) had —

no flowers; the second (7265) was flowering, but no mature

spikelets remain on any specimen. Both these collections are

represented only at Singapore; neither has leaves so hairy as the

Kew specimen of 10171, and neither has bristly leaf-auricles.

Vol. XVI. (1958).

This species is certainly a Bambusa, not a Dinochloa. It is very

near B. klossii and if the two are vegetatively similar they should

probably be united; the differences in spikelet-structure are slight,

and may disappear when more specimens of both are examined.

Further search for B. montana on Penang Hill should be made.

14. B. klossii Ridl., Flora Malay Penin. 5: 259. 1925.

Culms 2-5 cm. diameter (to 4 cm. ?), walls 6 mm. thick,

internodes to 40 cm. long; few white hairs near top of internode,

and a little white waxy powder below the node. Culm-sheaths

flushed purple when young, hairs on back nearly white, not very

copious, with some waxy powder; auricle forming a low rim

(5-6 mm. in lateral extent) to the top of the sheath on each

side of the blade, this rim purple, barely 1 mm. high, usually

bearing a few slender bristles 5-8 mm. long; ligule 2-3 mm.

tall, firm, with an irregular edge which bears a few very slender

bristles 5-9 mm. long (longest near ends of ligule). Leaf-blades

20-30 cm. long, 1-5—3-2 cm. wide, lower surface glabrous, stalks

of lower leaves 3 mm. long; ligule short, sometimes with a few

fine bristles, auricles usually not bristly. Flowering branches

slender, sometimes laxly branched, sometimes unbranched at the

ends of leafy branchlets, the ultimate flowering branchlets to

30 cm. or more long, bearing dense groups of spikelets at nodes

5S mm. to 5 cm. apart; distal internodes hairy along one side.

Spikelets 7-8 mm. long; perfect florets 1 or 2, in the latter case

separated by a rachilla-internode 1-2 mm. long, with a rudi-

mentary terminal floret jointed to a similar rachilla-internode;

lemma of perfect floret 6-7 mm. long, shortly apiculate, glabrous;

palea keeled, keels fringed, 2 veins between the keels and 2

between each keel and edge; lodicules 3, obovate, unequal, largest

2-25 mm. long, long-fringed; anthers 4. mm. long, purple, not

apiculate; ovary glabrous, with 3 hairy stigmas gradually diverging

from a short hairy style.

DISTRIBUTION: Kedah Peak, at about 3,000 ft. altitude.

The only two collections were made by Robinson and Kloss

(6069, 6109), both represented only at Kew (no. 6069 is the

type; Ridley labelled the other, with a query, Oxytenanthera

sinuata). 1 have dissected spikelets of both, and see no essential

difference.

The above description of culms and culm-sheaths was made by

me from living plants when I visited Kedah Peak in January 1954

(specimen K. 16, S.). The plants were not flowering, but they

Gardens Bulletin, S.

were abundant at about 3,000 ft. and I saw no other bamboos

at that elevation. The culm-sheaths are much like those of B.

pauciflora.

The spikelets of B. klossii (1 have only dissected one from

each specimen) differ little from those of B. montana, and it is

possible that the two species should be united under the name

B. montana, which is the earlier.

THYRSOSTACHYS

Thyrsostachys Gamble in Indian Forester 20: 1. 1894.

Type species: T. oliveri Gamble, |.c. Bor in Indian For. Rec.

N.S, Bot. 2:°222.. 3944.

Spikelets consisting of 2 or 3 perfect florets and an extension

of the rachilla bearing a minute rudimentary floret; palea of lower

florets deeply 2-cleft, with narrow tail-like divisions, 2-keeled,

ciliate on the keels, palea of uppermost floret not sharply keeled

and only a little cleft; lodicules 0, 2 or 3, narow; stamens 6,

filaments free, anthers with a short tip; ovary small, cylindrical,

its fleshy apex swollen, glabrous or nearly so, bearing a long

tapering glabrous style; stigmas 3, hairy; fruit cylindric, glabrous,

beaked.

This is a genus of two known species, from Burma and Siam.

These species are certainly very close to Bambusa, and have the

same type of ovary (except that the swollen top is almost or quite

glabrous); they agree with certain species of Bambusa in having

only 2 or 3 perfect florets, but differ in the deeply cleft paleas,

and also apparently in the extremely reduced nature of the

terminal floret. Only one species is known in Malaya; it is not

native, but is common in gardens.

T. siamensis Gamble in Ann. R. Bot. Gard. Calc. 7: 59, pl. 51.

1896.

Culms commonly 8-10 m. tall, densely tufted, basal part not or

little branched, 4-5 cm. diameter near the base, very thick-walled;

longest internodes about 30 cm. long, green. Culm-sheaths to 20

cm. long, pale, thin, persistent, back covered with inconspicuous

short pale appressed hairs, edges fringed towards apex when young,

line of junction with blade distinct, raised slightly in the middle

where blade is attached, upcurved again a little on each side of

base of blade; auricles very small; blade erect, narrowly triangular,

edges incurved, base slightly narrowed; ligule low, slightly and

irregularly toothed, glabrous. Leaf-blades commonly to about 12

cm. long and 8 mm. wide, glabrous or sometimes sparsely hairy,

Vol. XVI. (1958).

lower surface of same colour as upper, short-stalked; auricles very

small, sometimes with a few short thin bristles; ligule short, entire.

Spikelets in small groups in the axils of persistent short sheaths

on slender branchlets, or in larger groups on larger branches, each

spikelet about 17 mm. long, the bracts and glumes finely hairy;

perfect florets usually 2; internode between florets short; rachilla-

extension at top of spikelet 6 mm. long, thickened at the tip

(rudimentary floret not distinctly developed); palea of lower floret

cleft half way or more towards the base; top of ovary and style

apparently quite glabrous, base of style thick and persistent as a

beak on the fruit; lodicules 1-5 mm. long, narrow.

DISTRIBUTION: Burma and Siam; often cultivated in Malaya as an

ornamental plant.

This is a very graceful bamboo, owing to its compact clumps

of out-curving slender culms, bearing many small leaves on

slender branches. It very rarely flowers in Malaya. There are some

very poor specimens of a clump that flowered in Singapore after

transplanting about 1915; the flowers are not clearly distinguish-

able. Some plants transplanted to fill gaps in a hedge in Dalvey

Road, Singapore, flowered well in 1949 (Holttum, S.F.N. 37943);

and several old undisturbed clumps flowered simultaneously in

Penang in 1948 (Holttum, S.F.N. 38450).

It may be that the first name published for this\species was

Bambusa regia Thomson (ex Munro in Trans. Linn. Soc. 26:

116. 1868); but the description was from sterile specimens and

no good description of culm-sheaths was included, so that it is

open to doubt.

DINOCHLOA

Dinochloa Buse in Miquel, Plant. Junghuhn. 388. 1854.

Type species: Nastus (?) tjankorreh Schult., Syst. Veg. 7: 1358.

1830, from Java.

Climbing bamboos, with zig-zag culms; culm-sheaths thick, when

dry wrinkled at the base where a leathery ring of uneven width

remains after the sheath falls; Jeaves with distinct transverse veins

(seen in dried specimens on lower surface). Spikelets very short,

borne in groups at the nodes of slender leafless branched stems,

one-flowered, rachilla not produced beyond the flower; glumes

(lacking axillary buds) 1 or 2, ventricose with a distinct stiff tip;

lemma similar, commonly about 4 mm. long; palea broad and

convolute, usually longer than the lemma and thinner in texture,

not keeled but very slightly 2-pointed, the two points touching;

lodicules absent (present in Philippine species); stamens 6, free,

Gardens Bulletin, S.

short; ovary ovoid, glabrous, its widened apex gradually attenuate

upwards (style-like) and bearing 3 long slightly plumose stigmas;

fruit ovoid, shortly beaked, or spherical, longer than lemma and

palea; pericarp thin except at top of fruit, smooth and dry when

ripe, or fleshy and of even thickness throughout; seed consisting

mainly of endosperm, with embryo about 1/3 of its length, in

dry fruits (lacking endosperm in fleshy fruits ?).

A genus of few species, at low elevations in forest in Burma,

Indochina, Hainan, the Andamans and Nicobars, Western Malaysia

and the Philippines; in Malaya apparently represented by two

species which are not abundant and rarely flower. Ridley’s D.

montana, from Penang Hill, is a species of Bambusa.

The spikelets are not only reduced to a single floret; the lemma,

palea and floral parts are much shorter than in other bamboos.

As in other such cases in many plant families, the result of this

reduction is that characters which might indicate relationships to

other genera are so modified as to be difficult to interpret. The

traditional position of this genus is near Schizostachyum and

Melocanna, where it was placed by Munro, mainly on the evidence

of the fruit (which he never saw, relying on the description of

Schultes). The question of the fruit will be considered later; the

subjects of spikelet-morphology and ovary-structure need first to

be considered.

Bentham (Genera Plantarum 3: 1214) said he could see no

palea in Dinochloa. He was looking for a two-keeled structure,

which is certainly not to be found. But Munro and Gamble both

interpreted the Dinochloa spikelet as having one perfect floret

enclosed by lemma and palea, and that interpretation is here

accepted. Presumably Bentham thought the organ here regarded

as the lemma of the perfect floret to be the lemma of a missing

floret, the organ here regarded as the palea to be a lemma enclosing

floral parts, the true lemma being lacking. A similar interpretation

has been given in the case of Oryza (rice) but in the case of no

other genus in the sub-family Pooidez, and the case of Oryza is

doubtful.

In the genus Chloothamnus there is only one perfect floret and

no extension of the rachilla beyond it; the palea is quite unkeeled,

but has a slight thickening of two veins just at the tip, giving a

broad and very slightly two-pointed apex. Exactly this condition

occurs also in Dinochloa. In Bambusa (to which, on ovary-form,

I believe Chloothamnus to be related) there is always a more or

less reduced terminal floret on a rather thick rachilla-extension,

and the palea of the uppermost perfect floret is always strongly

keeled throughout. The two keels converge and almost meet at

Vol. XVI. (1958).

the apex of the palea, which is thus formed by the thickened

tips of the two keels almost touching. If one could smooth out

the rest of the keels and leave the tips only, one would have the

condition of Chloothamnus; and I believe that Dinochloa shows

the same characters, though in a reduced condition.

Dinochloa has traditionally been placed near Schizostachyum,

but the palea in that genus is distinctly different (again, Bentham

doubted if it was really a palea, presumably because it is not

two-keeled throughout). In Schizostachyum the sole or uppermost

perfect floret has a palea which is normally more or less grooved

near the apex, rarely to the base, the edges of the groove hardly

forming keels; at the apex there are always two distinct terete

(almost spine-like) tips which represent the ends of the keels.

These two tips are usually separate (sometimes almost contiguous )

and do not converge as in Bambusa. Thus the evidence of the

palea would place Dinochloa near Bambusa rather than near

Schizostachyum.

The ovary in Dinochloa has a fleshy apex which is gradually

attenuate upwards (thus forming what is called a short style)

with three stigmas at its apex; this is very much like the condition

of some Bambusa species except that the top of the ovary in

Dinochloa is not hairy. I regard the evidence of the ovary therefore

as pointing to affinity with Bambusa.

The fruit in Dinochloa scandens, as described from Java speci-

mens by Backer (the type was from Java, and Backer only reckons

one species in that island) is longer than the lemma and palea

(this is not surprising, in view of their small size), shining brown,

the embryo 1/2-1/3 as long as the fruit. Apparently the sides

of the pericarp are thin enough to show the position of the

embryo in an entire dry fruit. A single fruit of D. scandens var.

andamanica (Kurz) from Gamble’s herbarium agrees with this

description. The sides of the pericarp are quite thin, smooth and

shining externally; the seed consists mainly of endosperm, the

embryo being proportionately larger than in most bamboo seeds

but not otherwise abnormal.

The fruit in a Dinochloa from Borneo and the Philippines,

named D. scandens by Gamble, is quite different. It is larger, and

quite spherical, with a fleshy pericarp of even thickness throughout.

The surface of the pericarp of a dried fruit is wrinkled, not smooth.

The spikelets bearing such fruits are not different in essentials from

those of D. scandens var. andamanica. It thus appears that two

types of fruit occur in one genus. But the species bearing fleshy

fruits cannot be D. scandens; it needs a new name, and a new

description made from living plants.

Gardens Bulletin, S.

I conclude therefore that the genus Dinochloa is probably related

to Bambusa; and it is perhaps significant that the mountain species

of Bambusa in Malaya which have a reduced number of florets

in the spikelet also have slender more or less climbing stems

tending towards the condition of Dinochloa.

The rather distinct cross-veins in the leaves do not suggest a

near relationship to Bambusa, but this character in bamboos is

difficult to describe clearly from external features (it should be

described from anatomical features) nor is its value clear in

assessing relationships.

McClure, in Kew Bulletin 1936, p. 253, included Melocalamus

compactiflorus in Dinochloa. In my opinion he was wrong in so

doing, and I can only conclude that he did not dissect spikelets

of the type material, which clearly have two perfect florets (in

some other specimens there are three) and a rachilla-extension

with a rudiment at its tip, the paleas of all the perfect florets

normally two-keeled, a condition quite incompatible with Dinoch-

loa and with McClure’s description, but agreeing with descriptions

by Kurz and Gamble.

Species in the Malay Peninsula

Leaves commonly to 20 cm. long and 3 cm.

wide; leaf-auricles and ligules usually bear-

ing bristles; smaller branches not very rough

below the nodes .. fa .. 1. Dieseamadens

Leaves (at least on lower branches) to 5:5 cm.

wide; leaf-auricles and ligules lacking

bristles; smaller branches below the nodes

very rough from the bases of fallen hairs

(probably densely hairy when young) -.. 2. D. sp.

1. D. scandens (Bl.) O. Ktze, Rev. Gen. Pl. 773. 1891. Backer,

Handb. Fl. Jav. 2: 289. '

Basonym: Bambusa scandens Bl. ex Nees in Flora 7: 291.

1824.

Synonyms: Nastus ? tjankorreh Schult., Syst. Veg. 7: 1358.

1830.

Dinochloa tjankorreh Buse in Miquel, Pl. Jungh.

388. 1854. Gamble in Ann. R. Bot. Gard. Calc.

7: 112, pl. 98. Ridl. Flora 5: 267.

Culms climbing, the lower part unbranched and 8-25 mm.

diameter (Backer), with prominent nodes, internodes when young

bearing scattered ascending appressed stiff pale hairs, the bases

Vol. XVI. (1958).

of which later leave scattered minute warts. Upper culm-sheaths

(fide Backer, not seen in Malayan specimens) 9-12 cm. long,

narrowed and gaping towards the apex, suffused with purple,

hairs white to purplish; ligule 1-1-5 mm. high with shortly fringed

edge; auricles apparently low and firm, as in Bambusa pauciflora

(sometimes fringed with bristles?); blade at first erect, then

spreading, ovate-lanceolate, 5-8 cm. long and 10-20 mm. wide,

base rounded or slightly cordate. Leaf-blades on upper branches

commonly to 20 cm. long and 3 cm. wide, base cuneate, stalk

short; sheaths hairy when young, usually with a few bristles on the

small auricles; ligule short, bearing a few bristles each side near the

auricles. Larger spikelet-bearing branches somewhat rough (like

the culms), smaller ones quite smooth; pseudospikelets to about

5 mm. long and 2:5 mm. wide, terete, acute; lemma very broad

with a rounded top having a very short tip; palea longer than

lemma, 4—4:5 mm. long, very broad, apparently glabrous (Backer

says very shortly fringed), veins numerous, with cross-veins; fila-

ments short, anthers 4 mm. long including a distinct slender tip

bearing many short hairs; ovary short, narrowed to a (glabrous? )

style 1 mm. long which bears 3 gradually diverging stigmas; fruit

7 mm. long (Backer).

DISTRIBUTION: Western Malaysia (type from West Java) and also in

the region immediately north of Malaya; in Malaya collected only once

in flower, at Lumut in Perak (Ridley 3112; S,K). This flowering spe-

cimen has rather small groups of spikelets on very slender branches.

The discrimination of species in this genus is difficult, and it is possible

that the plants in Malaya differ from those in Java (from which part

of the above description, as indicated, is derived), in which case it may

be shown that they represent a distinct species. Further flowering

material, and full notes on vegetative parts of plants in Malaya, are

needed. The figures of lemma and palea given by Munro (Trans. Linn.

Soc. 26: pl. 5, figs. 4, 5), showing fringes of stiff hairs, are like no-

thing I have seen.

2. Dinochloa, sp. ined.

Specimens of what appears to be a second (and probably

undescribed) species of Dinochloa have been found at three

localities in southern Malaya; their distinctive characters are shown

in the key. The plant I saw on Gunong Panti had long unbranched

stems trailing along the ground in the forest, bearing leaves; I

could get no specimen of climbing stems, which would doubtless

have leafy branches at the nodes, the branch-leaves smaller than

those on the main stem which I collected. It looks as if other

specimens are from similar material. The specimens are:

Jouore, G. Panti, Holttum s.n. 8.9.1929 (S); Corner s.n. 13.10.1935

(S). G. Pulai, Ridley 12150 (S,K). Maracca, Bukit Lenggeh F.R.,

Dolman, F.D. 25254 (S).

Gardens Bulletin, S.

DENDROCALAMUS

Dendrocalamus Nees in Linnaea 9: 476. 1834.

Type species: Bambusa stricta Roxb., Corom. Pl. 1: 58. 1798.

Culms forming dense clumps, in the more slender species with

drooping upper parts; in D. strictus with branches often from

lower parts of culms, in other species not. Culm-sheaths usually

with rather pale hairs (sometimes sparse), in several species

mixed with white waxy powder; blade rigid, not green and leaf-

like on middle sheaths. Spikelets in dense groups at the nodes of

usually leafless branches; perfect florets 1-6, the apical floret

sometimes sterile or reduced and then on an elongated extension of

the rachilla (this character not constant even in one spikelet-

group), rachilla-internodes between perfect florets very short and

not articulated below the florets; lemmas broad, glabrous or finely

hairy, not long-fringed on the edges, many-veined, progressively

longer towards apex of spikelet; paleas of lower florets 2-keeled,

thin and translucent, veins between keels few, keels usually

fringed with hairs, sometimes hairs on back and edges also; palea

of uppermost perfect floret (or of sole floret if only one) not or

only slightly keeled, often glabrous; stamens 6, with free or more

or less completely joined filaments (the extended tube, if present,

very thin), anthers not or shortly pointed at the apex; lodicules

rare (not known in Malayan species); ovary with hairy apex

(except apparently in D. elegans) bearing a long slender hairy

style with usually an undivided stigma; fruit (seen in D. pendulus)

not longer than lemma and palea, pericarp thick at top of fruit

and easily separable from seed, very thin towards base of fruit;

embryo small and round, hilum linear, extending whole length of

seed (for floral details, see figs. 23 and 26).

A fairly large genus, widely distributed in India (especially the

north-east), Ceylon, Burma, Siam, Indochina and southern China,

apparently not native in Java and doubtfully in Borneo, but re-

ported in the Philippines and New Guinea; in Malaya represented

by one very abundant species, two of less clear status, and two

slender species on limestone in the north. In Malaya also one

species is very commonly planted (D. asper) and two others (D.

strictus and D. giganteus) less commonly; in the Botanic Gardens,

Singapore, are also D. hamiltoni and D. latiflorus, which I have

not included in the present treatment, as they are unlikely to be

generally planted.

Vol. XVI. (1958).

I would include in this genus most of the Indian species placed

by Gamble (following Munro) in Oxytenanthera (the others are

Gigantochloa). The type species of Oxytenanthera is O. abyssinica

(Rich.) Munro. This has an ovary very different from that of the

Asiatic species referred to the genus by Gamble, for which reason

I consider that these Asiatic species should be in another genus.

The only reason for separating them from Dendrocalamus is the

stamen-tube; but, as stamen-tubes occur also in other very widely

different genera of bamboos, I do not consider the presence of a

stamen-tube alone a sufficient character for the delimitation of

genera. In general spikelet-structure, and in the form of the ovary,

these species do not differ from Dendrocalamus strictus, the type

species of the genus. I have found it very difficult to see whether

in fact a stamen-tube is present in our native Malayan species; the

tube, if present, is much more delicate than that of Gigantochloa,

in which genus it is invariably firm and very distinct, even when

extended.

The type species, D. strictus, has two or three perfect florets;

all the native Malayan species except D. asper (and this is doubt-

fully native) have one, two or three florets. There are two species,

D. asper and D. giganteus, which have spikelets of a different

aspect (with looser or hairy lemmas) and containing a larger

number of florets; the culms also of these species are very large

and more rigid than the others. If one had only the species here

considered, there would be a good case for separating D. asper

and D. giganteus in a distinct genus from the others. But, looking

over the Indian species (including those now called Oxytenan-

thera) I do not see how to define two genera clearly. McClure has

proposed a new genus, Sinocalamus, the type species of which,

Dendrocalamus latiflorus Munro, is very like D. asper and would

certainly be in the same genus as D. asper (though McClure later

included D. asper in Gigantochloa). McClure states that his genus

differs from Dendrocalamus in having lodicules; but in the type

species there are none. The only possible distinctive character I

can see in his diagnosis is the broad ventricose shape of the lem-

mas, and this is not a precisely definable character; I could not

use it to my satisfaction as a means of sepatrating the species of

Dendrocalamus (in the broad sense here used) into two parts.

Field characters (native species). The long slender shining dark

green culms of D. pendulus, their upper parts curved downwards

or resting on forest trees, bearing long slender leafy pendulous

branches, are abundant and distinctive in the lower valleys of the

Main Range as one travels along the roads to Ginting Simpah,

The Gap, and Cameron Highlands. The white-waxy sheaths of

Gardens Bulletin, S.

erect unbranched young culms, with their rigid pinkish blades, are

also distinctive. The hairs on the sheaths also are nearly white,

but it may be (as certainly in D. hirtellus) that they are darker

on lower sheaths. There is no question that D. pendulus is an

abundant and distinct species, all along the Main Range. It is not

a village bamboo.

In Johore and Kedah, and in various other parts of Malaya

away from the Main Range, occur wild plants of Dendrocalamus

which are closely related to D. pendulus, with very similar culm-

sheaths and spikelets, but of more erect habit, with softly hairy,

usually larger, leaves. I believe also that these other plants differ

from D. pendulus in having a conspicuous white waxy powder on

the young culms (as distinct from the sheaths) and in having

darker (often fewer) hairs on the sheaths. In Kedah I found much

variation in size of ligule and auricles on the culm-sheaths, and

could not see any useful character in these organs by which to

distinguish species. In fact, the only apparently clear characters

by which one can separate these large-hairy-leaved native Dendro-

calamus plants into two species are in the number of flowers in a

spikelet and in some floral details; and these are not practicable

field characters. There is need for further study of these plants.

In Kedah and Langkawi, especially on limestone hills, are

slender short plants of Dendrocalamus which have been described

as two species; but again, I am not sure how good are the apparent

distinctions between them.

Uses. The principal useful species is D. asper, which is culti-

vated by Chinese market gardeners in most parts of Malaya for its

excellent and large edible shoots. It can also be found in Malay

villages, but I do not think it is systematically planted, and Malays

also eat the shoots of other bamboos. The only other species I

have found to be (occasionally) planted for this purpose in Singa-

pore is Gigantochloa levis, which resembles D. asper in several

ways and is said also to have shoots of excellent edible quality;

for notes on the distinctions between this and D. asper, see the

description of G. levis. The old culms of D. asper are large and

strong, useful for building purposes, but probably not abundant

enough to be much used, as only a limited number are allowed to

grow on each plant.

I found that D. pendulus was used for basket-making near the

Cameron Highlands road, but it was not the bamboo most used

for that purpose. Its long culms of very even thickness (not too

stout) look as if they would make useful poles, if fully mature and

Vol. XVI. (1958).

suitably treated, but probably they are not as strong as the culms

of D. strictus.

D. strictus, an Indian species, is an important source of bamboo

poles, and might profitably be more planted in Malaya for that

purpose. At present, the only plants of it that I know are several

large clumps in the Public Gardens at Kuala Lumpur, and a few

in the Botanic Gardens, Singapore.

D. giganteus, probably the largest of all bamboos, was reported

by Ridley as native in Malaya, but I have failed to trace the speci-

mens he quotes and have seen no others. There is a large clump at

the Forest Research Institute, Kepong, from which others could be

propagated. The culms are probably too large for some purposes,

but should be useful for building. Where the species is native, the

lower internodes are used as buckets.

Key based mainly on spikelet characters

Spikelets all with one floret, or some with 2

florets, glabrous

Palea of single (or upper) floret with 4

veins; stamens 6

Anthers 4 mm. long; tall bamboos

Leaves hairless on lower surface, 2-3

cm. wide x .. 1. D. pendulus

Leaves softly hairy on lower surface,

3—6 cm. wide + .. 2. D. hirtellus

Anthers 2:5 mm. long; slender small

bamboo - es .. 3. D. elegans

Palea of single floret with 7 veins; stamens

variable, 3-6 (?) ie .. 4. D. dumosus

Spikelets all with 2 or more florets; lemmas

more or less hairy

Florets 2 or 3

Native species; filaments united to form a

tube 7 es .. 5. D. sinuatus

Planted, introduced species, not common

in Malaya; filaments free .. 6. D. strictus

Florets at least 4, usually 5 or 6

Spikelets 6-9 mm. tong... <5.» iene aSper

Spikelets 15-20 mm. long .. .. 8. D. giganteus

Gardens Bulletin, Ss.

Key based mainly on vegetative characters

Very large culms, 18—25 cm. diameter; sai

planted, doubtfully native .. 8. D. giganteus

Culms smaller

Culms commonly 12 cm. diameter, basal

internodes densely brown-hairy .. 71. tes

Culms more slender, basal internodes not so

hairy

Auricles of culm-sheaths not bristly;

planted, not native 53 .. 6. Dace

Auricles of culm-sheaths bristly; native

bamboos, rarely planted

Sheaths or young culms (or both)

copiously white-waxy; culms c. 6—9

cm. diameter

Very long slender forest bamboo,

needing support of trees; young

culms not conspicuously waxy;

leaves glabrous... .. 1. D. pendulus

Less long and more erect, in more

open places; young culms white-

waxy; leaves softly hairy beneath

Spikelets with 1 or 2 flowers;

palea glabrous .. . 2. D. hirtellus

Spikelets with 2 or 3 flowers; eh

hairy .. 5. Dy SSeS

Sheaths and young aie not white-

waxy; culms 2:5 cm. diameter

Leaves soft-hairy beneath .4. 3. D. elegans

Leaves not hairy beneath .. 4. D. dumosus

1. D. pendulus Rid. in Journ. Str. Br. R. As. Soc. 44: 210. 1905.

Flora 5: 266.

Synonyms: Schizostachyum subcordatum Ridl. in J.S.B.R.A.S.

82: 204. 1920. Flora 5: 269. 3

Cephalostachyum malayanum Ridl. in J.S.B.R.A.S.

57: 118. 1910. Flore 5: 207.

Culms to at least 30 m. long, to about 9 cm. diameter near base,

when old smooth and dark green, nodes somewhat prominent, the

lower ones with many roots; branches long, slender, pendulous.

Gardens Bulletin, S.

“HW purysq opey{q JO oseq jo y1ed pue ojorine jo oZULIy YUM ‘Opn Jo aed SUIMOYS

‘MOIA JOUUI ‘Cl *¥Z X ‘opelq Jo pur yieoys

sovjins JouUl ‘gq ‘¢/, X ‘pousey ‘yeoys 9

Jo do} jo sjivd jus0elpe pue gfHne ue JO MIIA JING ‘O ‘ouIeS Jo

Jo[MWIOD JO sdRJINS JajnNoO ‘y ‘snjnpuad snupjpr04puaq JO yyeoys-wynd

Theis, haba beaile

My Mee

‘CZ “BIA

Vol. XVI. (1958).

Culm-sheaths to 25 cm. long, rigid except near edges at top, olive

green at base grading to pinkish at the top, covered with white

waxy powder and with copious felted almost white hairs; auricles

dark, to about 15 mm. in lateral extent and 3—4 mm. tall, upper

edge bearing a close row of slender slightly flexuous pale bristles

15—20 mm. long; blade to about 25 by 8 cm., glabrous, pinkish,

obliquely ascending, edges incurved, apex acuminate, base slightly

contracted; ligule 5-10 mm. tall, firm, edge irregularly toothed

but not bristly, usually lower in the middle than towards the ends

(fig. 22). Leaf-blades of smaller branches small, glabrous on

lower surface, with few or no bristles on the auricles, of larger

branches 15-30 cm. long and 2-3 cm. wide, all leaves with very

broadly rounded or truncate base (sometimes almost cordate)

above a stalk 2-6 mm. long (distal leaves sometimes cuneate at

the base); auricles of larger leaves sometimes bearing bristles to

10 mm. long; ligule short. Spikelets in dense heads at the nodes of

long slender branches, each spikelet about 10 mm. long, glabrous,

with a single perfect floret (rarely 2) and several smaller glumes

below it (fig. 23); lemma 8-9 mm. long, smooth, with a short

stiff point, veins about 12; palea of sole floret (or of upper floret

if there are 2) thin, glabrous, not keeled, with about 4 veins all

near the middle; palea of lower floret (if present) keeled but not

fringed; anthers 4 mm. long, rather broad, shortly pointed, tips

not hairy, filaments very delicate (forming a tube?); fruit 4 mm.

long, 1:5 mm. wide, smooth except for the truncate top which has

short hairs and bears a short hairy beak (base of style).

DISTRIBUTION: very abundant in the valleys of the Main Range, not

recorded outside Malaya.

RECORDED MALAY NAME: Buloh akar (in common with all long

slender bamboos).

Uses. Used for making baskets by Cameron Highlands road,

but not used so much as a Gigantochloa. Doubtless used also by

forest peoples in many ways, but no study of its particular useful-

ness has been made. The name Buloh akar is used for some of the

slender species of Schizostachyum, and so references to uses of a

bamboo with that name may not apply to D. pendulus.

This species is undoubtedly very abundant, and of quite uniform

character. An aberrant specimen was collected by Symington from

Ulu S. Termin (route to G. Korbu, Perak, F.D. no. 32264) at

2,800 ft. The diameter of the culm was said to be 4—5 inches,

which is larger than any specimen measured by me, and the lemma

of the (single) floret is only 7 mm. long.

The type specimen of Schizostachyum subcordatum Ridl. was

part of the original collection of Dendrocalamus pendulus (Ridley

no. 8482) sent to Kew unnamed, later discovered by Ridley when

Vol. XVI. (1958).

A B oe D E F

Fig. 23. A, Dendrocalamus pendulus, complete mature spikelet (2 empty

glumes and a lemma), x 5. B, palea of the single floret from

A. C, shrivelled not quite mature fruit from A. D, D. asper,

palea of uppermost floret. E, D. asper, palea of floret next below

that shown in D. F, ovary, style and stigma of D. giganteus,

x 44.

he was preparing his Flora and re-described. The type specimen of

Cephalostachyum malayanum Ridl. has large galled spikelet-

groups at the ends of leafy branchlets; the spikelets are hypertro-

phied and each flower contains an insect larva. The arrangement

of the parts of the spikelets is exactly as in D. pendulus, but lem-

mas and paleas are very long. The leaves are closely similar to

those of normal D. pendulus, and I do not think there can be any

doubt of the identity of the specimen; it is certainly no Cephalo-

stachyum.

Specimens: SINGAPORE, Botanic Gardens, Ridley s.n. 1908 (S); Md.

Nur s.n. 1922 (S); Pestana s.n. 1933 (S). NEGRI SEMBILAN, Jelebu,

Md. Tahir F.D. 9595 (S). Kenabai Road 39? mile, Buyong F.D. 11003

(K). Ulu Bendul, Holttum S.F.N. 9796 (S,K). Seremban District, per

A. Arber (K); Kinsey s.n. (K). SELANGOR, Ulu Gombak, Foxworthy

F.D. 6431 (S); Burkill S.F.N. 9973 (S,K). Pahang Track, Ridley 8482

(Type, S,K). Ginting Simpah, Holttum S.F.N. 38413 (S); Hume

8434A (S). Ulu Gombak, Hume 8908 (galled infl., S), 9067 (S),

9491 (S). PERAK, Temango, Ridley 14350 (with galled infil., type of

Cephalostachyum malayanum, S,K); Ridley 14384 (S). Jor Track,

Ridley s.n. Nov. 1908 (S). Grik, Hamid F.D. 6427 (S), 11616 (S). S.

Termin, 2,800 ft., Symington F.D. 32264 (Kep.). Fraser’s Hill, Bur-

kill & Holttum S.F.N. 7892 (S). Cameron Highlands Road 1,000 ft.,

Holttum S.F.N. 38418 (S). PAHANG, Temerloh, Hamid F.D. 10581

(K). Rompin, Md. Soh F.D. 15456 (S). P. Tioman, G. Kajang, Md.

ae S.F.N. 18914 (S). KELANTAN, Kuala Lebir, Ridley s.n. 12.2.1917

2. D. hirtellus Ridl. in Journ. Str. Br. R. As. Soc. 73: 146. 1916.

Flora 5: 266.

Culms to about 15 m. tall and 8 cm. diameter (perhaps also

larger), internodes white-waxy when young, apical part of culm

very slender and long-drooping with small leaves (7 cm. long).

Gardens Bulletin, S.

Culm-sheaths dull orange or yellowish towards apex, green at

base, covered with pale hairs and white waxy powder when young,

the middle ones soon hairless or nearly so, the lowest having

brown hairs; auricles dark green, more or less crisped, to about

15 mm. in lateral extent and 3 mm. high, edges bearing close

nearly straight pale bristles 20-25 mm. long; blade of middle

sheaths purplish towards base, green distally, commonly 18 cm.

long and 2-5 cm. wide, acuminate, base narrowed to about 12

m.; ligule 4-5 mm. high, firm, edge shortly toothed. Leaves

commonly 20-40 cm. long and 3-6 cm. wide, lower surface vel-

vet-hairy (hairs to 1 mm. long), base broadly rounded in lower

leaves, cuneate in upper ones, petiole to 10 mm. long, widening

upwards; sheaths glabrous or with some hairs and a waxy cover-

ing when young; auricles usually with a few bristles 10-15 mm.

long; ligule 1-2 mm. high, slightly toothed. Inflorescences usually

on leafless branches; internodes of branches glabrous, to 8 cm.

long on larger ones; spikelet-groups densely crowded, the larger

ones 2:5 cm. diameter. Spikelets slender, acute, green, glabrous,

1- or 2-flowered; empty glumes usually 2, the upper nearly as long

as the lemma and shortly apiculate; lemma 7-5 mm. long including

a pale tip of less than 1 mm.; upper or sole palea a little shorter,

with 4—7 veins unevenly spaced, not keeled; palea of lower floret,

if present, keeled but not fringed; anthers hardly 4 mm. long, tips

not hairy, filaments free or sometimes apparently forming a tube;

ovary nearly spherical, top sparsely hairy, stigma undivided; fruit

with truncate slightly hairy top.

DISTRIBUTION: Johore, Taiping, Kedah and Kelantan, in open places

in forest.

RECORDED MALAY NAME: Buloh kapor.

The type of this species was found at Genuang in northern

Johore. It is not at Singapore, and I have not found it at Kew. I

think however there can be no doubt it was of the same species as

numerous plants which flowered abundantly ‘between Labis and

Segamat (very near the type locality) in October 1946; the above

description was prepared from these plants. No other Dendro-

calamus has been found in northern Johore. I propose therefore

to regard Henderson’s collection (S.F.N. 38201) as a neotype. I

made notes on culms and culm-sheaths from the plants in situ.

These plants were growing in forest which had been cleared dur-

ing the Japanese occupation of Malaya. The cutting: and exposure

doubtless stimulated them to flower.

I subsequently found a few plants in Kedah, mostly in bamboo

thickets in the north of that State, which resembled the Johore

plants in essentials, but differed in details of size of auricles and

Vol. XVI. (1958).

ligule on the culm-sheaths. One plant had auricles 13 mm. high

(normal height in this species is 3-4 mm.) with very short bristles

(5-7 mm. instead of 20-25 mm.), and ligule to 13 mm. high;

another had normal auricles but the ends of the ligule 15—20 mm.

high, deeply incised. Those plants from Kedah which I found in

flower agreed with the Johore plants. A flowerless plant from

Taiping Hills also agreed in essentials with the Johore plants, but

had short bristles on the auricles and a ligule only 4-5 mm. tall;

the sheaths may not have been representative of the full develop-

ment on that plant. Other plants which might pass vegetatively for

D. hirtellus are here included under D. sinuatus, because of the

larger number of flowers in the spikelets, and hairiness of lemma

and palea. Vegetative characters have been insufficiently studied in

D. sinuatus. If such study confirms that there is no clear distinc-

tion between the species, then they should probably be united,

under the earlier name sinuatus.

Specimens: JOHORE, Ulu Kahang ,Lake & Kelsall s.n. 1892 (S). G.

Panti, 2,000 ft., Ridley s.n. Dec. 1892 (S.) G. Pulai, Holttum S.F.N.

38437 (S); Ridley s.n. Dec. 1905 (S). Near Labis, Henderson S.F.N.

38201 (Neotype, S,K, Kep.); Holttum S.F.N. 37790 (S), S.F.N.

38308 (S). KELANTAN, Near Kampong Kando, Symington F.D. 37815

(S). Kepau, Sungei Patani, Wolfe & Abdul Kadir S.F.N. 21466, 21469

(S). Sintok Road, Holttum s.n. 26.10.1946. Bukit Jamboi, 700 ft.,

Holttum S.F.N. 19816 (K,S). PERAK, Maxwell’s Hill, Wyatt-Smith

F.D. 76179 (Kep.).

3. D. elegans (Ridl.) Holttum in Gard. Bull. Singap. 11: 296.

1947.

Basonym: Schizostachyum elegans Ridl. in Journ. Str. Br. R.

As. Soc. 73: 146. 1916. Flora 5: 271.

Culms to about 6 m. long and 2-5 cm. diameter, smooth, walls

thick, upper part drooping, internodes to at least 26 cm. long.

Culm-sheaths not known. Leaf-blades to 12 by 1-2 cm., base

rounded or broadly cuneate, stalk very short (05-15 mm.),

lower surface densely and shortly velvet-hairy; auricles small,

sometimes with slender bristles; ligule short, edge nearly smooth.

Inflorescences on tufted branches 10—20 cm. long at the nodes of

the culms, flowering and leafy branches sometimes from the same

node, the spikelets in dense groups 5—20 mm. apart. Spikelets about

8 mm. long and 1-5 mm. wide near the base, terete, glabrous; perfect

florets 1 or 2; lemma of upper or sole floret 7 mm. long, of lower

floret (if present) shorter, short-pointed; palea of lower floret

2-keeled, short-hairy on the keels, 2 veins between the keels; palea

of upper or sole floret not keeled, glabrous, about as long as

lemma, with 4 veins; anthers 6, a little more than 2:5 mm. long;

Gardens Bulletin, S.

ovary gradually contracted at the apex, smooth, style very slender,

smooth, ending in a single plumose stigma.

DISTRIBUTION: described from a flowering specimen from a plant

cultivated in the Waterfall Garden, Penang, origin Langkawi (Burkill,

S.F.N. 785, S,K.); also collected in flower on limestone on Pulau

Timun, Langkawi, by Henderson (S.F.N. 29097, S,K).

This species appears to be very near D. dumosus vegetatively,

but differs in its softly hairy leaves, which appear also to be

smaller and with even shorter stalks. In flower-structure, D. elegans

appears always to have 6 stamens, with somewhat shorter anthers

than in D. dumosus, and the apex of the ovary is quite glabrous.

It also often has two florets, with keeled palea hairy on the keels,

a feature not seen in D. dumosus; and the paleas of upper florets

examined by me had 4 veins only. Culm-sheaths of the two species

need to be compared.

4. D. dumosus (Ridl.) Holttum in Gard. Bull. Singap. 11: 296.

1947.

Basonym: Schizostachyum dumosum Ridl. in Journ. Str. Br.

R. As. Soc. 61: 64: 1912. Flora 5: 271.

Culms 2-2-5 cm. diameter, thick-walled, Culm-sheaths thin, to

18 cm. long, back rather sparsely covered with appressed dark

hairs, top of sheath little over 1 cm. wide; auricles small, bearing

bristles 5 mm. long; ligule 1-5 mm. tall, with toothed edge bearing

slender bristles near each end; blade 3—5 mm. wide, reflexed.

Leaf-blades commonly to 18 by 1-8 cm., sometimes to 23 by 3

cm., surfaces not hairy, base rounded to cuneate, stalks of upper

leaves to 25 mm. long (lowest leaves almost sessile); auricles

small, sometimes with a few fine bristles 5—8 mm. long; ligule

short. Spikelet-groups at the nodes (5-15 mm. apart) of smooth

leafless branches 7-15 cm. long from the nodes of the slender

upper part of the culm. Spikelets 1-flowered, with about 4 glumes

of increasing size below the lemma; lemma 7,mm. long, with a

short stiff point, almost completely enfolding the palea; palea 7

mm. long, thin, glabrous, not keeled, with about 7 veins; stamens

varying from 3 to 6, anthers 3-5 mm. long; ovary glabrous, with a

narrowly conical slightly hairy apex connecting it to the very

slender sparsely hairy style; fruit 5-5 mm. long, cylindric, glabrous

except for short hairs on the truncate apex.

DISTRIBUTION: Rawei Island, Langkawi, and Baling Hill (South

Kedah), on limestone.

The original collection was from Rawei Island, in the Adang

group, west of Terutau (Siamese territory), where Ridley reported

it “on a dry rock face of the island, forming dense thickets, the

Vol. XVI. (1958).

stems usually short, 6 or 7 feet, but sometimes much longer”.

Flowering plants agreeing with the type were later found on Baling

Hill; the culm-sheaths described above were from a flowerless

Baling plant, so there is a little doubt whether they belong here.

A flowerless specimen from Langkawi which agrees vegetatively

with the Baling plants has also been here included.

Ridley stated that his specimens had three stamens with connate

filaments; but of two spikelets examined by me one had five, the

other six stamens, and the filaments (still not expanded) appeared

to be free. The Baling specimens also appear to have free fila-

ments. For differences from D. elegans, see that species.

Specimens: RAWEI Island, Ridley 15903 (S,K, Type collection).

KEDAH, Langkawi, Batu Ayam, Corner S.F.N. 37839 (S). Baling Hill,

700-1, 000 ft., Nauen S.F.N. 38017 (S); half-way up, Corner & Nauen

S.n. 25.11.1941 (S); summit, Corner & Nauen s.n. 25.11.1941 (flower-

ing, S); summit, Kiah S.F.N. 35409 (S,K).

5. D. sinuatus (Gamble) Holttum in Gard. Bull. Singap. 11: 296.

1947.

Basonym: Oxytenanthera sinuata Gamble in Ann. R. Bot. Gard.

Calc. 7: 71, pl. 62. 1896. Ridley, Flora 5: 264.

Culms apparently waxy near tops of internodes when young.

Culm-sheaths rigid, when young covered with rather fine loose

brown hairs, smooth when old; blade relatively narrow, slightly

constricted near base and decurrent on each side as a narrow band

close to the top of the sheath, ending in firm auricles 2-4 mm.

high bearing stiff bristles; ligule stiff, toothed, the teeth perhaps

prolonged into bristles, undivided part to 5 mm. high. Leaf-blades

commonly 20-30 cm. long and 3-4-5 cm. wide, base broadly

rounded and then decurrent on a broad often hairy stalk 5-10

mm. long, lower surface hairy throughout; auricles sometimes with

slender bristles; ligule short, edge irregularly toothed. Spikelet-

bearing branches long, slender, pendulous, internodes short-hairy;

spikelets to 13 mm. long, slightly flattened, 2.5 mm. wide, nar-

rowed evenly to pointed apex, glumes and lemmas stiffly pointed

(longest points on upper lemmas), surfaces with very short fine

erect hairs or not; empty glumes usually 3 or 4; florets 3 or 2,

all perfect; lemmas to 11 mm. long; paleas to 10 mm. long,

narrow, the lower ones with hairy keels, 2—4 veins between keels

and one between each keel and edge, uppermost palea keeled to-

wards apex only, with a few short hairs on the keels; anthers 5—6

mm. long, apiculate, tip toothed or short-hairy, filaments before

extension adherent, the tube when extended very delicate; top of

Gardens Bulletin, S.

ovary and base of style rather long-hairy; fruit glabrous below the

truncate hairy apex.

DISTRIBUTION: Malaya, Annam; in Malaya known from three flow-

ering collections (as listed below) from Negri Sembilan and Pahang;

two sterile specimens, from Grik and Kemaman, may also belong here.

RECORDED MALAY NAME: Buloh akar.

As noted in the generic discussion, this species is very closely

related to D. pendulus and D. hirtellus, agreeing with the latter in

its rather large hairy leaves and probably also in its white-waxy

young culms. All the flowering specimens agree in the larger

number of florets, and also in hairiness of lemmas and paleas and

longer anthers (but in Dendrocalamus the upper florets are always

longer than the lower, so longer anthers may merely be correlated

with a larger number of florets). Careful notes on culm-sheaths

and other vegetative characters of plants identified from their

flowers are needed. I have never collected specimens of this spe-

cies, and the above description of the culm-sheath is made from

rather inadequate dried specimens.

Specimens: NEGRI SEMBILAN, Seremban, Alvins 1988 (S, Type).

PAHANG, Raub, Burkill & Haniff S.F.N. 16895a (S). Sungei Chelia,

Chigar Perah, Henderson S.F.N. 19373 (S). TRENGGANU, Kemaman,

Sungei Nipa, Corner S.F.N. 30545 (S). PERAK, Grik, F. R. Hamid,

F.D. 8258 (S,K).

6. D. strictus (Roxb.) Nees in Linnaea 9: 476. 1834. Gamble in

Ann. R. Bot. Gard. Calc. 7: 78, pl. 68, 69. Backer, Handb. FI.

Jav. 2: 280.

Basonym: Bambos stricta Roxb., Corom. Pl. 1: 58, t. 80. 1798.

Culms growing very close together, 10-15 m. long, much curved

(not stiffly erect), 2:5—7:5 cm. diameter, thick-walled; internodes

pale blue-green when young, dull green when old, 30-45 cm.

long, nodes somewhat swollen, some of the lower nodes often

bearing branches. Culm-sheaths with rather sparse golden brown

to dark hairs or almost hairless, rounded at the top; auricles small

and short-hairy, not bristly; blade erect, triangular, with stiff nar-

row tip; ligule 2-3 mm. high, edge toothed, not bristly (fig. 24).

Leaves to 25 cm. long and 3 cm. wide, lower surface short-hairy;

auricles sometimes with a few slender bristles; ligule low. Spike-

lets borne in dense groups up to 2°5 cm. diameter on branching

leafless axes; perfect florets 2 or 3: lemmas about 8 mm. long,

densely and minutely hairy towards apex, ending in a sharp spine

2 mm. long; paleas of lower florets 2-keeled, ciliate on the keels

and hairy on surface near tip, uppermost palea not keeled, often

nearly glabrous; anthers 3 mm. long, filaments free; ovary hairy

at top, style long, bearing a single stigma; fruit ovoid, 7:5 mm.

long.

ou ;

‘ht,

’

' '

Kf:

‘y

‘

ue) i"

“fa . =

Fig. 24. Culm-sheath of Dendrocalamus strictus. A, outer surface of com-

plete flattened sheath, x 1/3. B, inner surface of same. C,

inner surface of top of sheath, showing part of ligule and base

of blade, x 3. D, outer surface of part of top of sheath and

base of blade (auricle hardly developed).

Gardens Bulletin, S.

DISTRIBUTION: throughout India; perhaps occasionally planted in

Malaya, but there are no certain records apart from plants in the Public

Gardens at Kuala Lumpur (several large clumps) and the Botanic

Gardens, Singapore. A clump in the Residency garden, Penang, flow-

ered heavily in 1938; whether it died after flowering is not certain.

(Abdul Kadir, S.F.N. 36176, S,K, Kep.). The plants at Kuala Lumpur

and Singapore sometimes flower a little on leafy branches, and no

general flowering has been noted, though the clumps at Kuala Lumpur

must be much more than 30 years old.

In the Botanic Gardens, Singapore, are two plants believed to

represent different forms of this species, agreeing in their rather —

slender thick-walled culms, glaucous when young, branched at all

nodes nearly to the base, and sheaths of the shape above described.

They differ in their mode of branching, one with horizontal bran-

ches, the other with the main branch from each node strongly de-

flexed; the former plant has copiously hairy sheaths, the latter

almest glabrous ones. One of the Singapore plants is growing in

quite wet ground, and still has very thick-walled culms, though

Gamble says that D. strictus has thinner-walled culms in wet

places.

Gamble remarks as follows of this species: “this is the most

common and most widely spread and most universally used of all

the Indian bamboos, and is commonly known as the male bamboo.

Its culms are employed for all purposes of building and furniture,

for mats, baskets, sticks and other purposes”. Further planting of

D. strictus in Malaya would be worth trying.

7. D. asper (Schult.) Backer ex Heyne, Nutt. Pl. Ned. Ind. ed. 2,

1: 301. 1927. Handb. Fl. Jav. 2: 279.

Basonym: Bambusa aspera Schult., Syst. Nat. 7: 1352. 1830.

Kurz in Journ. As. Soc. Beng. 39, pt. 2: 87. 1870.

Synonyms: Dendrocalamus flagellifer Munro in Trans. Linn.

Soc. 26: 150. 1866. Gamble in Ann. R. Bot.

Gard. Calc. 7: 91, pl. 80. 1896. Ridl., Flora 5:

265.

Gigantochloa aspera Kurz in Ind. Forester 1: 221.

1876. McClure in Fieldiana, Botany, 24, pt. I:

141.1955,

Culms to 20 m. tall, when young (and persistently near base)

covered with fine closely appressed brown hairs, later green, near

the base commonly 12 cm. diameter and often more, lower nodes

usually bearing many roots; middle internodes 40-50 cm. long.

Culm-sheaths on upper part of culm 30—40 cm. long, very pale

green when young, sparsely covered with loose pale hairs, hairs at

base of sheaths darker, lowest sheaths rather densely covered with

100

Vol. XVI. (1958).

dark brown hairs; blade reflexed, narrow (on a lower sheath 8 by

1:7 cm., on an upper sheath 25 by 3-5 cm.), on lower sheaths

hardly narrowed at base, base crisped and extending laterally to

slightly crisped auricles up to 2 cm. in lateral extent and 7 mm.

high, bearing rather slender bristles about 5 mm. long; ligule 7-10

mm. high and beyond this a fringe of fine hairs 3 mm. long (fig.

25). Leaf-blades commonly to 30 cm. long and 2:5 cm. wide, base

of lower ones unequally cuneate above a stalk 3-5 mm. long,

lower surface sometimes rather sparsely hairy near the base, often

quite glabrous; auricles small, not bristly; ligule about 2 mm. high,

entire or with teeth or bristles. Spikelets usually on long leafless

short-hairy branches, in small nearly spherical and rather dense

groups, each spikelet 6-9 mm. long, flattened, 4 mm. wide, apex

blunt; empty glumes 1 or 2; florets 4 or 5, also often a reduced

sterile apical floret; lemmas broad, fringed with pale hairs towards

apex, back covered with short fine hairs, upper lemmas longer

than lower, uppermost to 8 mm. long; paleas about as long as

lemmas, keeled, fringed on keels and edges, outer surface fine-

hairy towards apex (edges of palea of upper-most floret fringed

but Keels not fringed), veins between keels 1—3, between keels

and edges 1 or 2 (fig. 23 D, E); anthers 3—5 mm. long (longer

in upper florets) with short glabrous tip; ovary and style hairy,

style undivided.

DISTRIBUTION: much planted throughout Malaysia for its edible

young shoots; native country not certainly known, but possibly north-

ern Malaya and neighbouring regions (an apparently wild plant found

near Cameron Highlands).

RECORDED MALAY NAMES: Buloh Béting, Buloh Bétong; in Singapore

known as Rébong China (cultivated by Chinese).

Schultes’ name B. aspera and his description are based entirely

on Rumphius, who described this species in Amboyna. He stated

that it was planted from cuttings, but also thought it to be native.

Backer states that he doubts whether it is wild in Java, as no seeds

have ever been seen. I found a plant (not flowering) which

agreed in vegetative characters with this species in the forest at

3,000 ft. near the road to Cameron Highlands (S.F.N. 38416).

This plant could have been planted; but as it was in apparently

primitive forest it could also be native. Further evidence is needed.

I think Rumphius’s description is sufficient to indicate the present

species, as no other is so universally cultivated for edible purposes.

Gamble states that he has found one or two lodicules in some

spikelets, but I have never seen them in the many spikelets I have

examined. Gamble also figures three stigmas, a condition I have

not seen. The spikelets on a single inflorescence may vary in the

presence or absence of an imperfect terminal floret, which is borne

101

\ Se “a es a)

la ay oR BR 4s

TATTTN\\\N

Fig. 25. Culm-sheath of Dendrocalamus asper. A, outer surface of com-

plete flattened sheath, with inner surface of blade and top of

sheath superposed. B, top of sheath and blade in natural posi- — |

tion, x 4. C, outer view of part of top of sheath, showing

auricle overtopped by fringed ligule, x 3. D, inner view of end

of ligule, with part of base of blade.

Vol. XVI. (1958).

on a continuation of the rachilla about 1-5 mm. long. The imper-

fect floret varies in character; in one case I found in it 3 stamens

and a rudimentary ovary. McClure, when placing this species in

the genus Gigantochloa, did not explain his reasons for so doing.

The persistent brown-hairy covering of the bases of the culms,

and the many roots on the lower nodes, appear to be distinctive.

The pale green sparsely hairy sheaths on the upper parts of un-

branched young culms are also very striking. The only other

bamboo I have found which could be confused with this one is

Gigantochloa levis, under which are notes on distinguishing

characters.

In addition to the edible shoots of this species, the old culms

are also useful, being large and very strong.

Specimens: SINGAPORE, Experimental Nursery, Ridley 454 (S). Lower

Garden, Ridley 3945 (S,K), 10644 (S). Deer Shed, Botanic Gardens,

Ridley 5603 (S,K). Ang Mo Kio, Ridley 6681 (S,K). No locality,

Ridley 2929 (K); Cantley 2939 (S). Burkinshaw’s Jungle, Ridley s.n.

27.5.1893 (S). Tanglin Hill, Burkill s.n. 1.5.1913 (S). Jurong Road,

G. F. Hose 88 (S); Hullett 486 (S.K). Adam Road, Kiah s.n.

23.12.1932 (S,K). Dunearn Road, Pestana s.n. 15.10.1932 (S). Bukit

Brown, Pestana s.n. Oct. 1933 (S). MaLacca, Alvins 115 and s.n. (S).

Gaudichaud, s.n. (K). Griffith, s.n. (K). Bukit Sabukor, Derry 101

(S). Bukit Bruang, Alvins s.n. 20.1.1886 (S). PERAK, Taiping, Strouts

F.D. 9987 (Kep.). Cameron Highlands Road, 3rd mile, Holttum S.F.N.

38425 (S); 3,000 ft., Holttum S.F.N. 38416 (S). PENANG, Batu Ferin-

ghi, Ridley & Curtis 8363 (S,K). Balik Pulau, Ridley 9460 (S). Batu

Feringhi Road, Birch s.n. 27.3.1900 (S); Curtis 3565 (S,K), 3566

(S,K). No locality, Flippance s.n. 28.6.1934 (S).

8. D. giganteus Munro in Trans. Linn. Soc. 26: 150. 1868.

Gamble in Ann. R. Bot. Gard. Calc. 7: 87, pl. 76. Ridl., Flora

5: 265. Backer, Handb. Fl. Jav. 2: 281.

Culms to 30 m. tall, 18-25 cm. diameter near base, walls to

2:5 cm. thick, when young covered with a white waxy layer.

Culm-sheaths 25-50 cm. long, dark-brown-hairy on the back;

blade spreading at right angles, narrowly triangular on sheaths 6

ft. above ground, stiff, edges inflexed towards stiff acuminate apex,

base widened into large brown strongly crisped auricles which are

not bristly; ligule 8-12 mm. high, stiff, dark, bearing a short

fringe. Leaf-blades commonly 40 cm. long and 6 cm. wide, lower

surface slightly rough with distinct cross-veins, petiole to 5 mm.

long; auricles small and glabrous, ligule to 3 mm. high, irregularly

toothed. Spikelets in groups of about ten at the nodes of finely

hairy branches; detached mature spikelets 13-17 mm. long, flat-

tened, 4-5 mm. wide; empty glume one (fig. 26); florets 4—6, the

uppermost one sometimes imperfect in varying degrees down to a

small rudiment; lemmas broad, 8-13 mm. long with a short firm

tip, edges very shortly fringed, back bearing short fine hairs;

103

Gardens Bulletin, S.

paleas with long-fringed keels, rather narrow infiexed finely

fringed edges, short hairs all over back, 2—3 veins between the

keels (uppermost lemma only slightly keeled); anthers 7-10 mm.

long, tips pointed; fruit 7-8 mm. long with blunt hairy apex.

MW ittrrhs The

yet .

ttt OU as

rt mnearveevn

ye hots pene

A B Cc D E

Fig. 26. Dendrocalamus giganteus. A, complete spikelet, x 2. B—G, paleas

of its six perfect florets, in order (lowest on left), x 3. H,

imperfect terminal floret, x 3.

DISTRIBUTION: Lower Burma; reported by Ridley as native in Malaya

but evidence for this cannot be found.

The above description was prepared from a fine clump planted

at the Forest Research Institute, Kepong, Selangor, probably in-

troduced by the late Dr. F. W. Foxworthy; the clump flowered in

the years 1951-1953 (Wyatt-Smith, F.D. 55798, Kep; E. F.

Allen s.n., S). I have also seen plants (not flowering) in the

Botanic Garden at Bogor in Java. This most spectacular bamboo

should be planted for ornamental purposes in Malaya, and it

might well find practical uses also. There are specimens taken

from a plant formerly in Government House Domain, Singapore

(Anderson s.n., February 1914; S,K); this plant disappeared be-

fore 1922.

GIGANTOCHLOA

Gigantochloa Kurz ex Munro, Trans. Linn. Soc. 26: 123. 1868. —

Type species: G. atter Kurz ex Munro, l.c. 125, from Java; see —

also Kurz in Indian Forester 1: 344. 1876. Holttum in Taxon 5:

28-30. 1956.

Culms close and erect, mostly of medium size, forming dense

clumps, the upper parts curved outwards but never with very

slender pendulous tips, lower part of culm always unbranched,

upper part with small branch-groups at the nodes; walls of

erate thickness; young culms often bearing scattered coarse

104

Vol. XVI. (1958).

on their exposed parts, in G. apus a close covering of pale hairs.

Culm-sheaths usually dark-hairy, often with loose hairs; auricles

in nearly all cases low (often elongate laterally) and very firm,

dark green when young, with or without marginal bristles; blades

of middle sheaths usually rather narrow, sometimes green and leatf-

like; ligule well developed, usually thin and more or less lacerate,

or with marginal teeth. Spikelet-groups usually on leafless bran-

ches (often a whole culm bears flowers and no leaves); spikelets

consisting of several basal bracts and glumes, 2—5 perfect florets

and a terminal narrow empty lemma, the rachilla-internodes all

very short (under 0-5 mm.), the lower lemmas shorter than the

upper ones, the terminal empty lemma longest of all (fig. 27);

lemmas usually fringed, apiculate, many-veined; paleas thin and

translucent, all strongly keeled, narrow between the keels and with

narrow inflexed edges, keels usually fringed; stamens 6, the fila-

ments joined to form a firm tube which at flowering is a little

longer than the palea, anthers long, with a distinct tip; ledicules

usually absent; ovary with hairy top; style long, usually ending in

a single hairy stigma; fruit narrowly cylindric, hairy at the top

only, pericarp thin towards the base and showing position of

small round embryo, slightly grooved down the other side along

the line of the hilum.

E F G H

Fig. 27. Gigantochloa wrayi. A, complete spikelet with stamens and stig-

mas protruding from two florets, x 24. B, palea of lowest

floret. < 3. C, floral parts from B (ovary hidden by stamen-

tube) before anthesis. D, palea of floret next above B. E, floral

parts from D. F, ovary and style from D. G, palea of upper-

most perfect floret. H, empty lemma which represents terminal

floret.

I have selected G. atter as the type species, because Munro's

first species, G. verticillata, is a mixture; see discussion under G.

maxima, below.

Gardens Bulletin, S.

This genus appears to be confined to the region of Burma, Siam

and Indochina southwards to Malaya. Several species (or clones?)

are planted in Java, and one in North Borneo and the Philippines,

but these are not native and were probably brought from the region

of Lower Burma. In Malaya I have seen no wild plants south of

Tampin (i.e., the southern end of the Main Range); as one goes

northwards the variety of Gigantochloas increases, and in the

north of Kedah and Perlis they are very abundant in the bamboo

thickets near Sintok and elsewhere. My impression is that hybrid

swarms of Gigantochloa occur in the north of Malaya; the variety

of plants in the Sintok region is quite bewildering. Besides native

plants, which are locally very abundant on edges of forest and in

clearings, several Gigantochloas are common village plants in all

parts of Malaya; some of them have probably been propagated

from wild plants in the neighbourhood, some (like the Gigantoch-

loas of Java and the Philippines) brought from more remote areas.

The planted Gigantochloas are undoubtedly indications of the

migrations of men, and as such are of considerable potential

interest.

But the species of Gigantochloa are difficult to characterize

clearly; and there is no doubt that their area of greatest abundance

is in Lower Burma and Peninsular Siam, where they have been

little studied in the field. Malay Peninsula species in fact cannot

be properly understood until those of the region further north

have been carefully studied. And even those species which exist in

Kedah need much more field study; the present treatment is no

more than a tentative one.

The form of spikelet is very constant throughout the genus;

vegetative form is more variable, and in fact bamboos with quite

different vegetative characters may bear almost identical spikelets.

Backer’s solution has been to include in one species all plants with

the same kind of spikelet; this is perhaps inevitable, but it does

not help the man who wants to distinguish bamboos by vegetative

means for practical uses. The distinct vegetative forms, whatever

taxonomic rank they are given, need to be described. The various

forms included by Backer under G. verticillata are apparently

well known by name to village people in Java (though I do not

find that they have yet been clearly distinguished in botanical des-

criptions), and are in fact clones, propagated vegetatively. They

may be particular forms of variable natural species or of hybrid

swarms; and it may even be that new forms have arisen by the

germination of chance seeds in Java (though fruits are not abun-

dant, they are certainly produced occasionally). The largest form

of G. verticillata (in Backer’s sense), said to be widely planted in

106

Vol. XVI. (1958).

Java, is not generally planted in Malaya (I have only seen a

colour-variant of it, at Kota Tinggi in Johore), nor have I seen

any indication that it is wild in Malaya; its origin must be sought

further north. The same is true of Gigantochloa apus, said to be

the most useful species in Java and one of the most distinct vege-

tatively of all Gigantochloas; the only plants I have seen in

Malaya are at the Botanic Gardens, Singapore, and at Serdang,

brought in both cases from Java in recent years. Gigantochloas

will never be properly understood until a plantation, made from

many plants from many localities, is established and studied over

a period of years; and most probably also it will be necessary to

plant seedlings, to discover the extent of a heterozygous condition

in the genus. That however is counsel of perfection. The practical

thing needing to be done first is to select some clones which are

good for village planting, for particular purposes, and to propagate

them and test them. This genus provides more useful bamboos

than any other in Malaya.

The genus Gigantochloa has traditionally been separated from

Dendrocalamus by the supposed condition of the fruit. Munro pro-

posed a new sub-tribe Bacciferae (subdivided and re-named by

Bentham, but essentially unchanged) to include bamboos with a

fleshy pericarp in which the position of the embryo is not exter-

nally visible; he placed Dendrocalamus in this sub-tribe, and

Gigantochloa (with Bambusa) in a sub-tribe of genera having a

thin pericarp through which the position of the embryo could be

seen. I have compared fruits of Dendrocalamus pendulus and

Gigantochloa wrayi, and I can see no difference whatever in essen-

tial structure. In both cases the pericarp is very thin towards the

base and the position of the embryo is clearly seen through it.

This is true also of Dendrocalamus hamiltoni and of other species

of which fruits are figured and described by Gamble. Having re-

moved this supposed difference, one has to place the two genera

near together, and to find another basic distinction between them.

That this is not easy is indicated by the fact that Munro and

Gamble placed in Oxytenanthera a group of Asiatic species some

of which are Dendrocalamus and some are Gigantochloa, as

defined in the present work (true Oxytenanthera is confined to

Africa, and is characterized by a quite different kind of ovary).

The presence of a stamen-tube is a possible distinction. But in

fact it is not at all easy to observe, and as it occurs also in the

very different genus Schizostachyum (or in Neohouzeaua, if one

keeps that as a separate genus) I do not think it is necessarily an

important character. I find that the distinctive feature of the

Gigantochloa spikelet is the presence of the rudimentary terminal

107

Gardens Bulletin, S.

floret in the form of a long narrow empty lemma, longer than the

uppermost fertile lemma. Owing to the presence of this lemma,

the palea of the uppermost perfect floret, which backs on to it, is

two-keeled like the rest. In Dendrocalamus, the last floret is usually

perfect, in which case the palea has no further structure behind it

and is generally not keeled; or if a rudimentary floret is present,

it is of variable form and the uppermost palea is slightly keeled.

In some species of Dendrocalamus, as so separated, there is a

delicate stamen-tube, much less easy to see than that of Gigan-

tochloa. The rather firm stamen-tube in Gigantochloa is thus, |

think, a good character.

In the fruit of G. wrayi and G. ligulata, the apical part of the

pericarp is much the thickest part, and is separated by a distinct

gap from the top of the seed. Towards the base of the fruit the

pericarp is quite thin, in close contact with the seed, but in the

almost ripe fruit it can be separated easily from the seed (the

separation is not easy in a dried herbarium specimen). The peri-

carp is thin enough to show the position of the small round basal

embryo quite clearly. Apart from the embryo, the other external

features of the fruit are three very slight and narrow ridges

(grooves in the dry fruit), representing the course of the median

vascular bundles of the three carpels, and a broad darker band

(slightly depressed in the dry fruit) on the side opposite the

embryo, representing the position of the vascular tissue which

supplies the developing seed all along the linear hilum (see sec-

tion in Arber, The Gramineae, p. 122, fig. B1).

The fruit of Bambusa also is essentially similar in structure to

that of Gigantochloa and Dendrocalamus, but Bambusa differs in

spikelet-form, having elongated rachilla-internodes and lemmas all

about the same length; it has also paleas of firmer texture (the

paleas of Bambusa are partly exposed, but quite hidden in Gigan-

tochloa) and often a short style with three stigmas. One species

which in these characters belongs to Bambusa has been placed in

Gigantochloa because of the presence of a stamen-tube (G. hete-

rostachya Munro); as with the species of Dendrocalamus having

this character, I consider that the weight of the other evidence is

more important in deciding the relationships of this species, and

include it in Bambusa.

Field characters. Most Gigantochloas are moderate-sized bam-

boos with very straight culms growing close together, the tops of

the culms never very thin, pendulous and whip-like, as in Schizo-

stachyum. The culm-sheaths are usually not so thick as those of

Dendrocalamus and are pale papery-brown when old. The auricles

are nearly always low and very firm, dark green when young; the

108

Vol. XVI. (1958).

exception is the apparently little-planted G. levis, which has raised

auricles more like those of the larger species of Dendrocalamus.

The ligule is often tall and very thin, deeply lacerate in part, and

thus of fragile nature not easily preserved intact in herbarium spe-

cimens. The blade is either erect or deflexed; blades from the

middle of a culm upwards are often green and leaf-like. The culm

walls are of varied thickness, never very thin, and they do not

split so readily as those of Schizostachyum. In Schizostachyum the

walls are not only thin; the fibres are mainly concentrated in a

dense thin layer near the surface, the fibres of the inner vascular

bundles less well developed. In Gigantochloa, as in Bambusa,

there is a much more gradual lessening of fibrous structures in

the inner parts of the wail of the culm; the bundles are more

widely spaced as one passes inwards from the very dense layer

near the surface, but each bundle has equally large and strong

fibre-groups. The result is that the wall of the culm in Gigantochloa

and Bambusa is of much more even structure than in Schizosta-

chyum, and this allows their use in particular ways, as mentioned

below.

Spikelets of Gigantochloa are usually on all branches of a single

culm. They are of moderate size, and the fringed lemmas are often

distinctive. Sometimes the end of the stamen-tube can be seen

protruding beyond the tip of a lemma at the flowering stage.

In the region of Selangor and Lower Perak young culms of ihe

abundant species G. scortechinii are conspicuous by the contrast-

ing colours of their sheaths, bright orange near the top and covered

elsewhere by glistening brown-black hairs, the exposed parts of

the culms themselves often pale glaucous green. Further north

other species appear, especially G. ligulata in its varied forms. In

open country in Kedah, typical small G. ligulata is the common

bamboo, with its large leaves and very long leaf-ligules.

Uses. The larger Gigantochloas have strong culms useful for

general structural purposes, agreeing in quality in this matter with

Bambusa arundinacea. Hildebrand states that the culms of G.

' apus, though somewhat smaller than those of G. maxima, are

more durable. Neither of these is common in Malaya, but prob-

ably the large forms of G. ligulata, and G. scortechinii, are of

comparable quality. Gigantochloas with walls not too thick are

used for making floors and walls of houses, by a process of split-

ting and flattening. The medium-sized Gigantochloas, with walls

not too thick, are probably the bamboos most in use for basket-

making, and here the structural character of the culm-wall, des-

cribed above, is important. The culms are first split along radial

109

Gardens Bulletin, S.

planes to strips of suitable size, and then these are split tangen-

tially (i.e., in a plane parallel to the surface). The rather uniform

texture of the culm-walls makes this latter form of splitting

possible, and provides thin flexible strips of varying width which

can be used for weaving the sides of baskets.

Some Gigantochloas give good edible shoots (these are said to

be manis by Malays); the shoots of others are astringent (kélat)

or bitter (pahit). My impression is that clones of very similar

vegetative form may differ in edible quality. Malays near the road

to Cameron Highlands pointed out to me two Gigantochloas, one

good to eat, one bitter; they differed a little in hair-colour on the

sheaths and a little in the shape of the auricles, and I could see no

other clear distinction. There has been no systematic selection of

the best edible varieties. The one Gigantochloa which is certainly

of excellent quality for this purpose is G. levis, but this does not

seem to be native, nor commonly planted, in Malaya. The propa-

gation of this species would seem to be worth while. I was told in

Kedah that the common small form of G. ligulata gave good edi-

ble shoots, though of course small ones.

Key to the Malayan species of Gigantochloa

Biade of middle culm-sheaths spreading or

reflexed, not erect

Culm-sheath ligule not over 5 mm. tall

including fringe (if any)

Culms when young covered with white

appressed hairs 4% .. Lo Gira

Culms when young not so covered

Big bamboo, culms 10 cm. or more

diameter ;

Culms streaked light green and

yellowish ss .. 2. Gy. maxima

(typical). .

Culm uniform dark green .. GG. maxima var.

viridis

Smaller, culms to about 5 cm. dia-

meter

Culm-sheath auricles 2 cm. or

more in lateral extent, not

over 2 mm. high anywhere .. G. maxima vat.

minor

110 pete

Vol. XVI. (1958).

Culm-sheath auricles usually

shorter in lateral extent and

more or less raised at outer

ends :

Culm-sheath ligule more fee 5 mm. tall

including fringe ;

Culm-sheath auricles commonly 7 mm.

high

Culm-sheath auricles yes lower

Culm-sheath ligule with its fringe at

least 10 mm. tall; spikelets

densely hairy all over

Hairs on culm-sheaths very dark

brown

Hairs on culm-sheaths nearly white

Culm-sheath ligule with its fringe not

over 10 mm. tall; spikelets not

densely hairy all over

Culm-sheath auricle without bris-

tles; spikelets 20-25 mm.

long, lemmas _ dark-fringed;

leaves glabrous beneath

Culm-sheath auricles often with

bristles at the outer end; spike-

lets to 20 mm. long, lemmas

brown-fringed; leaves more or

less hairy beneath

Blade of middle culm-sheaths erect

Culm-sheath ligule not over 4 mm. tall

Culm-sheath blade broadly triangular,

not narrowed at base

Culm-sheath blade narrowly hasitadben

and narrowed at base

Culm-sheath ligule 10-25 mm. tall

Culm-sheath ligule much taller at ends

than in middle; leaf-ligule com-

monly 15-20 mm. tall

Culm-sheath ligule not taller at an

than in middle; leaf-ligule less than

10 mm. long, usually much less ..

111

3. G. hasskarliana

4. G. levis

5. G. scortechinii

(typical)

G. scortechinii var.

albovestita

3. G. hasskarliana

6. G. wrayi

7. G. ridleyi

3. G. hasskarliana

8. G. ligulata

9. G. latifolia

Gardens Bulletin, S.

1. G. apus (Schult.) Kurz in Tijdschr. Ned. Ind. 27: 226. 1864.

Munro in Trans. Linn. Soc. 26: 126. 1868. Backer, Handb. FI.

Jay, 2: 21.

Basonym: Bambusa apus Schult., Syst. Veg. 7: 1353. 1830.

Synonym: (?): Gigantochloa kurzii Gamble in Ann. R. Bot.

Gard. Calc. 7: 65, pl. 56. 1896. Parker in Indian

Forester 57: 108. (Not G. kurzii sensu Ridl.,

Flora'5:-2615

Young culms covered partly or wholly with fine whitish ap-

pressed hairs, thus appearing grey-green, smooth and green when

old; culms 10—20 m. tall, up to 9 cm. diameter, basal part un-

branched and of even thickness, longest internodes 45 cm. or more

long. Middle culm-sheaths dark-hairy on the back, commonly 25

cm. long (to 45 cm., fide Backer), middle part of the top (bear-

ing the blade) distinctly rounded, the edge on either side of the

blade bordered by narrow firm auricles (part near the blade com-

monly under 2 mm. high, outer ends about 3 mm.) their edges

with scattered rather slender bristles; blade reflexed, deciduous

when old, relatively narrow and narrowed at the base, when young

dark-hairy on both surfaces (4:5 to 18 cm. long, 2—6 cm. wide,

fide Backer); ligule 3-5 mm. high, irregularly toothed, not long-

bristly (fig. 28). Leaf-blades very variable in size on the same

culm or even on the same branchlet, distal ones often much

smaller than basal, 9-40 cm. long, 1-5—7:5 cm. wide, paler be-

neath than above and very finely hairy beneath when young,

petiole commonly 7-12 mm. long; auricles firm, rounded, usually

quite glabrous; ligule 2-3 mm. tall, edge finely hairy. Spikelets

15-20 mm. long, 3-5—4-5 mm. wide near the base; bracts and

glumes, of increasing size, 4—5; perfect florets 3 or 4; lemmas of

perfect florets to 17 mm. long, shortly pointed, fringed near the

apex with dark brown hairs, the outer surface bearing short fine

appressed hairs near the apex; paleas with long hairs on the keels,

4—S veins between keels and 1-3 between each keel and edge;

anthers to 8 mm. long, with slender tips; fruit 10-12 mm. long.

DISTRIBUTION: probably native in Tenasserim, introduced long ago

to Java and widely planted there, little known in Malaya.

This species is well known in Java by the names Bambu (or

Pring) Apus and Bambu Tali. The above description (apart from

the floral characters, which are taken from Backer) was prepared

from a plant near the Javanese gardeners’ quarters at the Botanic

Gardens, Singapore, said to have been brought from Java (Md.

Nur & Pestana, several specimens, S). The only other plants I

have seen in Malaya are at the Federal Experimental Plantation

142

Vol. XVI. (1958).

‘pus J9jNO SIE Ww Wt DurddoysoAro spstuine oy) pur opnFy

JUIMOYS “ 0} Surpuodsess09 MIA JoUUL ‘GE *Z & ‘ape Jo oseq puy (papeys) 9poLINe JO MaIA sano “> ‘OURS Jo

eoRJANS JoUUT JO Awd Joddn “gq “fF % “YRaYys oJa[dUWIOD poud}eYy Jo aouzANS J9jNO “y ‘sndp DO[YIOIUPBID) Jo yeoys-wuND ‘gz ‘BLY

KD Wi .

Dy Lge MDT) wry, {

anit a WILLY, EA

HTS

\'\

\ %

113

Gardens Bulletin, S.

of the Department of Agriculture, at Serdang, Selangor (Holttum

s.n. 21.10.1946, S). It is quite likely that plants exist also else-

where in Malaya, but they are not common village plants in any

places I have visited. As G. apus is considered the most useful

species in Java, its planting should be encouraged in Malaya.

Schultes’s original description is lengthy, but does not mention

the vegetative characters which appear to be most distinctive. The

association of the name apus with the species here described is

mainly due to tradition in Java. Young culms of Gigantochloa

nearly always have scattered coarse hairs on the exposed parts of

the internodes, but I do not know another that has a close cover

of pale hairs on the young culms. This covering is quite different

from the amorphous white waxy powder often seen on young

culms of Dendrocalamus, and also on Gigantochloa scortechinii.

The type specimen of G. kurzii Gamble was gathered by Kurz

in Burma. So far as I can see, there is nothing in it to disagree

with G. apus as described by Backer, so Backer and Parker may

well be right in ranking G. kurzii as a synonym of G. apus. But

the Peninsula specimens quoted by Gamble and Ridley as G.

kurzii are probably different, and I am placing most of them in

G. wrayi.

Uses. Hildebrand says that this is the most generally useful

bamboo in West Java, its poles being especially durable. Heyne

states that well matured culms are first dried in the shade and

then soaked in water for at least a month, to prevent subsequent

attack by beetle larvae. Hildebrand states that small culms are

used as a substitute for rotans in making furniture. The name tali

indicates a use for rope or string. It seems to mean that the cuims

can be cut up into fine strips for weaving hats and baskets and

other objects; when split fine, and the pieces bent, the surface does

not chip off. Ochse reports that the young shoots are bitter and

not good to eat when freshly cut, but that after immersion in wet

mud for 3 or 4 days they are good to eat.

2. G. maxima Kurz in Indian Forester 1: 343. 1876.

Synonyms: G. verticillata sensu Gamble in Ann. R. Bot. Gard.

Cale. 77°61, (hpi

G. verticillata p.p. sensu Munro in Trans. Linn.

Soc. 26: 124; and sensu Backer, Handb. FI. Jav.

23210,

G. verticillata var. Awi Gombong, Ochse, Veg.

Dutch E. Ind. 323, 325-6.

114

Vol. XVI. (1958).

Culms 15 m. or more tall, to 10 cm. or more diameter, rather

light green streaked with many longitudinal bands of yellow-

green, not hairy. Culm-sheaths to 30 cm. long, hairs on back copi-

ous, very dark brown, top broad and not raised at attachment of

blade; auricles to 20 mm. in lateral extent, 2-3 mm. high near

outer ends, very firm, not bristly; blade deflexed, on a large middle

sheath 25 cm. long and 5 cm. wide, narrowed at the base; ligule

3 mm. high.

DISTRIBUTION: commonly planted in Java; introduced to the Federal

Experiment Station, Department of Agriculture, Serdang; the above

description made from the Serdang plants on 21 October, 1946 (spe-

cimen in H.S.).

Var. viridis Holttum, var. nov.

Culmi magni, virides (non luteo-striati), in juventute pilos cons-

persos fuscos ferentes; vaginae culmorum eis Awi Gombong simi-

les, differunt auriculis 5 cm. horizontaliter extensis.

Culms 10 cm. or more in diameter, uniformly green, upper

parts of young internodes bearing scattered dark hairs. Culm-

sheaths like those described for typical G. maxima, orange-

flushed towards apex when young; auricles extending 5 cm. along

top edges of sheath each side of the blade, 3 mm. high. Leaf-

blades commonly about 25 by 2-5 cm., glabrous. Spikelets 16-18

mm. long (when detached), 3—4 mm. wide near base; perfect

florets 3; lemmas fringed with dark purplish hairs; paleas with 3-4

veins between keels and 1—3 between keel and edge, keels fringed

and outer surface minutely hairy; anthers to 8 mm. long.

TYPE: Kota Tinggi, Johore, Holttum S.F.N. 40201 (S).

The above description was made from a large plant on the

stream bank near the Mawai Road (north of the road) about two

miles from Kota Tinggi. A number of similar large clumps occur

along the banks of the stream, but no other has flowered. The

area has long been inhabited, and the bamboos are doubtless

planted. I am indebted to Mr. J. A. le Doux for pointing out the

plants to me and for observing the flowers. I have seen no other

similar bamboos in Malaya except those introduced to Serdang

from Java, as above described; the latter differ chiefly in their

streaked culms. The Serdang plants were not flowering when |

saw them, and it is not certain that their spikelets are exactly like

those at Kota Tinggi. Backer, who takes a very broad view of the

species G. verticillata (see discussion below) gives the spikelet-

length as 9-13 mm., and anther-length 4-6 mm. Possibly these

spikelets were from one of the smaller varieties of his comprehen-

sive species.

115

Gardens Bulletin, S.

Var. minor Holttum, var. nov.

Culmi c. 5 cm. diametro, interdum striati, parietibus crassis.

A small bamboo; culms not over 5 cm. diameter, sometimes

streaked with light green, thick-walled; culm-sheaths like those of

G. maxima but proportionately smaller (fig. 29); flowers not

known.

Fig. 29. Gigantochloa maxima var. minor. Upper part of culm-sheath with

blade raised into the same plane; inner view on left, showing

ligule; outer view on right, showing the long low auricles.

Type: Forest Research Institute, Kepong, Selangor, Holttum s.n.

September 1953 (S).

This is very like a miniature plant of G. maxima. It may be the

same thing as Kurz’s Bambu Andong ketjeel. The clump is a

large and vigorous one, and I think had reached its full growth as

regards size of culms. I have not noticed this bamboo in Malay

villages; its origin is not recorded. It is planted near the big clump

of Dendrocalamus giganteus, and may be another introduction

made about the time the Institute was established.

As indicated above, this species has usually been called G. ver-

ticillata, a name based on Bambusa verticillata Willd. (Sp. Pl. 2:

245. 1797). Willdenow’s specimen, which he described very

briefly, was seen by Munro, who reported that it consisted of a

single flowering branchlet, sent by Thunberg from Java; apparently

this specimen bore the local name Tring ater, but Munro ranked

G. atter Kurz as a separate species. Munro also included several

other specimens under G. verticillata, a procedure objected to by

Kurz in 1876; Kurz divided Munro’s species into two, namely G.

maxima and G. robusta, the former with two varieties. He doubted

whether Willdenow’s specimen belonged to either species, and so

rejected the name verticillata. Backer has reverted to Munro’s

116

Vol. XVI. (1958).

treatment, including all (and G. atter Kurz) in one species; un-

fortunately he does not indicate how to recognize the different

varieties or forms so included. Gamble includes as a synonym only

G. maxima from among Kurz’s species, citing Ridley 119 from

Singapore (which is G. Jevis) and no other specimens from the

Peninsula. Gamble’s description is taken partly from Munro, and

his drawings were made from Kurz’s Java specimens, or copied

from drawings made by Kurz; he never saw living plants.

Munro’s description naturally omits reference to characters of

culms and culm-sheaths. He describes lodicules as present in the

florets (3 in the uppermost one, | or 2 in the others); but did he

copy this from the earlier description of Schultes? Gamble also

figures a lodicule as seen in a specimen of Kurz. Backer makes

no mention of lodicules in his description of the species, and in

his generic description says lodicules are lacking. I have never

seen a lodicule in any spikelet I would consider to belong to the

genus Gigantochloa. Regarding vegetative characters, Kurz and

Backer (the two who both saw living plants in Java) disagree as

to the blade of the culm-sheath; Kurz describes it as reflexed,

Backer as erect. Kurz’s drawing of a culm-sheath (copied both by

Gamble and by Koorders) shows low fringed auricles and a low

toothed ligule. Backer does not describe the auricles clearly (he

says they are small and soon fall off).

In the Botanic Gardens at Bogor I examined plants of most of

the various forms included by Backer in G. verticillata. Their

culm-sheaths and other vegetative characters differed so much that

I cannot agree to include all in one species; for example, Awi

Andong had large raised and rounded auricles with many curved

bristles and almost erect broad blades, and a plant named Awi

Leah (I think this should have been Awi Gombong) had long

low auricles without bristles and reflexed narrow blades, as in the

Serdang plants above described. Backer does not indicate this

amount of vegetative difference in his description.

It is at any rate certain that the name verticillata, as typified

by Willdenow’s flowering specimen, cannot be attached with cer-

tainty to any one of the forms included in G. verticillata by

Munro and Backer. I consider therefore that this name should be

ignored as a nomen dubium, and that a new type for the generic

name should be proposed. In view of the descriptions of Kurz and

Ochse, and of the traditional use in Java of the name Awi Gom-

bong, I do not doubt that the Serdang plants are G. maxima

Kurz; the only discrepancy is the fringe of hairs on the auricles

shown in Kurz’s drawing (as copied by Gamble) and his state-

ment “rigidly fringed at the auricles”. I offer no explanation of

117

Gardens Bulletin, S.

this discrepancy. Typical G. maxima is the largest of the Gigan-

tochloas, and is considered a very valuable bamboo for structural

purposes in Java. It would surely be worth planting more widely

in Malaya.

3. G. hasskarliana (Kurz) Backer ex Heyne, Nutt. Plant. Ned.

Ind., ed. 2, 1: 299. 1927. Backer, Handb. Fl. Jav. 2: 277.

Basonym: Schizostachyum ? hasskarlianum Kurz in Indian

Forester 1: 352. 1876.

Synonym: Oxytenanthera nigrociliata sensu Munro, p.p., m

Trans Linn. Soc. 26: 128. (Not Bambusa nigro-

ciliata Buse).

Culms to about 5 cm. diameter, 5—6 m. tall, thin-walled, green,

with dark hairs near tops of internodes when young. Culm-sheaths

dark-haired on the back, when young black-ciliate on the edges;

auricles low, stiff, dark green, entire, 10 mm. or more in lateral

extent on large sheaths, the end usually somewhat raised and

rounded; blade spreading or deflexed, lanceolate, narrowed at the

base, hairy on upper surface near base; ligule 2-4 mm. high, edge

toothed, bearing short bristles on the teeth when young. Leaf-

blades 15—40 cm. long, 2-45 cm. wide, lower surface glabrous,

stalk 2—5 mm. long; auricles small, glabrous; ligule short, entire

or toothed. Spikelets few in each group, 2—2:5 cm. long, slightly

flattened near the base and hardly 4 mm. wide, gradually nar-

rowed to the slender apex; glumes and lemmas glabrous except

for the fringes of dark hairs 1 mm. long on their edges; florets

commonly 4—5 including the sterile apical lemma; longest lemma

20 mm. long; paleas almost as long as lemmas, hardly 2 mm. wide.

between the keels, keels short-fringed, veins 5 or 6 between the

keels and 2 between each keel and edge; anthers 7 mm. long, with

slender toothed or slightly hairy tips; fruit narrow, hairy only at

apex.

DISTRIBUTION: originally described from Java, where it is probably

not native; formerly planted rather extensively for tall hedges in Sin-

gapore (Thomson and Serangoon Roads) and less commonly in

Penang; not known elsewhere in Malaya, probably native in Burma.

The above description was made from Singapore plants. Penang

plants, both in the Waterfall Garden and at Hutchings School,

have somewhat wider and less reflexed blades on the culm-sheaths,

and auricles more truly auricular in shape; but the general habit

and spikelets are identical. Kurz, in his original description, gives

the name Bamboo Lengka Tali; a plant so named at Bogor agrees

well with those in Singapore. Backer’s description also agrees, ex-

cept as follows: he says the fringing hairs on the lemmas are

118

Vol. XVI. (1958).

yellow-brown, and the anthers 8-10 mm. long (I may not have

seen anthers in the longest floret on a spikelet). Munro, under

‘Oxytenanthera nigrociliata, cites Wallich 5033 from Burma,

which has long slender spikelets very like those of the Singapore

plants of G. hasskarliana; it is however possible that the vegetative

parts of Wallich’s plant were different. Kurz and Backer both give

descriptions of Java plants of Gigantochloa nigrociliata (Buse)

Kurz which certainly indicate something different from Wallich’s

specimen (see Kurz in Journ. As. Soc. Beng. 39 (pt. 2): 88.

1870). Bambusa nigrociliata Buse, described from Java, is un-

doubtedly a Gigantochloa. Munro put it into Oxytenanthera, and

many others have copied him, including Gamble and Camus.

Ridley used the name O. nigrociliata for various different Gigan-

tochloas, rather uncritically.

The name Bambusa auriculata Kurz (later transferred to Gigan-

tochloa by Kurz himself, but only on vegetative evidence), given

to a bamboo from Burma, needs also to be considered. Gamble’s

figure of a culm-sheath attributed to this species (his plate 49)

resembles considerably the lower sheaths of the Penang plant here

included in G. hasskarliana; it was from a plant of which Gamble

had not seen the flowers, and he retained it in Bambusa. I judge

from Kurz’s description that G. auriculata is a big bamboo, which

G. hasskarliana is not. Possibly Wallich 5033 represents the

flowers of G. auriculata.

: Specimens: SINGAPORE, Thomson Road, Ridley 8063 (S); Burkill s.n.

5.8.1918 (S); Holttum s.n. 13.12.1934 (S). Geylang, in a hedge, Ridley

8087 (S,K). Ang Mo Kio, roadside, Ridley 6682 (S,K). Bedok, Ridley

12198 p.p. (S,K). Government House Domain, Holttum S.F.N. 21163

(S,K); Henderson s.n. September 1933 (S). Kampong Teban, Sinclair

S.F.N. 38863 (S). PENANG, Hutchings School, Cheang Kok Choy

S.F.N. 37942 (S). Waterfall Garden, Holttum s.n. 24.10.1946 (S).

4. G. levis (Blanco) Merr. in Amer. Journ. Bot. 3: 61. 1916.

Enum. Philip. Pl. 1: 96 (synonymy).

Basonym: Bambusa levis Blanco, Fl. Filip. ed. 1: 272. 1837.

Synonyms: Gigantochloa scribneriana Merr. in Philip. Journ.

Sci. Suppl. 1: 390. 1906.

G. verticillata sensu Ridl., Flora 5: 260.

G. verticillata sensu Backer, Handb. Fl. Jav. 2:

276, p.p.

Culms to 15 m. or more tall, to about 10 cm. diameter, when

young bearing both pale and brownish hairs on upper part of

internodes, when old smooth and dark green; internodes to about

45 cm. long; lowest nodes somewhat swollen and root-bearing.

Middle culm-sheaths to 30 cm. long, densely dark-hairy (hairs

119

Gardens Bulletin, S.

loose and spreading) almost all over the back when young, top

almost truncate, bearing a raised auricle commonly 7 mm. high

(to 12 mm.) on each side connected by a narrow rim to the base’

of the blade, auricles bearing a few marginal bristles 10-15 mm.

long; blade dark purple on lower sheaths, green on upper ones,

reflexed, relatively narrow, narrowed at the base, hairy towards

base on upper surface; ligule lacerate, undivided part 5-7 mm.

high, lacerations with bristle-like tips, ligule with its fringe of

bristles 15-20 mm. high (fig. 30). Leaf-blades commonly to 30

em. long and 4 cm. wide, larger on small culms, lower surface

finely hairy; sheath-auricles usually distinct, bearing a few bristles

to 4 mm. long; ligule shortly toothed. Spikelets usually on leafless

culms, in groups at the nodes of the branches; groups 1-5 to 12

cm. apart, those on the large branches often of many spikelets,

densely spherical, surface of internodes softly short-hairy; mature

detached spikelets 10-15 mm. long, 4 mm. wide, somewhat flat-

tened, apex acute; basal empty glumes 2 or 3; lemmas 4 or 5,

6-10 mm. long, edges with rather long pale hairs, surface very

finely hairy, veins many, with distinct cross-veins, apex shortly

pointed; paleas narrow, long-hairy on the keels, finely hairy on

edges and back, veins between keels 2—4, 1 or 2 veins between

each keel and edge; uppermost palea slightly keeled, short-hairy

on keels; anthers 4—6 mm. long, tip slightly hairy; style hairy

throughout, stigma undivided.

DISTRIBUTION: originally described from the Philippines where it is

widespread but probably not native; planted also in Borneo (specimens

seen at Kew) and Indochina (Camus); in Malaya found as a village

bamboo or planted in Singapore, Johore, Malacca and Selangor; pro-

duces excellent edible shoots.

RECORDED MALAY NAMES: Buloh Bisa (doubtful); Buloh Suluk (N.

Borneo).

Merrill’s description of his G. scribneriana agrees in spikelet

characters with Malayan specimens except that he describes the

anthers as 7-8 mm. long; he does not describe the culm-sheaths.

He states that the species is “usually if not always planted”, and

no other Gigantochloa is known from the Philippines (also that

no other Philippine bamboo corresponds to Blanco’s description).

In the Kew herbarium good culm-sheaths of the following Philip-

pine collections agree well with those seen in Malaya: Luzon,

Prov. Bulacan, Bur. Sci. no. 11838 (C. B. Robinson); Leyte,

Prov. Visayao, G. M. Weber 1527. ;

I judge that this Gigantochloa would be regarded by Backer as

a variety of G. verticillata. As Ochse (Veg. Dutch E. Ind. p. 317)

states that G. atter produces young shoots which are among the

best for eating, this seems the most likely among the species in-

cluded in G. verticillata by Backer, but Ochse’s accompanying

120

Vol. XVI. (1958).

‘ope[q JO oseq pue s[n3I] sjoym

oy} Surmoys ‘yyeoys Jo do} Jo dovyAns Jouul ‘q *(po}wo ajorine uo sapstiq) Z/¢ X ‘WI pulyoq s[nsI] posiour A[doop

oy} pue (popeys Ajsnbijqo) spore ue suIMoys ‘Yy}eOYsS

JO do} Jo sdvjins s9UUI ‘g *¢/]

X “yleays oj9[dul09 pausyy

| ) I

Jo do} jo jaed Jo sovjins J9jno ‘Og ‘apetq pur yyeoys

BY JO SdvJAINS INO “YW ‘sJ4a] VOTYIOIUDSIH JO YeoYs-WIND ‘OE “B14

A AY

121

Gardens Bulletin, S.

description only describes the sheaths of the shoots as cut for

eating, and these do not show the distinctive characters of sheaths

on actively growing young culms. The figure of G. atter given by

Koorders (Exkurs. Fl. 4, fig. 171) shows a culm-sheath with ap-

propriately large auricles but a tall ligule which is not long-fringed.

My observation of Bamboo Ater at Bogor agrees with Koorders’

drawing and I saw nothing at Bogor quite like the Malayan plants

here described. Kurz’s description of G. atter also differs from my

observations of G. levis. It is possible therefore that G. levis is not

in Java. It may be a Gigantochloa selected originally in Burma

and carried along a different stream of human migration; to the

Philippines via North Borneo ? or to North Borneo from the

Philippines and so to Singapore ? .

I have not noted this as a common species in Malaya; it is cer-

tainly not common in Singapore. Ridley notes on his specimen

no. 119 “this is the common campong (i.e., village) bamboo”,

but I doubt is he was right. (Gamble thought Ridley intended the

word campong as the name of the bamboo). A plant was brought

to the Botanic) Gardens, Singapore, many years ago by Ridley, its

source unrecorded. This plant never grew to a large size, because

it was well known as a source of good edible shoots, until special

arrangements were made to protect it.

This species shows some resemblances to Dendrocalamus asper

in general habit and in appearance of spikelets, and both yield ex-

cellent edible shoots. The differences are as follows: bases of

culms in G. levis are not persistently brown-hairy, upper culm-

sheaths of G. levis have dark hairs, the auricles are not crisped

and have longer bristles, the ligule has a long fringe of bristles (a

much shorter one of fine hairs in D. asper), and spikelets are

longer with longer fringes on the lemmas.

Specimens: SINGAPORE, Bamboo no. 48, cult. H.B.S., and others pro-

pagated from it (origin unrecorded), J. L. Pestana s.n. 26.8.1937 (S,K).

Tivoli, Ridley 119 (S). Pasir Panjang 64 mile, Holttum s.n. 9.12.1934

(S). JoHore, Kota Tinggi, le Doux s.n. 5 November, 1948 (S).

Ma.acca, Maingay 1730 (K). SELANGOR, Salak South, Sth mile from

Kuala Lumpur, Jagoe S.F.N. 37932 (S). Kepong, cult., Symington F.D.

23138 (S, Kep.). .

5. G. scortechinii Gamble in Ann. R. Bot. Gard. Calc. 7: 62,

pl. 53. 1896. Ridl., Flora 5: 261.

Culms 10-20 m. tall, to 12 cm. diameter, close, straight, covered

with a fine white waxy powder when young, green when old, inter-

nodes to about 40 cm. long. Culm-sheaths light orange at the top

when young, rest covered with appressed coarse nearly black

hairs, similar hairs fringing edge of sheath; blade much narrower

than top of sheath, reflexed at right angles, green (except on low-

est sheaths) and leaf-like, at 200 cm. from the ground nearly as

122

Vol. XVI. (1958).

long as the sheath, base somewhat constricted and then spreading

to a dark green narrow rim (auricle) along the top of the sheath,

ends of the rim hardly raised, sometimes with a few stiff pale

bristles; ligule deeply incised, the lobes produced into bristles, the

whole to 10 mm. tall or at the ends to 18 mm. Branches few at

each node but foliage dense; leaf-blades to 35 by 5 cm., lower

surface softly hairy throughout (hairs 0:5 mm. or more long),

stalk 2-4 mm. long; auricles small, usually with a few bristles;

ligule short, sometimes bearing bristles. Spikelets usually on slen-

der leafless branches (often a whole culm flowers without any

leaves) in groups up to 5 cm. apart; mature spikelets 18-20 mm.

long, distinctly flattened and widest near base, 6 mm. wide, all

glumes and lemmas densely brown-hairy on the surface and

brown-fringed on the edges; perfect flowers 4 or 5; lemmas 12-16

mm. long; paleas about 2 mm. wide between the keels, keels

fringed, veins between keels 5 or 6, and one between each keel

and edge; anthers to 9 mm. long.

DISTRIBUTION: Malaya, from Tampin northwards, one of the com-

monest bamboos in the foothills of the Main Range.

- RECORDED MALAY NAMES: Buloh Sémantan, Buloh Télor, Buloh

Rayah, Buloh Pa-aao, Buloh Gala; Buloh Sérémai (var. albovestita).

I have several times seen whole culms of this bamboo flowering,

the remaining culms on the same clump leafy, and I have never

seen a gregarious flowering of all plants in one district. The broad

and very hairy spikelets are characteristic; the culm-sheaths not

so clearly different from what I am calling G. wrayi. The leaves

are very like those of other species of ie ph OA and there is

some variation in hairiness.

There appear to be two slightly differing varieties of this species

in Selangor, both seen near together at Rawang; further study

may reveal more and different variation. One of the two varieties

has strongly orange-coloured young culm-sheaths and a white

waxy covering of the young culms; this has been clearly associated

with typical spikelets of G. scortechinii. The other form has less

orange colour in the young sheaths, and the upper parts of the

internodes of the young culms ae little waxiness but carry a few

pale and brown hairs.

Specimens: NEGRI SEMBILAN, Kangkoi, F.Gd. Hussein F.D. 9596

(S); Buyong F.D. 11005 (K). Jelebu, Kinsey s.n. (K). SELANGOR,

Bukit Hitam, Ridley 7786 (S,K). Gua Batu, Ridley 8170 (S,K). 15th

mile Pahang Track, Ridley 8481, 8483 (S,K). Klang Gates, Ridley s.n.

1.1.1921 (K). Ulu Gombak, Foxworthy F.D. 6429 (S). Gombak Re-

serve, Strugnell F.D. 10549 (K). Salak South, Kuala Lumpur, F.Gd.

Md. Soh F.D. 13820 (S, Kep.) 43rd mile Gap Road, Wyatt-Smith

F.D. 76168, 76169 (Kep.). Near Rawang, Holttum S.F.N. 37789 (S).

Ginting Simpah Road, Holttum S.F.N. 38415 (S). Gunong Semangkok,

123

Gardens Bulletin, S.

Gap, 2,000 ft., Curtis 3746 (S,K). Near Rawang, Holttum s.n. October

1946 (S). Ulu Kerling & Ulu Selangore, Kunstler 8572 (K). PERAK,

without locality, Scortechini (K, type). Temengoh, Ridley 14383 (S).

Behrang F.R., C. Smith F.D. 1160 (S). Grik, F.R. Hamid F.D. 6419

(S), F.D. 8253 (S,K). Near Grik, Burkill & Haniff S.F.N. 12479

(S,K). Jor, Burkill & Haniff S.F.N. 14238 (S). Upper Perak, Wray

3433 (S,K). PAHANG, Kuala Lipis, Somerville F.D. 10488 (K), F.D.

10490 (K); Marshall F.D. 21385 (Kep.). Ulu Chineras, Burkill &

Haniff S.F.N. 15690 (S). Bentong, Burkill & Haniff S.F.N. 16417 (S),

S.F.N. 16435 (S). Near Raub, Burkill & Haniff S.F.N. 16855 (S). Ulu

Kuantan, Craddock s.n. 4.3.1903 (S). PENANG, Cult. Waterfall Garden,

Curtis 10849 (S,K); Pestana s.n., April 1934 (S,K). KEDAH, Ulu Pan-

tai Mulik, Sow F.D. 34625 (Kep.). Kampong Naka, Holttum S.F.N.

19826 (S). KELANTAN, Tanjong Telok Lalu, Haniff & Nur S.F.N.

10206 (S,K). Sungei Betis, Henderson S.F.N. 29704 (S,K).

Var. albovestita Holttum, var. nov.

Vaginae culmorum extus pilis albis vestitae; internodia culmo-

rum apicem versus pilis albis rigidis instructa.

Culms slender (2:5—3-5 cm. diameter); leaves on smaller

branchlets 12-18 cm. long, 10-15 mm. wide, on larger ones to

25 cm. long and 20-26 mm. wide; culm-sheath ligule 13-15 mm.

tall at ends, often lower in the middle, cut almost to the base

throughout, auricles low, firm, not bristly; no flowers.

TYPE: Kedah, 9th mile from Alor Star to Pokok Sena, Holttum,

K 11, December 1953 (S). ;

Specimens from Grik probably also belong here. These have

culms 7 cm. diameter, and white-hairy sheaths. The ligules on the

sheaths are broken so much that their original height cannot be

judged. The specimens are: Grik, F. R. Hamid, F.D. 6425 (S),

8252 (S,K).

6. G. wrayi Gamble in Ann. R. Bot. Gard. Calc. 7: 64, pl. 55

(as regards flowering specimens only). 1896. Ridl., Flora 5:

261.

Synonym: G. kurzii Gamble |.c. 65, as regards specimens from

Malaya. é

Culms 2-7 cm. diameter, not hairy, walls of moderate thick-

ness. Culm-sheaths to at least 27 cm. long, dark-hairy when young,

line of junction with blade nearly horizontal; auricles firm, low,

raised near the outer ends, often bearing a few bristles near the

outer ends, edge otherwise smooth; blade reflexed, green when

young, relatively narrow and narrowed at the base, of middle

sheaths about 12 by 3-5 cm., of upper sheaths to 22 by 4 cm. or

longer, dark-hairy on upper surface; ligule deeply incised through-

out, the lobes prolonged into bristles, the whole (including bris-

tles) 6-10 mm. high (fig. 31). Leaf-blades 9-40 cm. long, 1-2-6

cm. wide, lower surface finely hairy, stalk to about 8 mm. long;

auricles small, often with bristles; ligule short, edge irregularly

124

Vol. XVI. (1958).

toothed but usually not bristly. Spikelets to about 2 cm. long and

5 mm. wide, with 3 or 4 perfect florets (fig. 27); lemmas to 17

mm. long, with a short stiff apex, edges fringed with brown hairs

1 mm. long, exposed parts of the surface very finely short-hairy;

paleas fringed on the keels (fringe sometimes near apex only);

anthers to 8 mm. long; ovary hairy at apex only, style long, stigma

undivided; fruit about 10 mm. long.

DISTRIBUTION: as here interpreted (not with certainty; see discussion

below) native in the northern part of Malaya, and probably planted in

villages; probably native also in the region north of Malaya.

RECORDED MALAY NAMES: Buloh Béti, Buloh Mata Rusa, Buloh

Manis, Buloh Semantan, Buloh Minyak.

Gamble’s type specimens (Wray 1895 from Bukit Gantang,

Perak) are mounted on two sheets, on one a leafy branch, on one

a flowering branch. The leafy branch is to me indistinguishable

from Schizostachyum brachycladum, the shape of the well-devel-

oped leaf-auricles being very distinctive and unlike anything |

have seen in Gigantochloa. The Malay name recorded with the

specimen is Buloh Plang, commonly used for Schizostachyum. |

consider therefore that the species should be typified by the

flowering branch only; but spikelets alone are not very satisfactory

for characterizing a species in this genus. As regards Gamble’s

description of the spikelets, he makes the statement that the keels

of the palea are “not or only very faintly ciliate”. I examined a

spikelet from the Calcutta sheet of the type collection (kindly sent

to me in Singapore by Dr. K. Biswas) and found that the paleas

had quite normal long fringes; a spikelet from the Kew sheet (from

Gamble’s own herbarium) had keels rather sparsely ciliate, to-

wards the apex only. The hairiness of the keels thus seems vari-

able, and I do not think it has any diagnostic value. The Singapore

specimen has no spikelets.

Gamble identified no other specimen as G. wrayi. I have how-

ever included here, with some doubt, the specimens from Malaya

called G. kurzii by Gamble and Ridley, and some others. They

have on the whole somewhat larger spikelets than those of the

type of G. wrayi and the lemmas have usually fringes of a lighter

brown. These specimens are vegetatively difficult to separate

clearly from G. scortechinii; their spikelets, where known, are

smaller and much less hairy than those of G. scortechinii. There is

need for a more careful survey of Gigantochloas in the north of

Malaya in order to distinguish this species (or species-group)

more clearly. It may be that the extreme forms of G. scortechinii

and G. wrayi are connected by a number of intermediate hetero-

zygous forms.

125

Gardens Bulletin, S.

‘apeyq JO oseq pue o[NsI] SuIM

aseq (pepeys Ajasoys) 9[d1in

yoyjoue JO ooRyINs Joyo YUM “Yyeoys pouoHey soyM Jo govjins JOUUl

A ot!

eB ue SUIMOYS

AW \\w \Wilt Mil if

: pa gipentenn iene Munn FR hey BOM E99 yy) 91m,

oys ‘yyeoys Jo do} Jo y1ed Jo advjiNs JouUl fe are

X ‘apn31] oy}. Jo Jed pue ‘gpeyq Jo a

‘yyeoys JO do} Jo wed JO 9d¥JINS 19}NO0 ‘gq ‘2 X ‘posodurtiedns yyesys

x 1

‘wh

at hd MOOT ny

ee ey 7

\y will ,

‘y ‘1KD4M vO]YIOJUDSIHH JO Yeoys-wWIND “TE ‘SI

a Ge? g

vil ;

Vol. XVI. (1958).

The type of G. kurzii Gamble came from Burma. Parker sug-

gested that it was not different from G. apus. Gamble later agreed.

with this suggestion (as evidenced by a note in his copy of the

monograph), and Backer has united G. kurzii and G. apus. There

is no doubt that the Peninsula specimens called G. kurzii by

Gamble are not G. apus, which is separately described in the

present work.

Specimens: SINGAPORE, Lower Garden, Ridley 12501 (S,K). Garden

Road near Office, Ridley s.n. 1903 and 27.4.1908 (S). Cluny Road near

Gardens Office, Pestana s.n. 9.9.1937 (S,K). SELANGOR, Kanching F.R..,

Abd. Rahman F.D. 37600 (Kep.). PERAK, Bukit Gantang, Wray 1895

(the flowers only; type of the species, S,K). Lumut, Ridley 3114 (S,K).

Grik, F.R. Hamid F.D. 6420 (S.K), F.D. 8254 (S,K). Taiping, Bur-

kill & Haniff S.F.N. 13118 (S,K); Wray 134 (S,K). Tapah, Pahang

Road 10th mile, Burkill & Haniff S.F.N. 13452 (S,K); Holttum S.F.N.

38420 (S); 44 mile, Holttum S.F.N. 38426 (S). South of Lahat, Bur-

kill & Haniff S.F.N. 13920 (S,K). Kuala Kangsar, Haniff S.F.N. 14921

(S). Changkat Jong, Holttum S.F.N. 38431 (S). Batu Gajah, Barnard

s.n. August 1915 (S,K). Tapah, Curtis s.n. October 1894 (S). Pro-

VINCE WELLESLEY, Permatang Bertam, Ridley 6999 (S,K). PENANG,

Balik Pulau, Cheang Kok Choy S.F.N. 37941 (S). Kepan, Bukit

Jamboi, Holttum S.F.N. 19819 (S). Lower Siam, Kopah, Haniff

12608 (S). Kasoom, Curtis 3237 (S).

G. wrayi Gamble, large variety ?

Near the road to Cameron Highlands, at Jor (1,500 ft.) I was

shown a large Gigantochloa (culms 10-12 cm. diameter) with

culm-sheaths corresponding to the above description of G. wrayi

(a description made from Singapore plants). The Jor plant had

no flowers. The local Malays gave the name Buloh Périm to this

bamboo, and said it was the best kind for making baskets (i.e.,

good pliable strips could be obtained by first radial and then tan-

gential splitting).

Specimens: Jor, Holttum S.F.N. 38423 (S). Third mile from Tapah,

in a Malay compound, Holttum S.F.N. 38424 (S). (Latter plant had

no good sheaths).

7. G. ridleyi Holttum in Gard. Bull. Singap. 15: 275. 1956.

Culms to about 16 m. tall, and to 10 cm. diameter; internodes

glabrous, green. Culm-sheaths long-persistent, to 25 cm. long, the

back covered with almost black appressed hairs, the top united to

the blade without any external line of separation; blade not decidu-

ous, erect, when young appressed to the culm, triangular, line of

junction with the sheath (as shown by ligule) upcurved, blade of a

middle sheath 8—10 cm. long, 6 cm. wide, not narrowed at the base

but decurrent along the top of the sheath as a low rigid auricle on

each side, the edge of the blade near its base, and adjacent parts

of the auricles, bearing short curved bristles; ligule to 3 mm. high,

almost entire, the edge somewhat undulate and slightly toothed,

eat

Gardens Bulletin, S.

not bristly (fig. 32). Leaf-blades to 40 cm. long and 6 cm. wide,

glabrous on the lower surface apart from a few hairs near the

midrib, stalks of lower leaves 4-5 mm. long, of upper leaves 10

mm. long; young sheaths hairy, auricles small and glabrous, ligule

low and entire.

Fig. 32, Culm-sheath of Gigantochloa ridleyi. A, inner surface of whole

flattened sheath, showing complete ligule with the ends of the

auricles rising above it. B, outer view of half of top of sheath,

showing relation of the low auricle to the base of the blade. C,

inner view corresponding to B, showing ligule and part of base

of blade.

Type: cultivated in Botanic Gardens, Singapore, lawn W, origin

Province Wellesley; specimens coll. J. L. Pestana in herbaria at Singa-

pore and Kew.

128

Vol. XVI. (1958).

The plant from which the type specimen was taken was brought

by Ridley from Province Wellesley to Singapore, at a date not

recorded (Ridley s.n., S). The Singapore plant now forms a large

clump, and in more than 40 years has not been known to flower.

The auricles and ligule of the culm-sheaths are very like those of

G. maxima, for which reason I have no doubt of the genus; but I

have never seen another Gigantochloa with the peculiar broadly

triangular, erect, persistent culm-sheath blades, nor have I seen

such described. The original plant in Province Wellesley was in a

village, not wild, and probably came from further-north. I once

saw a plant near Kota Tinggi, Johore, very similar to the type of

G. ridleyi, but took no specimen.

The young shoots of this bamboo are astringent, not good to

eat. The culms are strong and quite large, and would probably

serve the same purposes as those of G. maxima.

8. G. ligulata Gamble in Ann. R. Bot. Gard. Calc. 7: 67. pl. 58.

1896. Ridl., Flora 5: 262.

Culms 6-9 m. tall, to 4 cm. diameter, walls usually about 10

mm. thick; internodes to 37 cm. long, sometimes slightly streaked

with paler green; upper part of young internodes bearing scattered

stout very dark brown hairs nearly 1 mm. long, and very fine

white hairs 0-2 mm. long; branches from middle nodes few, none

from upper nodes. Culm-sheaths 14-22 cm. long, rather thin,

hairs very dark brown, not very copious and quite deciduous

from old sheaths; top of sheath upcurved at attachment to blade

but line of attachment not clearly marked externally and blade not

deciduous; auricles more or less down-curved on the edges of the

top of the sheath on each side of the base of the blade, forming

low firm smooth rims 1 mm. high and 1-5 to 2:5 cm. in lateral

extent, quite lacking bristles; ligule 5-6 (—10) mm. high in the

middle, 20 mm. high at the ends, incised 4/5 of its width through-

out, the middle incisions closest; blade erect, commonly 44:5

cm. wide, on a middle sheath 10 cm. long, on higher sheaths to

18 cm. or more long, thin, hardly hairy on inner surface, base

narrowed a little and then decurrent into the auricles (fig. 33c).

Leaf-blades on flowering stems to 27 cm. long, 3-7-6 cm. wide,

glabrous or very short-hairy beneath, base of lower leaves broadly

cuneate, of upper ones narrowly cuneate, petiole about 3 mm.

long; half of callus below attachment of blade of upper leaves

produced into a thin blade 4-16 mm. long; auricles hardly more

than raised lines, upcurved along the line on which the top of the

Sheath meets the ligule; ligule thin, varying in height, highest in

the upper (distal) leaves on a branchlet and then 2-3 cm. long,

129

Gardens Bulletin, S.

deeply bilobed; leaves on sucker shoots to 10 cm. wide, ligule to

4 cm. high, callus-lobe to 2 cm. long. Flowering branches droop-

ing, 80 cm. or more long, little or not branched; nodes bearing

spikelet-groups at base 4—5 cm. apart, near tips of branches 1 cm.

apart, each group at first enclosed by a sheath 15 mm. long;

mature spikelets 12-15 mm. long when detached, lemmas fringed

with dark hairs (old ones may fade to a lighter colour), perfect

flowers 2—4; paleas fringed on the keels towards the apex, veins

between keels 2—5, between each keel and edge 1-2; anther

6-8 mm. long. 7

Fig. 33. Gigantochloa latifolia. A, complete flattened culm-sheath, inner

surface, showing the tall deeply incised ligule. B, top of outer

surface of culm-sheath, showing the low auricles with the ligule

rising above them. C, G. ligulata, part of a leafy branchlet,

showing bases of two leaves with their very long ligules, a callus

at the base of each leaf, and a narrow auricle rising from behind

each callus to join the side of the ligule.

DISTRIBUTION: northern part of Malaya; southern Siam.

RECORDED MALAY NAMES: Buloh Tikus, Buloh Tilan (Pahang).

This species was originally described from two flowering speci-

mens, one from Perak (Wray 845) and the other from Pahang

(Kuala Tahan, Ridley 5597), the latter in poor condition. These

specimens lack culm-sheaths, in which I find much variation in

Kedah, and the enlarged callus below a leaf-stalk is only 2-3 mm.

130

Vol. XVI. (1958).

long. The description given above is based on specimens gathered

by me from open country near Alor Star, Kedah, where the species

is abundant. I suggest that this should rank as the typical form of

the species; as so interpreted, it is a smallish, thick-walled bamboo.

The young shoots, though small, are said to be of good edible

quality. In bamboo thickets in northern Kedah there are larger

bamboos with similar ligules of culm-sheaths and leaves, some-

what variable in detail, as described below; and similar bamboos

occur in Perak near the road to Cameron Highlands, so that they

are probably abundant in northern Malaya generally.

G. ligulata, larger varieties.

Culms 5—6 cm. diameter (to 8 cm. ?); sheaths to 30 cm. or

more long, more rigid than in the typical form, junction between

blade and sheath ultimately distinct; culm-sheath ligule always

c. 2 cm. high at ends, lower in the middle, the ends in some cases

entire, in some cases lacerate, the middle part always in some

degree lacerate, sometimes deeply so; half of callus of upper

leaves on each branchlet often elongate and membranous, variable

on the same plant.

RECORDED MALAY NAMES: Buloh Bilalai, Buloh Gala, Buloh Mata

Rusa, Buloh Bélang, Buloh Hitam, Buloh Méliau.

The following key is an attempt to indicate the nature of the

variation observed among specimens of the larger varieties of G.

ligulata. The numbers prefixed by the letter K refer to specimens

collected by me in northern Kedah in 1953; letters A—D refer to

previous collections from the same area.

Culm-sheath auricles bearing long bristles

(2 cm.)

Leaf-blades glabrous beneath; about 40 bris-

tles on each auricle ie it: ae

Leaf-blades very short-hairy beneath; bris-

tles on auricles fewer me ahs sO

Leaf-blades very short-hairy beneath; bris-

tles on auricles many; lemmas not fringed,

or with few pale hairs $3 ap gs

Culm-sheath auricles without bristles

Leaf-blades glabrous beneath .. .. A; S.F.N. 38430,

38422, 38429,

38421

Leaf-blades short-hairy beneath ae BP SPIN, 38487.

131

Gardens Bulletin, S.

Apart from the vegetative variation, there appears to be varia-

tion in the spikelets, smaller spikelets having only 2 perfect flow-

ers, larger ones 3 or 4; how far there may be such variation on

one plant is not known. Lemmas are normally dark-fringed, but

K7 has lemmas almost or quite lacking a fringe, and short pale

anthers (5 mm. long). The other specimens with bristly cuim-

sheath auricles lack flowers; if it should be shown that all such

plants have glabrous lemmas and small anthers, they might con-

stitute a new species, but such an association of characters cannot

be assumed from one specimen. Another specimen (D) found on

a previous occasion in N. Kedah had similar spikelets, but the

leaves had ligules only 2 mm. high; the culm-sheaths were very

imperfect.

I am not sure that a sharp distinction between G. ligulata and

G. latifolia can be maintained: The present distinction is on length

of leaf-ligule and relative width of culm-sheath blades, but in both

of these there is much variation, even on the same plant. Perhaps

all the bamboos here included under G. ligulata and G. latifolia

constitute an inter-breeding complex of hybrids. Certainly some

seeds are formed by these bamboos in Kedah, and the plants do

not die after flowering.

Specimens of small form of species or size unrecorded: SELANGOR,

Kuala Lumpur, per A. Arber (K). PERAK, Tapah, Ridley 14034 (S,K).

Sungei Siput, F.Gd. Suleiman F.D. 8325 (Kep.). Cameron Highlands

Rd. 10th mile, Holttum S.F.N. 38428 (S). Ipoh, Curtis 3180 (S). G.

Kerbau, Haniff 3991 (S). Kuala Wok, Wray 845 (K). PAHANG, Kuala

Tahan, Ridley 5597 (S,K). Endau Rompin, Mahamud F.D. 15537

(S). KELANTAN, Sungei Keteh, Md. Nur S.F.N. 11951 (S,K). KEDAH,

Tangga Reserve, Dolman F.D. 21544 (S). Bukit Tangga, Symington

F.D. 46968 (Kep.). North of Changlon, Symington F.D. 56959 (Kep.).

Nerang Rd. 18th mile, Burkill & Haniff S.F.N. 13319 (S,K). Langkawi,

G. Raya, Corner & Nauen s.n. 21.11.1941 (S). Sintok Rd., Holttum

s.n. 26.10.1946 (S). PERLIs, Near Kangar, Ridley s.n. March 1910

(S); Henderson S.F.N. 22999 (S). Rinta Mas F.R., Symington F.D.

57000 (Kep.). SETUL, Ridley 14837 (S,K). Lower Siam, Haniff &

Nur S.F.N. 2939, 3887 (S).

Larger varieties: PERAK, Grik, Hamid F.D. 6424 (S), F.D. 8256

(S,K). Cameron Highlands Rd. 2,500—3,000 ft. Holttum S.F.N. 38417

(S); do., 15th mile, Holttum S.F.N. 38421 (S), 38422 (S); 11th mile,

Holttum S.F.N. 38429 (S); 14th mile, Holttum S.F.N. 38430 (S).

KEDAH, Sintok Road, Holttum s.n. (A), (B) and (C) 26.10.46 (all S).

North Kedah, Holttum K2, K3, K7 (December 1953, S,K).

9. G. latifolia Ridl., Flora Mal. Penin. 5: 262. 1925.

Closely related to G. ligulata, differing as follows: culm-sheath

blades proportionately wider (on an eye-level sheath of a large

culm 15 by 8-5 cm., on a smaller plant 9 by 4 cm.); ligule of

culm-sheath not taller at the ends than in the middle, 15 mm. tall

132

Vol. XVI. (1958).

throughout on a large sheath, 7 mm. on a small one, incised 1/3

to 1/2 way towards the base in the middle and more deeply to-

wards the ends; leaf ligule commonly 3—4 mm. high, sometimes to

8 mm., sometimes with a few fine bristles on its edge; callus on

back of leaf-sheath not produced to form a membranous wing;

leaf-sheath auricles occasionally with small bristles; leaf-blades

usually in part very short-hairy beneath and slightly glaucous be-

neath when dry; spikelets 2—4-flowered, lemmas fringed with pale

hairs (fig. 33 A, B).

DISTRIBUTION: northern Malaya.

The type of this species was collected at Kuala Teku, Pahang,

by Seimund (no. 368, at Kew only). Ridley stated that the spike-

lets contained only one flower, but I have found two in a spikelet

from the type specimen. The type has no culm-sheaths, and I am

therefore not certain that specimens from Kedah are the same

species; the above description of culm-sheaths is taken from

several Kedah specimens, and seems uniform.

This species can grow to a fairly large size, similar to that of

the large varieties of G. ligulata; I have seen culms about 8 cm.

diameter, with walls 14 mm. thick. The young internodes bear

stout dark hairs where exposed just below the nodes. The culm-

sheath auricles are not bristly; they form firm smooth rims 2-3

mm. high along the top of the sheath each side of the blade.

Specimens: PAHANG, Sungei Tahan, Holttum S.F.N. 20075 (S). S.

Tahan, Kuala Teku, Seimund 368 (K, type), 227 (S). Fraser’s Hill,

4.000 ft., Holttum S.F.N. 39460 (S). PERAK, Tapah, Ridley 14035

(S.K). KepanH, Alor Star, Ridley 14838 (S,K). Kedah Peak 200 ft.,

Haniff & Nur S.F.N. 5153 (S). Langgar, Burkill & Haniff S.F.N. 13311

(S,K). Bukit Jamboi 700 ft., Holttum S.F.N. 19818 (S). Sintok Road,

Holttum s.n. (D) 26.10.1946; K4, K5 (28.12.1953; S,K).

Var. efimbriata Holttum, var. nov.

Lemmata non fimbriata, fere glabra; laminae foliorum infra bre-

viter pilosae.

SINGAPORE, Ridley 12198, p.p. (S,K).

Var. alba, Holttum, var. nov.

Pili vaginarum et internodiorum culmorum albi; laminae vagi-

narum culmorum supra prope basin dense albo-pilosae.

KEDAH, Langgar, in village by river, Holttum K12, December 1953

(K). Local name: Buloh Pahit (young shoots bitter).

The culms of this bamboo were 5:5 cm. diameter, walls 10 mm.

thick; a sheath from a low node on the culm was 20 cm. tall, with

blade 7 by 3:2 cm., an upper sheath 32 cm. tall with blade 35 by

4-5 cm.; culm-sheath ligule 8-10 mm. high throughout, incised

more than half way to the base.

133

Gardens Bulletin, S.

RACEMOBAMBOS

Racemobambos Holttum in Gard. Bull. Singap. 15: 267. 1956.

Type species: Bambusa gibbsiae Stapf in Journ. Linn. Soc.

Bot. 42: 189. 1914.

Spikelets in a raceme at the end of a leafy branchlet (fig. 34);

lateral spikelets shortly stalked, in the axils of very small bracts,

terminal spikelet on a longer stalk; each spikelet consisting of two

empty glumes (the lower one usually very narrow), several per-

fect florets separated by rather long rachilla-internodes, and an im-

perfect terminal floret; lodicules usually 3, fringed with hairs;

stamens 6, filaments free, anthers not apiculate; ovary small,

cylindric, its hairy apex globose or ovoid, wider than the cylindric

lower part; stigmas 3, slender, arising directly from the swollen

apex of the ovary.

A genus of five known species, one from Gunong Pulai in

Malaya and the others from the mountains of Borneo. The ovary

is quite like that of some species of Bambusa, but the arrangement

of the spikelets is quite different, resembling Arundinaria.

R. setifera Holtt. in Gard. Bull. Singap. 15: 271. 1956.

Rhizome sympodial, creeping, horizontal length between one

culm and the next c. 10 cm., 5 mm. diameter, internodes 5-10

mm. long. Culms seen less than 10 mm. diameter, length not re-

corded, sheaths not known; leafy branches 20—25 cm. long, bear-

ing 5—8 leaves and sometimes spikelets towards the apex, the

spikelet-bearing part of the branch little longer than the last leaf-

sheath. Leaf-blades 10-18 cm. long, 8-12 mm. wide, lower sur-

face bearing scattered hairs 1 mm. long, stalk hardly 1 mm. long;

auricles slightly raised, bearing slender bristles; ligule small.

Axillary spikelets few, on stalks 2-3 mm. long; lower glume 6

mm. long, 1 mm. wide, 1-veined; upper glume 10 mm. long, 2:5

mm. wide, 9-veined, edges fringed, slightly keeled near apex but

not apiculate; rachilla-internode between upper glume and first

lemma 1-5 mm. long, between first and second lemmas 3 mm.,

remaining internodes 4:5 mm. long; lemmas 14-15 mm. long,

edges fringed, apex setiform and 3 mm. long; paleas 10 mm. long,

apex shortly bilobed, keels shortly fringed near the apex; lodi-

cules 3 mm. long including a fringe of hairs 0-5 mm. long; anthers

5 mm. long; ovary proper 0-8 mm. long, the swollen apex nar-

rowed upwards and 2 mm. long.

DISTRIBUTION: Gunong Pulai, Johore; possibly also on the hills near —

Batu Pahat. |

134

Vol. XVI. (1958).

The only flowering specimen was collected by Best in G. Pulai

(S.F.N. 7707, in Herb. Singap.). Ridley collected a sterile speci-

men of a slender climbing bamboo at a locality named Minyak

Buku (no. 10991, S), apparently on one of the hills near Batu

Pahat; this specimen agrees vegetatively with Best’s so far as one

can judge from leaves.

——_—_—— -— oe ee _—— © aeee oe oe @

Fig. 34. Racemobambos setifera. A, a raceme of spikelets at the end of a

leafy branchlet, with part of the last leaf, x 24. B, one rachilla-

internode bearing lemma and palea. C, diagrammatic interpre-

tation of A, with stalks of spikelets and rachilla-internodes

nee outlines of last two leaves, and of their sheaths,

dotted.

The Eugleninae of Malaya

By G. A. PROWSE

Fish Culture Research Station

Batu Berendam, Malacca

Summary

AN ACCOUNT is given of 125 species and forms of Eugleninae

found in Malaya, of which 11 are described for the first time.

The Eugleninae form a very important part of the micro-flora

in standing freshwater in Malaya, particularly where there is a

great deal of organic decay going on, or where the water harbours

an abundant plant-growth. Indeed, they may be so abundant as to -

colour the water; for example, the brick red scums so common

on Chinese fish-ponds, and on newly flooded rice-fields, are al-

most entirely due to Euglena sanguinea Ehrenberg. However, de-

spite their abundance, the Eugleninae have been very little studied

here, and I have been able to find reference to only three species

in all the literature on the Malayan freshwater micro-flora. This

is because the amount of algological work which has been done

in Malaya has been very limited, and most of it has been carried

out on preserved material taken to laboratories elsewhere. The

Eugleninae, on the other hand, must be studied in the living, and

freshly killed state, if sufficiently accurate identification is to be

carried out. The present paper deals with 125 species and forms,

of which 11 appear to be new. This is certainly not the total

number of species to be found in Malaya, and already one or two

other forms have turned up, differing from those described in this

paper, but which will need further study before they can be ade-

quately identified. Slides of the new species and varieties will be

deposited in the Herbarium, Botanic Gardens, Singapore.

The Eugleninae form a class of flagellates which are highly dif-

ferentiated, but whose origins are obscure. Characteristically they

are naked, without a cellulose cell wall as in the Chlamydomona-

daceae of the Chlorophyceae. The periplast, or bounding mem-

brane of the protoplast may be relatively soft so that the cell may

be very metabolic, changing its shape with ease, as in Euglena,

or it may be rigid, permitting very little change in shape, as in

Phacus. In most cases the periplast is striated, although the stria-

tions may be fine or coarse, and in some cases the membrane may ~

136

Vol. XVI. (1958).

be ribbed as well. In Trachelomonas, Strombomonas and Ascog-~

lena, the protoplast is enclosed in a firm envelope, or lorica,

usually of very distinctive shape, and the protoplast can often be

seen squirming within the envelope. These encapsuled types par-

allel the occurrence of Phacotus in the Chlamydomonadaceae. One

genus, Colacium, forms attached dendroid colonies, the cells ac-

quiring flagella and leaving the colony only during reproduction.

Two species of Euglena also form attached stages, but they can

readily leave the colony, even when not dividing.

All the Eugleninae have a large, centrally-placed nucleus, but

the most characteristic feature of the class is the vacuolar system.

At the anterior end an invagination of the periplast forms a

narrow canal, the cytostome, which leads inwards to an enlarged

vacuolar swelling, the reservoir. These are practically always visi-

ble, and their presence is indicated by an apical notch, which in

some cases may be distinctly one-sided. There may be one or two

flagella passing in through the cytostome and terminating at the

base of the reservoir. In the case of uniflagellate species the single

flagellum is forked at the base, where it enters the reservoir, sug-

gesting that the biflagellate condition is probably primitive, and

that the single flagellum has arisen by fusion of two. (Some

authors claim that the flagellar ends extend right down to the

nucleus, but evidence on this point is both varied and confusing. )

In all the Eugleninae the products of assimilation appear as solid,

often quite large granules of paramylum, either as rods, discs,

rings or other shapes, the shape usually being constant for a par-

ticular species. Paramylum is a polysaccharide, allied to starch,

but which does not stain with iodine or chlorzinc-iodide, is in-

soluble in boiling water, but which will dissolve in concentrated

sulphuric acid or in potash.

The Eugleninae can be divided quite naturally into the pig-

mented forms, comprising the family Euglenaceae, which contain

chloroplasts and are usually green in colour, and the colourless

forms consisting of the two families Astasiaceae and Peranemaceae.

Cyclidiopsis Korschikow, which is usually separated in the family

Cyclidiopsidaceae, is completely devoid of chloroplasts, but pos-

sesses a stigma; it shows such close resemblance to species of

Euglena that perhaps it ought to be included in the Euglenaceae.

Some species of Euglena have been observed to lose their chloro-

plasts and live saprophytically under special cultural conditions.

The Astasiaceae are without stigma (except Khawkinia Jahn &

McKibben) and chloroplasts and live saprophytically, but one or

two species of Astasia come very close to colourless forms of

137

Gardens Bulletin, S.

Euglena. The Peranemaceae are decidedly more specialised, show-

ing a marked tendency towards holozoic nutrition. There is in

most cases a special organ, the siphon, which in some cases

appears as two short parallel rods next to the cytostome and reser-

voir. In Entosiphon it forms a cone shaped tube running nearly the

full length of the cell, and capable of being extruded (fig. 8k).

The function of these structures is still obscure, and they have

been variously described as pumping organs, or the means of.

ingestion of solid particles.

For the purpose of identification I have largely depended on the

following works: “Das Phytoplankton des Siisswassers” Vol

iv ‘Euglenophyceen’ by Hiiber-Pestalozzi 1955, “The genus Eug-

lena” by Gojdics 1953, “Materiaux pour un Monographie de

Trachelomonas, Strombomonas et Euglena” by Conrad & van

Meel 1952, “Etude systematique du genre Lepocinclis” by Conrad

1935, “Synopsis der gattung Phacus’” by Pochmann 1942, and

various papers by Déflandre, Skvortzow, Playfair and others. In

some genera there: has been a tendency for the Malayan specimens

to be larger than those elsewhere, whereas in some other genera

they have been smaller. Such size variations, unless they are very

marked, are not of great significance, especially when we know so

little about the nutrition and growth of these organisms. In such

cases I have preferred to retain the Malayan specimens under the

species name rather than separate them as varieties.

Euglenaceae

Possessing green chloroplasts, ranging from discoid to band-

shaped, with or without pyrenoids. The green colour may in some

cases be obscured by the production of haematochrome. Colour-

less forms appear, but are so obviously related to the pigmented

forms that there is no difficulty in identification.

Key to Genera

Cell with one flagellum

I. Cell freeliving, solitary.

a. Cell without an envelope.

1. Cell decidely metabolic. Paramylum of various

SADCS os sinic sos p>, 0's 0) Sr Euglena.

2. Cell rigid

(a) Cell more or less round in cross-section,

paramylum two lateral rings

Lepocinclis.

(b) Cell more or less flattened ..... Phacus.

138

Vol. XVI. (1958).

b. Cell with an envelope

1. Envelope with neck or porus distinctly set off;

often sculptured, or bearing spines

a Trachelomonas.

2. Envelope with the neck tapering off from the

main body; rarely sculptured .. Strombomnas.

II. Cell forming attached dendroid colonies, only leaving them

cy Tretia fics ale eer Riess oes Colacium.

Euglena Ehrenberg

Cell usually free swimming, or crawling, fusiform to elongate-

cylindrical, often twisted, usually very metabolic. Flagellum one,

of varying length. Periplast usually distinctly striate. WVacuolar

system a typical cytostome and reservoir. Chloroplasts band-

shaped, reticulate or discoid, with or without pyrenoids. Paramy-

lum long rods, discs, and also sheathing pyrenoids in some species.

Green colour often masked by haematochrome.

Key to the Malayan species

Se OTOP lasts eesens eta pOd 6) es bbl sera ek wie ne i

yeinloroplasts rencwlate 24S ee ee. (iii) E. mainxi.

. Chloroplasts small (less than 6) discoid or ovoid ..... 3;

. Chloroplasts discoid, larger (over 6) ..........-..4. 12.

PR AMOLOPlasts Symes aped = os...) a 5 os ale bs acwin sn oss’ 14.

. One chloroplast, long and flattened ...... (i) E. elongata.

. Many chloroplasts arranged spirally, ends nearly touching

(ii) E. synchlora.

Sody CILGUIAL an cles SECON. 315652 0b ee i 4.

Deueeeny TiAMTOMCU nN Eel il. oe Pec cee ee ena 8.

poay triancular im Cross-Section . 0... cies enn. 11.

Soe AL ACIAN AE MEINEM A 5 gs tases di a Winn! Surredmin mie 6 aces os 3.

. With haematochrome, colouring cell red

(vi) E. chlorophoenicea.

ee APA YIUIEN CRC VON oi aes cand o cine oem 6.

. Paramylum long rods or rectangular plates ........... ci

. Pellicle delicately striated, cells not attached

(iv) E. proxima.

139

10.

£2.

12.

13.

14.

Gardens Bulletin, S.

. Pellicle faintly striated, cells attached to crustacea but readi-

ly leaving the, host shai aii-= -s ee (v) E. cyclopicola.

. Cell very long, ending in a sharp point. Flagellum short

(vii) E. acus.

. Paramylum granules small’. ©. 34. . 5.07. 9,

. Paramylumi in large links ©... >... <. . se 10.

. Paramylum in short rods, ellipsoids or spheres. Body band-

shaped, rounded at both ends .... (viii) E. ehrenbergii.

. Paramylum oval, hexagonal in outline. Body band-shaped,

with posterior end truncate or hollowed

(ix) E. subehrenbergii.

Pellicle yellowish brown, bearing squarish beaded excres-

cences along: the Stidetgc s...00:.) 5.8). eee (x) E. fusca.

Pellicle hyaline without excrescences, but with a long dorsal

keel-like fold. Cell twisted throughout its length

(xi) E. oxyuris var. charkowiensis.

Body triangular in cross-section, twisted. Paramylum in long

rods or flat plates ................ (xii) E. tripteris.

No pyrenoids, Chloroplasts oval or round. Paramylum in

short rods. of TeCtamgies 3.3.08) oi. (xiil) E. refringens.

With paramylum-sheathed pyrenoids .............. Ta;

No conspicuous sub-pellicular granules even when stained

with neutral red. Cell spindle-shaped .. (xiv) E. caudata.

With conspicuous sub-pellicular granules, staining with

neutral red and arranged in spirals between the striae.

Cell spindle-shaped, but very metabolic, withdrawing the

tatl ou es See gh ape ahs (xv) E. granulata.

No haematochrome. Chloroplasts spindle-shaped, spirally

arranged. Cytoplasm contains rows of granules staining

with neutral red, and arranged between the chloroplasts

(xvi) E. splendens.

Usually containg haematochrome granules, colouring the cell

red. Chloroplasts spindle shaped, but not arranged in any

marked spirals. No rows of granules staining with neutral

ted S650 eee ee (xvii) E. sanguinea.

(i) Euglena elongata Schewiakoff 1891 (Fig. 1a).

Cell fusiform, tapering to a point at the posterior end, truncate

at the anterior end, 65-70» long < 6-7» wide, Flagellum about

4 body length, but rarely seen, the cell progressing by squirming

140

Vol. XVI. (1958).

movements without appreciably changing its length and breadth.

Periplast apparently smooth. Cytoplasm containing a few small

scattered granules which stain with neutral red. Paramylum not

observed. Chloroplast, single, long band-shaped, stretching most

of the length of the cell. Pyrenoids none. Eye-spot small, anterior.

Collected from the lake in the Botanic Gardens, Singapore, and

from fish-ponds and padi swamps in Malacca.

Recorded also from New Zealand and U.S.A.

The Malayan specimens differ from the descriptions given by

Johnson (1944) and Lemmermann (1910) in the smaller eye-

spot, and the frequency with which the cell loses its flagellum, but

in all other respects they are similar, so that it seems justifiable to

place them under this species.

(ii) Euglena synchlora Gojdics 1953 (Fig. 1k).

Cell broadly fusiform, tapering gradually to a point at the poste-

rior end, round anteriorly, with the cytostome placed distinctly

laterally, 70u long « 22u wide. Flagellum about twice body-

length, very active, spiralling slowly even when the cell was sta-

tionary. Pellicle finely but distinctly striated in steep spirals.

Cytoplasmic granules staining with neutral red, very few. Para-

mylum granules ovoid, up to 6y long, abundant. Chloroplasts

elongate band-shaped, 12-154 long »< 3, wide, arranged end

on end to almost touch, so that they look like continuous bands,

five spirals being visible at a time. The chloroplasts are often diffi-

cult to distinguish, but show up in specimens freshly killed with

osmic acid vapour, and viewed with phase contrast. Pyrenoids

none. Eyespot, large, deep crimson. Cysts not seen. The cells were

very metabolic, contracting in length at the slightest stimulus.

Collected from the lake in the Botanic Gardens, Singapore, and

once from a drain in Malacca.

Reported from U.S.A. by Gojdics.

The characteristic shape and arrangement of the chloroplasts

make it certain that the Malayan specimens belong under this

species.

(iii) Euglena mainxi Déflandre 1928 (E. reticulata Mainx 1926)

(Fig. 1p).

Cell ovoid to elongate ovoid, tapering gradually to a point at

the posterior end, broadly rounded at the anterior end, 45, long

x 12-15. wide. Flagellum about body length, but often with-

drawn. Pellicle apparently smooth. Chloroplast reticulate, extend-

ing the full length just within the periplast, except at the very ends.

The ramifications were often so fine, that the whole chloroplast

- appeared uniform, but in favourable specimens the reticulations

141

Ce,

OSLO TERS

a ROTTS Rese

ASI

REQ OK

ne,s

— os

7 m7, 5

end tee Ses

— as

“A °

5 1

SF War

-4—¢—¢

SAREE

09.8.

Rin’:

FIGURE 1

Malayan species of Euglena Ehrenberg:—

a, E. elongata Schewiakoff; b, E. cyclopicola Gicklhorn; c, E. proxima

Dangeard; d, E. subehrenbergii Skuja; e, E. fusca (Klebs) Lemm.; f, E.

oxyuris Schmarda var. charkowiensis (Swirenko) Chu; g, E. splendens

Dangeard; h, E. ehrenbergii Klebs; i, E. caudata Hiibner; j, E. sanguinea

Ehrenberg; k, E. synchlora Gojdics; 1, E. tripteris (Dujardin) Klebs; m, E.

acus Ehrenberg: n, E. granulata (Klebs) Schmitz; 0, E. refringens Gojdics;

p, E. mainxi Déflandre; gq, EF. chlorophoenicea Schmarda.

Figures d & e are at a smaller magnification than the rest.

= =

Vol. XVI. (1958).

could be followed, while in others they showed up with staining,

or with the use of phase contrast..Pyrenoids two, sheathed, often

quite plainly visible in the chloroplast. Paramylum, in addition to

the sheaths of the pyrenoids, small grains scattered throughout

the cytoplasm. Eye-spot crimson, small, spherical. Cysts thin-

walled and often forming palmella-like masses.

Collected from the lake in the Botanic Gardens, Singapore, fish-

ponds, Malacca and Seremban, and padi swamps and drains,

Malacca, common. Originally described from Czechoslovakia.

The reticulate nature of the chloroplasts separates this off from

all other species except E. reticulata Sjostedt, marine and without

pyrenoids, and E. limosa Gard, larger and with many pyrenoids.

(iv) Euglena proxima Dangeard 1901 (Fig. Ic).

Cell fusiform, tapering to a point posteriorly, rounded at the

anterior end, 45-60» long “15-20, wide. Flagellum about two-

thirds body length. Periplast very finely striated. Cytoplasm gra-

nular, the granules appearing as spindles arranged spirally when

stained with neutral red. Paramylum ovoid granules of varying

size up to 6u in diameter, often so densely crowded as to make

other details difficult to distinguish. Chloroplasts parietal, lenti-

cular to oval, 3u in diameter, numerous. Eyespot bright crimson,

prominent. Cysts spherical, with a dense brownish wall imme-

diately around the cell, paler outside. The cells swim by slowly

rotating forward, but are so very metabolic that they easily round

up.

Collected from fish-ponds, swamps and drains, Malacca, Port

Dickson and Seremban. Cosmopolitan in distribution.

The slightly smaller size of the Malayan specimens is not suffi-

cient reason to separate them off as a variety.

(v) Euglena cyclopicola Gicklhorn 1925 (Fig. 1b).

Cell ovoid to globose when attached, cylindrical to ovoid in

the free-swimming state, 14—-20u long « 10-12, wide. Flagellum

about body length when the cell is swimming, but not visible in

the attached stage. Periplast very faintly striated, often appearing

smooth. Cytoplasm, when stained with neutral red, showing scat-

tered small, spherical granules. Paramylum granules small, dis-

coid. Chloroplasts parietal, 8-10, lenticular bodies 3—5,, diameter.

This species lives attached to small crustacea by the anterior end,

the mode of attachment being a short gelatinous stalk terminating

in a disc at the point of attachment. This species has been in-

cluded under the genus Colacium by Bourelly (1947) but it so

readily leaves the host, and in the frequent free-swimming state

143

Gardens Bulletin, S.

shows remarkable resemblance to various obscure, small forms of

Euglena which have been described from time to time.

Collected regularly from a fish-pond in Malacca.

Reported also from Czechoslovakia, Hungary and U.S.A.

(vi) Euglena chlorophoenicea Schmarda 1846 (Fig. 1q).

Cell cylindrical, tapering at the posterior end to a short point,

which may become blunt or even retracted, at the anterior end

tapering slightly while swimming, but becoming broadly rounded

when stationary, 100u long « 20 wide when swimming, 80y long

x 40 wide when static. Flagellum less than half body length.

Periplast prominently, spirally striate. Colour of the cells almost

brick-red due to large numbers of orange-red haematochrome

granules distributed more or less evenly throughout the cytoplasm.

Other cells in which the haematochrome had clumped in the cen-

tre, appeared dull green in colour, since the colour of the chloro-

plasts could then show through. Paramylum ovoid bodies up to 5

long, often densely packed. Chloroplasts parietal, ovoid, 4—5y

long, numerous. Eyespot large, consisting of a prominent clump

of crimson granules. Cyst spherical, with laminated gelatinous

walls.

Collected by Mr. H. M. Burkill (see acknowledgements) from

a small pond in Singapore. Not yet found elsewhere in Malaya.

Reported from Latvia, Germany and U.S.A.

The Malayan specimens were slightly smaller than those re-

ported elsewhere, but this difference is of no significance taxono-

mically.

(vii) Euglena acus Ehrenberg 1830 (Fig. 1m).

Cell cylindrical, tapering to a long point at the posterior end,

narrowed and truncate at the anterior end, 30-250» long xX

5—15,y wide. Flagellum not longer than quarter body length. Peri-

plast finely striate longitudinally. Paramylum consisting of rods of

varying length, some very long, and up to 10, in number. Chlo-

roplasts numerous small parietal discs. Pyrenoids none.

Found in practically all collections from fish-ponds, padi

swamps and drains, Malacca, Singapore, Port Dickson and Serem-

ban. Cosmopolitan in distribution.

Too well known a species to need much description. It is very

variable, and numerous varieties have been described, most of

which have been suppressed by Gojdics (1953).

E. acus var. hyalina Klebs, differing from the type only in the

complete loss of chloroplasts, but still retaining an eye-spot, has

occurred from time to time. It may be only a nutritional form.

144

Vol. XVI. (1958).

(viii) Euglena subehrenbergii Skuja 1948 (Fig. 1d).

Cells elongate, compressed, very truncate or slightly hollowed

at the posterior end, rounded at the anterior end, 170» long x

20 wide. Periplast firm with spiral, punctate striae which were

not always easy to see. Paramylum oval or hexagonal granules

(not spherical), slightly larger than the chloroplasts. Chloroplasts

parietal, discoid, 3—5u in diameter, numerous. Pyrenoids none.

No flagellum was seen, the cell progressing by squirming move-

ments without much change in length or breadth, metaboly being

slight.

Collected from fish-ponds and padi swamps (several times),

Malacca.

Reported originally from Sweden.

This species differs from E. ehrenbergii Klebs in the hexagonal

paramylum granules (not spherical as in that species) and the

markedly truncate posterior end. The Malayan specimens differ

from Skuja’s original description in that the striae are not nearly

so prominent, but this hardly seems an important difference, since

varying nutritional changes may affect the nature of the periplast.

(ix) Euglena ehrenbergii Klebs 1883 (Fig. 1h).

Cell flat to cylindrical, with sides nearly parallel, but sometimes

twisted, rounded at both ends, but slightly narrower at the ante-

rior end, 110-200 long «15-20, wide. Flagellum not seen, the

cell being very metabolic, contracting and expanding, and twist-

ing, sometimes to as short as half normal length. Cytoplasm very

finely granular. Periplast plainly, but delicately striate in spirals.

Paramylum spherical granules with depressed centres, about half

the size of the chloroplasts. Chloroplasts parietal, discoid, 3-4

in diameter, numerous. Pyrenoids none.

Collected from the lake in the Botanic Gardens, Singapore,

and from padi swamps and rain pools, Malacca, but not together

with the preceding species.

Reported also from Germany, Russia, Hungary, U.S.A. and

China.

Distinguished from the preceding species by the marked meta-

boly, the shape of the paramylum granules and the rounded poste-

rior end. Gojdics (1953) states that this species is larger, but

from the range of sizes quoted by her this is not particularly

evident.

(x) Euglena fusca (Klebs) Lemmermann 1910 (Fig. le).

Cells long, flat, with sides nearly parallel, but often twisted, end-

ing posteriorly in a sharp tail-piece, rounded at the anterior end,

145

Gardens Bulletin, S.

150-230 long * 15-20 wide. Periplast brownish in colour,

with spiral rows of hemispherical to squarish excrescences. Rows

often incomplete or even absent. Flagellum always very short.

Paramylum two large links, usually one anterior and one posterior

to the nucleus, and occasionally several smaller ones as well.

Chloroplasts parietal, ovoid, 3 long, numerous. Eyespot dark

red, prominent. Cysts, with the contents rounded up as in other

species were not seen, but frequently extended specimens were

found surrounded by a tightly-fitting, firm hyaline wall, from which

the cell later burst. The cells swam by a slow, forward, rotating

movement, and were weakly metabolic, able to contract to about

~ normal length.

Collected from the lake in the Botanic Gardens, Singapore, and

from fish-ponds, padi swamps, drains and pools in Malacca and

elsewhere. The species is probably cosmopolitan in distribution.

This species is so closely allied to E. spirogyra Ehrenberg that

it is not easy to separate them. Because of the brown colour, the

much greater size, and the fact that the chloroplasts seem to

touch, I have placed the Malayan specimens under E. fusca.

The variation in the size and shape of the excrescences, as des-

cribed by Lefévre (1934) and Gojdics (1953) I have been un-

able to demonstrate except in a few specimens only, even using

detached strings of the excrescences under phase contrast. Chu

(1946) claims to have demonstrated the identity of the two spe-

cies in culture, but he makes no mention of the nature of the

excrescences. Further work with pure cultures ought to clarify the

matter.

(xi) Euglena oxyuris Schmarda var charkowiensis (Swirenko)

Chu 1946 (Fig. 1f).

Cell elongated, somewhat flattened, with a dorsal keel-like fold

producing a shallow groove, frequently spirally twisted, terminating

in a prominent colourless tail-piece at the posterior end, rounded

at the posterior end, 150-160 long « 20-22, wide. Flagellum

about half body length. Periplast steeply, spirally, striate from

right to left. Paramylum two large links, one anterior and one

posterior to the nucleus. Chloroplasts parietal, discoid, 3 in dia-

meter, numerous. Pyrenoids none. Eye-spot red, about 3, in

diameter. Cyst not observed.

Collected from padi swamps and fish-ponds, aes

Reported from Russia, China and India.

The Malayan specimens all belong to this variety, which differs

from the type species in possessing a keel-like fold along the dorsal

surface.

146

Vol. XVI. (1958).

(xii) Euglena tripteris (Dujardin) Klebs 1883 (Fig. 1L).

Cell elongate, commonly twisted throughout its length, ending

in a long tail-piece at the posterior end, rounded at the anterior

end, distinctly triangular in cross-section, with the sides of the

triangle incurved, 100 long X 15, wide. Flagellum about half

body length. Periplast clearly striated, the striations following the

twists of the cell body. Cytoplasm containing scattered, small gra-

nules, which stain with neutral red. Paramylum two rods, one

anterior, one posterior to the nucleus. Chloroplasts parietal,

small, discoid, numerous. Pyrenoids none. Eye-spot red, promi-

nent. Swims rather rapidly in a rotating manner, but can bend

slightly laterally, and can twist to varying degrees about its longi-

tudinal axis.

Collected from fish-ponds, padi swamps in a number of locali-

ties in Malaya. Cosmopolitan in distribution.

(xiii) Euglena refringens Gojdics 1953 (Fig. 1o).

Cell cylindrical or slightly flattened, often curved or even sig-

moid at times, pointed at the posterior end, rounded at the ante-

rior end, 70 long & 20, wide. Flagellum about a third body

length. Periplast faintly striated, with the striae running in a steep

spiral. Cytoplasm containing a very few granules which stain with

neutral red. Paramylum granules very small rods, which when

numerous give the cell a distinctly refractive look, and may even

make it appear bumpy. Chloroplasts large, lenticular to ovoid,

7-10 in diameter, 2 thick, 10-15 in number, strictly speaking

parietal, but often becoming displaced with the movements of the

cell. Pyrenoids none. Eye-spot light crimson in colour. The cell

is very metabolic in movement.

Collected from a padi swamp in Malacca.

Reported from U.S.A.

Despite the fact that the Malayan specimens are larger in every

way than those described by Gojdics, the thick chloroplasts, and

the highly refractive paramylum granules makes it clear that they

belong here.

(xiv) Euglena caudata Hiibner 1886 (Fig. 1i).

Cell elongate fusiform, tapering strongly to the posterior end,

rounded anteriorly, 80u long x 10, wide. Periplast spirally stri-

ated, with very fine distinct lines. In addition to the sheaths on

the pyrenoids there are numerous small oval or rod-shaped gra-

nules scattered throughout the cytoplasm. Flagellum body length.

Chloroplasts 15-30 in number, parietal, discoid, with lobed mar-

gins and each bearing a doubly-sheathed pyrenoid. Eye-spot

147

Gardens Bulletin, S.

generally very prominent, crimson. The cell is very metabolic,

changing shape frequently.

Collected from fish-ponds and padi swamps, Malacca.

Reported from Germany, Russia and China.

Despite the relatively longer shape of the Malayan specimens,

the lobed chloroplasts, each with a doubly-sheathed pyrenoid, and

the small rod-shaped paramylum granules, make it seem better to

place them under this species. In any case, in a markedly metabolic

species, shape assumes a relatively lesser importance.

(xv) Euglena granulata (Klebs) Schmitz 1854 (Fig. in).

Cell fusiform, tapering to a point posteriorly, rounded at the

anterior end, 80 long X 20. wide. Flagellum about body length.

Periplast very distinctly striate. In the cytoplasm, just below the

periplast are prominent spherical granules, which readily stain

with neutral red; these are arranged in spirals, each two spirals

enclosing 3—6 spiral striae of the periplast. Chloroplasts parietal,

flat, elliptic 11-15 in number, each bearing a doubly sheathed

pyrenoid. In addition there are small ovoid granules of paramy-

lum, about 45 x 3, scattered in the cytoplasm. Eye-spot small,

yellowish-red in colour. Cyst not observed. The cell was very

metabolic, the posterior point often being withdrawn.

Collected from a small pond in Singapore (by Mr. H. M.

Burkill), where it was abundant, and more rarely in a drain in

Malacca.

Reported from Europe and U.S.A.

The prominent sub-pellicular granules make it certain that the

Malayan specimens belong here.

(xvi) Euglena splendens Dangeard 1901 (Fig. 1g).

Cell broadly fusiform, tapering posteriorly, but more frequently

rounded owing to the extreme metaboly of the cell, anterior end

rounded, 45-60, long & 15, wide. Flagellum about body length.

Periplast showing prominent spiral striae. Cytoplasm containing

spiral rows of granules, evident when stained with neutral red,

appearing spindle-shaped at the edges and in front views, lying

between the rows of chloroplasts; two rows of granules enclose

4—5 striae. Chloroplasts numerous, spindle-shaped, arranged in’

spiral rows, but appearing radial in optical section. Pyrenoids

three to six, lying within the chloroplasts, sheathed with paramy-

lum. Paramylum, in addition to the pyrenoid sheaths, flat discs up

to 3 long, scattered and rather scarce. Eye-spot crimson, granular.

The cell swims with constant rotation, but is very metabolic, so

that the tapered posterior end rounds up at the slightest stimulus,

even while swimming.

148

Vol. XVI. (1958).

Collected from the Botanic Gardens, Singapore, and from fish-

ponds and padi swamps Malacca. Cosmopolitan in distribution.

The spiral rows of spindle-shaped chloroplasts, the neutral red

granules, and the sheathed pyrenoids make it certain that the

Malayan specimens should be included under this species. The

commonly rounded shape of the posterior end is not of import-

ance, since the cell is very metabolic.

(xvii) Euglena sanguinea Ehrenberg 1830 (Fig. 1)).

Cells broadly fusiform, gradually tapering to a point at the

posterior end, rounded anteriorly, 60-100, long « 25-30, wide.

Flagellum about body length, but easily shed. Periplast conspicu-

ously and steeply striate. Colour brick red due to the abundance

of haematochrome granules dispersed throughout the cytoplasm;

when these granules clump in the centre the colour appears olive

green, due to the chloroplasts, and in some specimens collected

there was no haematochrome at all (except for the eye-spot).

Paramylum spherical or ovoid, up to 8 in diameter, visible in

cells with little or no haematochrome. Chloroplasts numerous long

spindles, radially arranged, but sometimes showing a tendency

to a spiral arrangement. Pyrenoids not visible in the living cell,

but showing up on staining in the centre of some of the chloro-

plasts. Eye-spot crimson, prominent. Cysts spherical, stalked in a

jelly-like layer. The cell is somewhat metabolic, often rounding up.

Collected from fish-ponds, Malacca, Seremban, Singapore, and

from padi fields, Malacca. Cosmopolitan in distribution.

This species is responsible for the brick red scums on Chinese

carp ponds, and on recently flooded padi fields. Its abundance

appears to be an indication of heavy application of organic

manures.

Lepocinclis Perty 1852

Cell solitary, free-swimming, spherical, ovoid, fusiform or ellip-

soidal, circular in cross-section. Periplast usually spirally, some-

times longitudinally striate. Cell rigid, not metabolic. Flagellum

one, usually long (2-3 times body length). Vacuolar system

typical for the family, with a cytostome and reservoir. Paramylum

usually two large rings, laterally placed to curve just inside the

periplast; in a few cases it consists of several discs or spheres, and

the genus then approaches the condition in Phacus. Chloroplasts

parietal, numerous, discoid, but sometimes so close together as to

become polygonal. Stigma usually well marked. Lives holophytic-

ally, but can live saprophytically.

149

Gardens Bulletin, S.

Key to species

. Striations running spirally to the right (in the direction of

the posterior end)"; 02s .3. 6+ wc en 0 + 5 pr Bi

. Striations longitudinal... 3... Js... «:: »« «5 10.

. Striations running to the left... ..........> .. 4) if;

. Cell much longer than broad (about three times), fusiform

to cylindrical, 23 long & 8u wide .

(xiii) L. acicularis.

. Cell relatively much shorter 2. ...../. /). . ee a

. Cell with a very short tail-piece, not much more than a wart-

like outgrowth, or none’atiall......... sag eee 4.

. Cell with a longer conspicuous tail-piece ............. pe

4. Cell large, broad, with many disc-shaped and ovoid paramy-

lum granules. 60 long * 40u wide .. (i) L. texta var.

mamillata.

. Cells much smaller (30 or less) with two lobed lateral

ring shaped paramylum granules .......’..\.sseaee =

. Cells with a distinct anterior neck-like outgrowth, sub-glo-

bose. 20u long *« 15-5u wide, neck 1-5—2y long

(x) L. ovum var. colliferum.

5. Cell without a distinct neck-like outgrowth ........... 6.

6. Cell elongate ovate, with sides nearly parallel, tail short and

blunt. 20-24, long & 15u wide .... (in) L. ovum var.

dimidio-minor.

. Cell ovate, wider than the preceding, with a distinct anterior

notched beak (not a neck). 20-24, long & 18, wide

(iv) L. ovum var. major.

. Cell rounder, with no beak, tail short and blunt. 19» long <

15g Wilkes tect eae (v) L. ovum var. deflandriana.

. Cell larger than preceding, posterior end wider than anterior

end. Very short beak at anterior end. 33,long & 15, wide.

(vi) L. ovum var. conica.

. Tail-piece well-marked but comparatively short (1/6 length)

cell broadly ovate. 28—30u long & 18-20, wide

(11) L. ovum forma ovum.

7. Tail-piece Jong 6. ii 3.0% ou. + sw ete & +100 oe gr 8.

. Striations verrucose. Cell oblong ovoid. 35 long & 17» wide

(xiii) L. ovum var. verrucosum.

150

Vol. XVI. (1958).

TEI IOS IeEBOOUNL o Sou eitkide PO eee o Sear... 9,

9. Cell oblong ellipsoidal with almost parallel sides. Tail long

and slender (4 body length); 42 long « 22 wide

(viii) L. ovum var. gracilicauda.

9. Cell ellipsoidal, narrowing towards both ends, with charac-

teristic swelling at base of tail. 50u long K 23, wide

(vii) L. ovum var. Butsciilii.

9. Cell broadly fusiform, almost rhombic, with smaller swelling

at the base of tail. 454 long « 20, wide

(xi) L. ovum var. angustata.

10. Striations ribbed; cell fusiform to elongate ellipsoidal; peri-

plast often yellowish 30u long & 10, wide

(xiv) L. steinii.

10. Striations not ribbed, fine; cell comparatively longer with a

narrow neck at anterior end; periplast hyaline. 33-66

los Sets wide: ia. 2a tae (xv) L. Marsonni.

11. Cell large, broadly ovate, rounded at posterior end; paramy-

lum numerous small discs; striations running to the left. 50u

lone ree: ces arin Sue od (xvi) L. salina.

Lepocinclis texta (Duj.) Lemm. em. Conrad var. mamillata (Da

Cunha) Conrad (Fig. 2a).

Cell broadly oval, tapering to a very short, blunt, tail at the

posterior end, rounded at the anterior end, 60u long « 40x wide.

Flagellum body length. Periplast spirally striate, the striae running

to the right. Paramylum one large central disc, and numerous,

scattered, small ovoid granules. Chloroplasts numerous, discoid,

parietal. Stigma large, crimson.

Collected from padi swamps, Malacca.

Reported from Brazil. .

This variety differs from the type species in that it tapers to

a very short blunt tail.

(11) Lepocinclis ovum (Ehrenberg) Lemm. f. ovum (Figs. 2c, d).

Cells broadly ovate, rounded at both ends, with a short blunt

tail at the posterior end, 28-30, tong & 18-20. wide, tail 5 long.

Flagellum about body length. Periplast spirally striate to the right,

the density of the striations being variable. Paramylum two large

rings, one on either side of the cell. Chloroplasts parietal, discoid,

numerous.

Collected from fish-ponds, swamps and ditches in various local-

ities in Malaya, common. Distribution cosmopolitan.

Gardens Bulletin, S.

(ili) L. ovum var. dimidio-minor Défl. (Figs. 2e, f, v).

Differing from the type in the smaller size, elongate ovate shape,

with very steep spiral striae. 20-24 long * 15, wide.

Collected from fish-ponds and padi swamps, Malacca. Common.

Cosmopolitan in distribution.

(iv) L. ovum var. major (Hiiber-Pestalozzi) Conrad (Fig. 2g, h).

Differing from the type in being narrower, and with a distinct

beak like outgrowth at the anterior end, and bearing a very short

tail at the posterior end. 20-24» long « 18, wide.

Collected from padi swamps, Malacca, occasional.

Reported from Germany and South Africa.

(v) L. ovum var. deflandriana Conrad (Fig. 21, j, k).

Smaller and rounder than the type, broadly rounded at the ante-

rior end, with a very small blunt tail at the posterior end. 19» long

x 15, wide.

Collected from fish-ponds, Malacca, occasional.

Reported from France and Indonesia.

(vi) L. ovum var. conica Allorge and Lefévre (Fig. 21).

Wider at the posterior end, bearing a very short tail there, and

a very short beak at the anterior end. 33u long «K 25, wide.

Collected from fish-ponds, Malacca, occasional.

Reported from France.

(vii) L. ovum var. butschlii Conrad (Fig. 2m).

Cell ellipsoidal, narrowing towards both ends, bearing a sharp

tail at the posterior end, with a characteristic swelling at the base.

5Ou long (with tail) « 23, wide, tail 15, long.

Collected from fish-ponds, Malacca, fairly common. Cosmo-

politan in distribution.

(vill) L. ovum var. gracilicauda Défl. (Figs. 2n, 0).

Cell oblong ellipsoidal with almost parallel sides, broadly

rounded at both ends, and bearing a long slender tail (about

+ body length) at the posterior end. 42 long & 22, wide, tail

12, long.

Collected from padi swamps, fish-ponds and rain pools, Malacca

and Singapore, fairly common. .

Reported from France.

(ix) L. ovum var. globula (Perty) Lemm. (Fig. 2p).

Differing from the type by being almost spherical and with a

very long flagellum (about 2—3 times body length) 20» long <

19u wide.

Collected from fish-ponds, Malacca, occasional. Cosmopolitan in

distribution.

152

FIGURE 2

Malayan species of Lepocinclis Perty:—

a, L. texta (Duj.) Lemm. emend. Conrad var. mamillata (Da Cunha)

Conrad; b, L. salina Fritsch; c-d, L. ovum (Ehr.) Lemm.; e-f, L. ovum

var. dimidio-minor Défl.; g-h, L. ovum var. major (Hiiber-Pest.) Conrad;

i-k, L. ovum var. deflandriana Conrad; 1, L. ovum var. conica Allorge &

Lefévre; m, L. ovum var. butschlii Conrad; n-o, L. ovum var. gracilicauda

Défi.; p, L. ovum var. globula (Perty) Lemm.; q, L. ovum var. colliferum

Prowse; r, L. ovum var. angustata (Défl.) Conrad; s, L. ovum var. verru-

cosum Prowse; t, L. acicularis Francé; u, L. steinii Lemm.; v, L. ovum

var. dimidio-minor Défl.; w-x, L. marsonni Lemm.; y-z, Colacium vesi-

culosum Ebr.

Gardens Bulletin, S.

(x) Lepocinclis ovum var. colliferum Prowse var. nov. (Fig. 2q).

Differing from L. ovum var. punctato-striato Lemm., which also

has an anterior neck-like outgrowth, by its smaller size, and by

the striae not being punctate.

Cell subglobose, broadly rounded at both ends, bearing a coni-

cal tail at the posterior end; at the anterior end there is a short

cylindrical neck-like outgrowth, from which the flagellum projects.

Cell 20n long & 15-5 wide, neck 1-5-2, long.

A L. ovo var. punctato-striato Lemm., quod etiam os colliferum

producit, haec varietas dimensione minore, striis non punctatis

valde recedit.

Cellula subglobosa, utrinque late rotundata, posteriore caudam

conoideam, brevissimam gerens, anteriore ore 1-5—2-0y alto cylin-

drico flagellifero praedita, flagello cellulae aequilongo, 20u longa,

15-54 lata. Membrana hyalina striis ad dextram spiraliter tortis

praedita. Paramyla duo annularia, lateralia. Stigma sanguineum,

conspicuum, anteriore dispositum. Chlorophora parietalia, discoi-

dea, plurima.

Habitat: Malacca in locis paludosis orizicultis (Prowse 182a).

(xi) L. ovum var. angustata (Défl.) Conrad (Fig. 2r).

Differing from the type in the broad, fusiform, almost rhombic

shape, and the slight swelling at the base of the tail spine. 45

long & 20 wide.

Collected from fish-ponds, Malacca, occasional.

Reported from Venezuela.

This variety comes very near to var. Butschlii (vide) and pos-

sibly should not be separated from it, although no intermediate

forms have been seen.

(xii) Lepocinclis ovum var. verrucosum Prowse var. nov. (Fig. 2s).

Differs from L. ovum f. ovum by the sharp pointed tailpiece,

and the spiral rows of small verrucae. From L. quadratum (Kuff. )

Conrad it differs by its smaller size, more ovoid shape and by the

complete absence of a neck at the anterior end. .

Cell oblong-ovoid, tapering posteriorly to a short tail, anterior

end truncate-rounded, 35 long & 17,» wide. Flagellum about

body length. Periplast bearing prominent spiral rows of small

warts or verrucae, about 17 rows being visible at a time. Para-

mylum forming two rings, one on each side of the cell. Chloro-

plasts parietal, numerous, discoid, small, 1-54 in diameter.

A L. ovo var. ovo haec varieta differt: posteriore cuneatim

caudata, membrana verrucis minutis seriatim ornata.

154

Vol. XVI. (1958).

A. L. quadrato (Kuff.) Conrad dimensione minore, forma ovoi-

diore, ore haud eminenti haec varietas facile recognoscenda.

Cellula oblongo-ovoidea, posteriore in caudam brevem angus-

tata, anteriore truncato-rotundata, flagello aequilongo praedita, 35

longa, 174 lata. Membrana verrucis minutis seriatim ornata; 17

series simul visibiles. Paramyla annularia, utrinque lateraliter sita.

Chlorophora parietalia, numerosa, discoidea.

Habitat: Malacca in locis oryzalibus paludosis (Prowse 239a).

(xiii) Lepocinclis acicularis Francé (Fig. 2t).

Cell fusiform, tapering to a short wedge-shaped tail at the pos-

terior end, notched at the anterior end, 23 long 8, wide. Fla-

gellum about 2 body length. Periplast hyaline, steeply, spirally

striate to the right. Paramylum two large rings one on each side.

Chloroplasts parietal, discoid, many.

Collected from a drainage channel, Malacca, occasional.

Reported from Hungary and the Netherlands.

(xiv) Lepocinclis steinii Lemm. (emend. Conrad) (Fig. 2u).

Cell fusiform to ellipsoidal, narrowing towards both ends, taper-

ing to a sharp conical tail at the posterior end, slightly truncate

and notched at the anterior end, 30u long & 10, wide, tail 9u

long. Periplast hyaline to yellowish, with well-marked, costate,

longitudinal striations. Flagellum about 14 body length. Chloro-

plasts parietal, discoid, numerous. Paramylum 2 rings, one on each

side.

Collected from fish-ponds, Malacca. Cosmopolitan in distribu-

tion.

(xv) Lepocinclis marsonni Lemm. emend. Conrad (Fig. 2w, x).

Cell elongate fusiform, bearing a long sharp tail at the posterior

end, and drawn out to a narrow straight neck at the anterior end,

33-60 long & 10-12 wide. Flagellum about ? body length.

Periplast hyaline, faintly striate longitudinally. Chloroplasts parie-

tal, discoid, numerous. Paramylum 2 large, laterally-placed rings.

Collected from padi swamps, and fish-ponds in several localities

in Malaya. Common. Cosmopolitan in distribution.

(xvi) Lepocinclis salina Fritsch (Fig. 2b).

Cell broadly ovate, flattened somewhat, broadly rounded at the

posterior end, slightly narrower at the anterior end, 50 long X

43 wide. Flagellum up to twice body length. Periplast spirally

Striate to the left. Paramylum numerous small discs, sometimes

with hollow centres. Chloroplasts parietal, numerous, discoid, but

often so close together that they become polygonal.

155

Gardens Bulletin, S.

Collected from padi swamps and fish-ponds, Malacca, common.

Cosmopolitan in distribution.

This species closely resembles L. texta (Duj.) Lemm. (vide)

but can easily be distinguished by the direction of the striations.

The paramylum granules are somewhat different as well, being

distinctly discoid, and not ovoid.

Colacium Ehrenberg 1833

Cells gregarious or colonial, living attached to small crustacea

or insects, often forming dendroid colonies. The individual cells

are attached by the anterior end, as can be seen by the position of

the stigma, the means of attachment being a gelatinous stalk,

which may be branched in dendroid fashion to hold several cells;

the base of the stalk is usually discoid. Flagellate stages only occur

during reproduction, and closely resemble Euglena, especially

small forms with discoid chloroplasts.

Colacium vesiculosum Ehr. (Figs. 2y, Z).

Cells fusiform to ovoid, solitary, or 2—4 together, 13-20, long

< 9-11, wide, attached by short gelatinous stalks to small crus-

tacea. Chloroplasts parietal, several, ovoid discs without pyrenoids.

Stigma usually quite clearly visible at the anterior end. Paramylum

not observed. Flagellate stage not seen.

Collected from a fish-pond in Malacca, where it occurs quite

frequently. Widespread in distribution.

I have included Euglena cyclopicola Gicklhorn (Colacium

cyclopicola (Gicklh.) Bourr.) under Euglena because it so readily

leaves its host without undergoing reproduction, and can reattach

itself again. The distinction is a slender one, but names are after

all artificial. It is of interest that the two species do not occur in

the same pond, but in different ones, where they are frequent.

Phacus Dujardin 1841

Cells solitary, free-swimming, not metabolic, very much flattened

dorsi-ventrally, generally ovoid or ellipsoidal, often with a marked —

tail-piece. Periplast firm, usually distinctly striate, longitudinally

or spirally, ribbed or with verrucae or hooks in some species.

Flagellum single, of varying length. Vacuolar system a cytostome

and reservoir. Stigma prominent, near the reservoir. Chloroplasts

parietal, numerous discoid. Pyrenoids absent. Paramylum discoid,

ring-shaped or lenticular, usually 1-2. Holophytic, but saprophytic

colourless forms have been reported.

156

Vol.

XVI. (1958).

Key to the species

Section Proterophacus: the periplast bears fine striations, not

ribs, and the cells are usually somewhat flattened, dorsi-

ventrally.

. Section Pleuraspis: the periplast is distinctly ribbed (costate)

and the cells more massive in cross-section.

. Section Acanthopeltis: the periplast bears longitudinal or

Spiral rows of wart-like or hooked excrescences.

A. Proterophacus

. Cells without any tail-piece, the posterior end being com-

pletely rounded, the cell outline being round or elliptical.

Generally flattened dorsi-ventrally ................ zB:

. Cells bearing at the posterior end a blunt, very short, wart-

like outgrowth instead of a tail-piece .............. dD:

. Cell posterior bearing a very short, but sharp, tail-piece,

Hardly set air from the main body 2.5. eee ae

. Cell bearing a well-marked tail-piece, straight or bent, more

or less short and pointed, clearly set off from the main

LS BE 2.0. G0. el cre Bae fe Soe ae Ty 6a 8.

. Cell bearing a long thin pointed tail-piece, markedly set off

from the main body, and about body length ........ 14.

Cells almost circular in outline, large: 00.5. es... os

. Cells elongate oval, about twice as long as broad, small .. 4.

. Cells flattened. Paramylum a single large disc, and some-

times many very small ones as well. 45-55 « 40-48.

(i) Ph. Stokesii.

. Cell much thicker. Paramylum several medium-sized sub-

spherical granules with hollow centres. 55 & 46u

(ii) Ph. Lefevrei.

. Cell not grooved throughout its length ............... ».

. Cell grooved throughout its length. Paramylum 1 small ring.

Re EE shan an ll a (iv) Ph. pusillum.

. Cell longitudinally striate, paramylum two oval discs placed

longitudinally 18 «K 8-5y........ (iii) Ph. Wettsteini.

. Cell spirally striate, paramylum two rings placed longitudi-

Haye 25 a hw Se (v) Ph. dangeardii.

. Cell twice as long as broad, cylindrical. Paramylum a single

large cylinder. 20 x 7-10p ........ (vi) Ph. granum.

157

Gardens Bulletin, S.

. Cell elongate ovoid, narrowing posteriorly. Paramylum a

large cylinder nearly filling the cell, with a smaller one

lying posterior to it. 22 & 8yu ... (vii) Ph. polytrophos.

. Cell ovate, broader than preceding, ventral face inrolled to

form a groove. Paramylum two laterally placed discs or

rings. 22+26 (iq =8ee eas ee (viii) Ph. oscillans.

. Cell almost quadrate ovate, paramylum two large curved

plates just within the chloroplasts. 125 x I1lpu

(ix) Ph. agilis.

. Cells flattish, elliptical or plano-convex in cross-section,

broadly ovoid in outline (x—xii) Ph. acuminatus & vars.

. Cells triangular in cross-section, ovoid in outline, twice as

long as broad. 22-28 * 11:5-13u (xiii) Ph. trifacialis.

. Cells in section flattened, elliptical, plano-convex, or con-

cavo-convex, not markedly twisted ....... 2.5. due o:

. Cell in section markedly triangular, with a dorsal ridge, out-

line broadly ovoid, tail prominent, deflected to one side.

80: Ao i Ca a. (xxiv) Ph. triqueier.

. Cell in section bearing three radiating wings spirally twisted,

in outline smoothly orbicular, with a sharply pointed tail.

22K EG CRY h Rae odes (xxv) PA. tricarinatus.

. Cell section of two unequally thickened lobes, twisted on

each other, cell outline ovate, tail pointed, sharply bent to

one side. 30-40 « 25-30u ...... (xvii) Ph. anomalus.

. Cell of two wings folded back towards dorsal side, outline

irregularly elongate ovate, markedly twisted. 45 «K 18y.

~ (xviii) Ph. raciborskii.

. Cell outline usually smooth)... .. 02s 15 tase 10.

. Cell outline usually notched .... .. . s,s eles ee £3.

. Cell outline broadly ovoid, bearing a distinct hook on the

left lateral edge. Tail stout, sharp-pointed, inclined. Para-

mylum 1 large disc. 102-110 « 65-70y

(xxiii) Ph. unidentatus.

. Body of cell distinctly longer than broad (about 13 times) 11.

10.

11.

Body of cell nearly as broad as long, or broader ...... V2.

Cell ovate, flattened, slightly twisted, tail straight, paramylum,

one large disc. 45 CMigie Ss ah i Grae (xiv) Ph. caudatus.

158

Vol.

iz.

LZ.

12.

£2

12.

13.

14.

iS.

tS.

XVI. (1958).

Cell ovoid, convexo-concave in section, tail bent to one side

or even hooked, paramylum two unlike concentric discs,

one lying on the other. 37-45 & 20-25»

(xxii) Ph. hamatus.

Cell somewhat rounded—triangular in outline, wider poste-

riorly, tail broad wedge-shaped, obliquely inclined, para-

mylum two large discs sometimes with hollow centres.

DT) WR ees 4 ee (xvi) Ph. angulatus.

Cell roundish in outline, wider to the posterior, tail short,

sharp, very abruptly turned to one side, paramylum two

unequal lateral discs. 23-28 & 22-25u

(xv) Ph. curvicauda.

Cell almost perfectly circular, tail relatively short, sharp,

turned to one side, paramylum one large central disc.

ee SOs, 2's ski KS (xix) Ph. circulatus.

Cell broadly ovate, flattened, tail slender, sharp-pointed,

turned slightly to one side, paramylum one large central

OS eI a ok 5s ices cg (xx) Ph. platalea.

Cell broadly ovoid to sub-orbicular, slightly asymmetrical,

tail stout, sharp-pointed, turned obliquely to one side,

paramylum one or two discs or rings. 60 &K 50u

(xxi) Ph. pleuronectes.

Cell rounded trapezoidal in outline, broadest to the posterior,

tail short, thick, inclined to one side, sides bearing several

notches, paramylum 1-2 large central discs. 33-30

(xxvi) Ph. Myersi.

Cell rounded trapezoidal in outline, broadest and very abrupt

at the posterior end, tail very stout, inclined obliquely

upwards in the dorsal direction, paramylum 1 large central

disc or ring. 50-55 « 35-37y ...... (xxvii) Ph. onyx.

Rare OGY UIE Le eee ea a a cele es 15:

BR ES ST A a 16.

Cell oval, tail long and curved, paramylum 1 central disc.

95-140 (with tail) «K 35-50u

(xxvill) Ph. longicauda var. rotunda.

Cell elliptical, tail long and distinctly bent to kinked, para-

mylum numerous small discs. 90 & 37, (with tail)

(xxxi) Ph. ranula.

159

Gardens Bulletin, S.

16. Cell twisted once through 180°, paramylum one large central

disc. . 80-85. 27=36a". 212). eee (xxix) Ph. tortus.

16. Cell twisted through 13 to 2 turns showing at least three

twists, paramylum 1 large central disc. 100 K 42y

(xxx) Ph. helikoides.

B. Pleuraspis

1. Cell ovoid, widely elliptical in cross-section, costae spiralling

to the right. 32°C) Tar te ee (xxxii) Ph. pyrum.

1. Cell oval, much flattened dorso-ventrally, costae spiralling to

the right. 37 & 17y .... (xxxiii) Ph. pseudonordstedtii.

C. Acanthopeltis

1, Cell. wall smooth,.26)\< 403 ae (xxxv) Ph. glaber.

1. Cell wall bearing longitudinal rows of blunt warts. 25-36 «

15235 a ge on el RR ee (xxxiv) Ph. suecicus.

1. Cell wall bearing longitudinal rows of down-curved hooks.

ASX SS a eee ee (xxxvi) Ph. horridus.

(i) Phacus stokesii Lemm. (Fig. 41).

Cell suborbicular, distinctly flattened, with a central groove

running the full length of the cell, at the posterior end bearing a

very small swelling, 45-55, long & 40—48, wide. Flagellum about

body length. Periplast hyaline, longitudinally striate. Paramylum

one large disc, and sometimes many very small granules as well.

Chloroplasts discoid, parietal, numerous. Stigma light crimson,

distinctly cup-shaped.

Collected from padi swamps, Malacca, fairly common.

Reported also from U.S.A., Poland, China and Java.

(11) Phacus lefevrei Bourelly (Fig. 4m).

Cell suborbicular, much less flattened than the previous species,

with the central groove only reaching ? length of the cell, and

ending posteriorly in a broad, shallow swelling, 55u long X

46u wide. Flagellum about body length. Periplast hyaline, longi-—

tudinally striate. Chloroplasts parietal, discoid, numerous. Para-

mylum several medium sized subspherical granules sometimes

with hollowed centres. Stigma large deep crimson, granular.

Collected from padi swamps, Malacca, occasional.

Reported also from Guadelope.

Close to Ph. stokesii but differing in being much less flattened,

the groove being shorter, and by the nature of the paramylum and

the stigma.

160

FIGURE 3

Malayan species of Phacus Dujardin:—

a, P. wettsteinii Drez.; b, P. acuminatus Stokes var. javana (Poch.)

Hiiber-Pest.; c, P. acuminatus var. discifera (Poch.) (Hiiber-Pest.; d-e, P.

acuminatus Stokes; ti ‘EP, curvicauda Swirenko; g-h, P. oscillans Klebs;

i, P. circulatus Poch. small form; j, 0, P. myersi Skv.: Lago a (PO ORPS

Poch.; m-n, x, P. anomalus Fritsch & Rich; p, P. circulatus Poch.;

a FP . platalea Drez.; r, P. trifacialis Prowse; s, P. angulatus Poch.;

u, y, P. raciborskii Drez.; v, P. hamatus Poch.; w, P. caudatus Hiibner;

z, P. agilis Skuja; c1-di, P . triqueter (Ehr.) Dujardin; e1, P. pleuronectes

(OF. M.) Duj.

FIGURE 4

Malayan species of Phacus Dujardin:—

a, P. longicauda (Ehr.) Duj. var. rotunda (Poch.) Hiiber-Pest; b-c, P.

tortus (Lemm.) Skv.; d, P. glaber (Défl.) Poch.; e, P. horridus Poch.;

f, P. helikoides Poch.; g-h, P. ranula Poch.; i-j, P. pseudonordstedtii Poch.;

k, P. pyrum (Ehr.) Stein; 1, P. stokessii Lemm.; m, P. lefevrei Bourrelly;

n-p, P. suecicus Lemm.; q-s, P. trifacialis Prowse.

162

Vol. XVI. (1958).

(iii) Phacus wettsteinii Drez. (Fig. 3a).

Cell elongate oval, rounded at both ends or slightly tapered at

the posterior end, apical groove reaching to the posterior end,

18 long < 8-5 wide. Flagellum about body length. Periplast

hyaline, longitudinally striate. Paramylum two large oval discs

placed longitudinally in the cell. Chloroplasts a few large discs,

parietal.

Collected from fish-ponds, Malacca, occasional.

Reported from Poland and Czechoslovakia.

(iv) Phacus pusillus Lemm. (Fig. Se).

Cell elongate oval, rounded at both ends, slightly flattened, with a

wide groove running the full length of the cell, slightly twisted, 19

long < 8, wide. Flagellum about 4 body length. Periplast finely

striate longitudinally. Paramylum 1-2 small rings. Chloroplasts

parietal, small, discoid, numerous.

Collected from padi swamps, Malacca, occasional.

Reported from Europe.

(v) Phacus dangeardii Lemm. (Fig. 5d).

Cell elongate oval, flattened, rounded at both ends, with no

longitudinal groove, 25 long & 14, wide. Flagellum about body

length. Periplast spirally striate. Paramylum two rings placed

longitudinally. Chloroplasts parietal, discoid, numerous.

Collected from padi swamps, Malacca.

Reported from France.

(vi) Phacus granum Drez. (Fig. 5g).

Cell nearly cylindrical, rounded anteriorly and ending at the

posterior end in a very short conical swelling; apical groove very

short. 20. long x 7—-10y wide. Flagellum about body length.

Periplast delicately striate longitudinally. Paramylum a single

large cylinder with rounded ends, placed usually longitudinally.

Collected from padi swamps, Malacca.

Reported from Germany, Poland and Bali.

(vii) Phacus polytrophos Pochmann (Fig. 5f).

Cell elongate ovoid, narrowing towards the posterior end, slightly

obliquely truncate at the anterior end, 22» long * 8» wide. Fla-

gellum about body length. Periplast delicately spirally striate to-

wards the left. Paramylum a large cylinder filling most of the cell

and a smaller one lying posterior to it. Chloroplasts parietal,

discoid, numerous.

Collected from fish-ponds, Malacca, occasional.

Reported from Russia, Poland, China and Germany.

163

taal

Te aa

LOTT)

Ss!

TT}

50M

FIGURE 5 :

Malayan species of Phacus Dujardin:—

a, P. unidentatus Prowse; b, P. tricarinatus Prowse; c, P. oscillans Klebs;

d, P. dangeardii Lemm.; e, P. pusillus Lemm.; f, P. polytrophos Poch.;

g, P. granum Drez. |

Strombomonas Défl. species in Malaya:—

h, S. schaunslandii (Lemm.) Défl.; i-j, S. gibberosa (Playf.) Défl.; k, S.

praeliaris (Palmer) Défl. var. brevicollaris Prowse; 1, S. girardiana (Playf.)

Défl.; m-p, S. australica (Play.) Défl.; q, S. fluviatilis (Lemm.) Défl.;

r, S. deflandrei (Roll) Défi.

Vol. XVI. (1958).

(viii) Phacus oscillans Klebs (Figs. 3g, h, 5c).

Cell ovate, broadly rounded at the anterior end, narrowed to-

wards the posterior, ending in a very short blunt projection;

ventral face inrolled forming a groove running obliquely the full

length of the cell; 22—26y long « 14-18, wide. Periplast spir-

ally striate. Flagellum about body length. Paramylum two discs or

rings. Chloroplasts parietal, small round discs, numerous. The cell

swims very rapidly rotating about the long axis.

Collected from fish-ponds in various localities in Malaya. Com-

mon. Cosmopolitan in distribution.

The Malayan specimens are relatively broader than described

elsewhere, but the species is a variable one, and it is doubtful if

this apparent difference is of any real significance.

(ix) Phacus agilis Skuja (Fig. 32).

Cells quadrate-ovate in outline, slightly narrower at the anterior

end and notched, and with a very short wart-like tail at the poste-

rior end, 12-54 long & 11, wide. Flagellum about body length.

Paramylum two large curved discs lying just within the chloro-

plasts. Chloroplasts parietal, two large discs lying just within the

periplast, lateral.

Collected from fish-ponds, padi swamps and drains in various

localities in Malaya.

Reported from Europe and Indonesia.

This is a very characteristic, if slightly variable species.

(x) Phacus acuminatus Stokes (Figs. 3d, e).

Cell broadly ovoid, bearing posteriorly a very short, broad,

wedge-shaped, but pointed tailpiece, narrowing towards the ante-

rior end, 22—26u long * 17-22 wide. Flagellum about body

length. Periplast longitudinally striate. Paramylum one or two

rings. Chloroplasts parietal, discoid, numerous.

Collected from fish-ponds, Malacca.

Reported from U.S.A., Russia and S. Africa.

(xi) Ph. acuminatus var. disciferus (Pochmann) Hiiber-Pestalozzi

(Fig. 3c).

Differs from the type in being rounder, slightly narrowed to-

wards the anterior, bearing on the posterior end a very short,

blunt wedge-shaped tail, and in the two large unequal imperforate

discs, the larger in the middle, the smaller lying excentrically. 22

long X 20, wide.

Collected from fish-ponds, Malacca, common.

Reported from France.

165

Gardens Bulletin, S

(xii) Ph. acuminatus var. javanus (Pochmann) Hiiber-Pestalozzi

(Fig. 3b).

Differing from the preceding by the apical groove reaching nearly

to the posterior end, and the two equal, symmetrically placed

paramylum rings. 234 long « 20y wide.

Collected from padi swamps, Malacca, common.

Reported from Java.

The above varieties have been separated by both Pochmann and

Hiiber-Pestalozzi, and I have done likewise, but it is uncertain to

what extent such a separation is justified.

(xiii) Phacus trifacialis Prowse sp. nov. (Figs. 3r, 4q-s).

Distinguished from the other similarly sized caudate species by

the distinct ridge running the full length of the dorsal surface, and

by the characteristic triangular cross-section.

Cell ovoid, gradually narrowing to a point posteriorly, rounded

anteriorly, 22—28u long & 11-5-13, wide; a distinct ridge runs

the full length of the dorsal face, so that in transverse section the

cell is like a broad shallow triangle, with the ventral surface

slightly hollowed; sometimes the cell is twisted throughout its

length. Flagellum about half body length. Paramylum granule a

single small ring, usually central. Chloroplasts parietal, small,

1-5 in diameter.

Inter omnes Phaci flagellati caudatique, sectione transversali

triangularis carinis inaequilbus haec species differt.

Cellula ambitu ovoidea, posteriore paulatim angustata acuta,

anteriore rotundata flagello dimidio longo praedita paulo torta vel

non, dorso per totam longitudinem conspicue uni-carinata sec-

tione transversali triangulata, ventre leviter concava, 22-28

longa, 11-5—13, lata.

Paramylum unicum, plerumque centrale, annulare. Chlorophora

parietalia, parva, 1-5, in diam.

Habitat: Malacca in locis oryzalibus paludosis et in piscinis

(Prowse 186a, 192a).

(xiv) Phacus caudatus Hiibner (Fig. 3w).

Cells ovate, flattened, but often slightly twisted, so that one side

seems more convex than the other, giving an asymmetrical appear-

ance; occasionally there may be a small notch on one side; poste-

rior end tapered to form a well-marked, sharp pointed tail-piece,

the anterior end broadly rounded, 45, long (including tail)

22 wide, tail 15, long. Flagellum about body length. Periplast

longitudinally striate. Paramylum one large ring, with sometimes a

smaller one as well. Chloroplasts parietal, discoid, numerous.

166

Vol. XVI. (1958).

Collected from padi swamps and fish-ponds, Malacca.

Reported from Europe, Asia and America.

This is a somewhat variable species, and the presence of a late-

ral notch in some specimens is no justification for separating them

off as a variety.

(xv) Phacus curvicauda Swirenko (Figs. 3f, k).

Cell roundish in outline, sometimes slightly wider than longer,

narrowing towards the anterior end, bearing at the posterior end a

short sharp tail turned markedly to one side. Apical groove vari-

able in length, ranging from short to nearly the full length of the

cell. 23—28u long 22-25, wide. Periplast longitudinally stri-

ate. Paramylum usually two unequal discs lying laterally. Flagel-

lum about body length. Chloroplasts parietal, discoid, numerous.

Collected from fish-ponds and padi swamps, Malacca. Cosmo-

politan in distribution, including Indonesia.

This is distinctly variable species, but always with the sharply

bent tail-piece. The length of the apical groove varies even within

the same population, so it cannot be considered a satisfactory

diagnostic character.

(xvi) Phacus angulatus Pochmann (Fig. 3s).

Cell somewhat rounded triangular in shape, broader to the pos-

terior end, with a short, wide, curved tail; narrower anteriorly; 37

long X 29 wide. Flagellum about body length. Periplast longi-

tudinally striate. Paramylum two large discs, sometimes slightly

hollowed in the centres. Chloroplasts parietal, rounded, numerous.

Collected from padi swamps, Malacca.

Reported also from Burma.

(xvil) Phacus anomalus Fritsch & Rich (Figs. 3m, n, x).

Cell body consisting of two unequal halves, one half being wider

than the other, twisted in opposite directions; in outline the cell

may appear ovate, or sometimes narrower at the anterior end,

and bearing posteriorly a sharp pointed tail markedly set off

from the main body of the cell, 30-40 « 25-30, wide, thick-

ness of narrower wing 9-10, that of the wider wing being

6—7u, the wings being unequal in thickness as well as in breadth.

Flagellum about body length. Striations of the periplast longitudi-

nal, following the bends of the two parts. Paramylum one or two

thickish discs. Chloroplasts parietal, discoid, numerous.

Common in the lake of the Botanic Gardens, Singapore, fish-

ponds, padi swamps, rain pools and drains, Malacca and else-

where.

Reported from S. Africa, Europe and Java.

167

Gardens Bulletin, S.

This is unmistakable species which cannot be confused with any

other, especially when seen in the living state.

(xviii) Phacus raciborskii Drez. (Figs. 3u, y).

Cell irregularly elongate-ovate in outline, very variable accord-

ing to the aspect; the cell body is made up of two wings folded

back towards the dorsal surface, and tapering posteriorly to a

sharp pointed, often curved tail, the whole cell being markedly

twisted throughout its length; from the ventral face the cell may

display a long twisted keel, and from the dorsal side a similar

hollow groove; 45 long x 18, wide. Flagellum about body

length. Striations of the periplast following the curves of the

wings longitudinally. Paramylum usually one, occasionally two

hollowed discs. Chloroplasts parietal, discoid, numerous.

Common in the lake of the Botanic Gardens, and fish-ponds,

padi swamps in various localities in Malaya.

Reported from France, Poland and Hungary.

This species is so characteristic, that anyone who has seen it,

especially in the living state, could not confuse it with any other

species.

(xix) Phacus circulatus Pochmann (Figs. 3p, i).

Cell almost perfectly round with a relatively short, sharp tail

turned to one side, 34 long (with tail) « 30, wide. Flagellum

about body length. Periplast longitudinally striate. Apical groove

reaching to about halfway. Paramylum one large central disc.

Chloroplasts parietal, discoid, numerous.

Collected from padi swamps, fish-ponds, Malacca.

Reported from Germany and Poland.

A much smaller form (fig. 31), 13 16, has occurred in the

same waters, but I have been chary of separating it as a variety,

as we know too little about nutritional and growth forms.

(xx) Phacus platalea Drez. (Fig. 3q).

Cell broadly ovate, flattened, with a prominent sharp tail in-

clined to one side; apical groove reaching up to halfway; 43.

long (with tail) 30, wide, tail 94 long. Flagellum about body

length. Periplast longitudinally striate. Paramylum one large

central disc. Chloroplasts parietal, discoid, numerous.

Collected from fish-ponds and padi swamps, Malacca.

Reported from Poland and France.

(xxi) Phacus pleuronectes (O.F.M.) Dujardin (Fig. 3e,).

Cell broadly ovoid to suborbicular in outline, slightly asymmet-

trical, produced posteriorly to form a sharp pointed tail turned

obliquely to one side, broadly rounded anteriorly, 60 long x

168

Vol. XVI. (1958).

50 wide. Flagellum body length or longer. Periplast longitudi-

nally striate. Paramylum one or two discs, often hollowed to

form rings. Chloroplasts parietal, discoid, numerous.

Collected from the lake in the Botanic Gardens, Singapore, fish-

ponds and padi swamps, Malacca. Cosmopolitan in distribution.

This is a common and rather variable species.

(xxii) Phacus hamatus Pochmann (Fig. 3v).

Cell ovoid, narrowing at both ends, but slightly wider in the

posterior half, bearing a sharp-pointed tail, usually somewhat

hooked and turned to one side, dorsal surface convex, ventral sur-

face concave, 37-45, long K 20-25, wide, tail 8-12, long. Fla-

gellum 13 times body length. Periplast longitudinally striate.

Paramylum usually two unequal concentric discs, one lying on top

of the other so that superficially they look like a ring. Chloroplasts

parietal, discoid, numerous.

Collected from fish-ponds, padi swamps and from the lake in

the Botanic Gardens, Singapore, common.

Reported also from Europe, Russia and Argentine.

(xxiii) Phacus unidentatus Prowse sp. nov. (Fig. 5a).

Allied to Ph. pleuronectes (O.F.M.) Duj. but differing in the

tooth on the left lateral edge.

Cells broadly ovoid in outline, bearing a distinct hook on the

left lateral edge towards the posterior end (when viewed from the

dorsal face), tapering towards the posterior end to form a stout,

sharp pointed tail, obliquely turned to the same side as the hook;

anterior end broadly rounded; flagellum about body length; cell

slightly hollowed ventrally, slightly convex dorsally, 102-110, long

(with tail) * 65-70» wide; tail 15-30, long. Longitudinal stria-

tions of the periplast prominent, with finer cross-connections.

Paramylum body a single large, central disc. Chloroplasts parietal,

discoid, 3-4, in diameter.

A Ph. pleuronectes (O.F.M.) Duj., cui affinissima, cellulis in

laevum unidentatis, flagellis multo longioribus haec species facile

distinguenda.

Cellula ambitu late ovoidea, posteriore caudiculata, anteriore

rotundato-truncata, flagello eae fere aequilongo praedita, supra

basin sinistro unidentata, ventre concaviuscula, dorso leviter con-

vexa, cum caudicula 102-110 longa, 65—70, lata; caudicula in

laevum oblique flexa, cuneata, apice acuta, 15-30, longa. Mem-

brana longitudialiter prominente striata, reticulationibus gracillimis

praedita. Paramylum magnum, unicum, centrale, disciforme. Chlo-

rophora parietalia, discoidea 3-4 in diam.

169

Gardens Bulletin, S.

Habitat: Malacca, in locis oryzalibus paludosis (Prowse 239b).

(xxiv) Phacus triqueter (Ehr.) Dujardin (Figs. 3c, , d, ).

Cells broadly ovoid, rarely with a constriction at one side, nar-

rowed asymmetrically posteriorly to form a prominent, sharp

pointed tail, slightly inclined to one side and deflected upwards

from the dorsal surface, broadly rounded at the anterior end, with

a prominent cleft marking the cytostome; on the dorsal surface a

high ridge runs the full length of the cell, so that in cross-section

it is triangular, slightly hollowed on the ventral surface; 80» long

(with tail) « 45, wide, tail 25 long. Flagellum about 12 body

length. Periplast longitudinally striate. Paramylum one or two large

discs or rings. Chloroplasts discoid, parietal, numerous.

Collected from the lake in the Botanic Gardens, Singapore, and

from fish-ponds and padi swamps, Malacca.

Reported from Europe, India, Java and Venezuela.

(xxv) Phacus tricarinatus Prowse sp. nov. (Fig 5b).

Allied to Ph. Warszewiczii Drez., but differing in that the stria-

tions run strictly longitudinally down the wings and the edges of

the wings, so that outline is quite smooth, and not crenulate in

that species, which has oblique striations.

Cell smoothly orbicular in outline, ending posteriorly in a

wedge-shaped tail; anterior end broadly rounded. Cell body con-

sisting of three equally-radiating longitudinal wings, slightly

twisted from left to right in the posterior direction; 224 long x

16, wide. Flagellum about body length. Striations of the periplast

running longitudinally, both along the ridges and in the hollows

between the wings. Paramylum granules usually one, sometimes

two central discs. Chloroplasts parietal, small, discoid, 1-5 in

diameter.

A Ph. Warszewiczii Drez. cui affinissima sed striationibus in

carina et ejusdem margine verticalissimis (non obliquis), cellulis

ambitu integerrimis (non crenulatis) haec species differt.

Cellula ambitu orbicularis, posteriore cuneata, anteriore late

rotundata, flagello aequilongo praedita, in tres carinas inter se fere

aequidistantes ad dextrum paulo reflexas vel tortas longitudinaliter

divisa, verticaliter striata, 224 longa, 16u lata. Granula paramy-

lonica plerumque singula, interdum bina, disciformia centraliter

sita. Chlorophora parietalia, parva, discoidea, 1-5» in diam.

Habitat: Malacca, in locis oryzalibus paludosis (Prowse 239c).

(xxvi) Phacus myersi Skvortzow (Figs. 3j, 0).

Cell rounded, trapezium shaped, broadest near the posterior

vend, bearing a short thick, obliquely inclined tail; cell narrower

170

Vol. XVI. (1958).

towards anterior end, sometimes rounded truncate; both sides dis-

tinctly notched; 33.long 30y wide, tail 4-5y long. Flagel-

lum body length. Periplast longitudinally striate. Paramylum 1-2

large central discs or rings. Chloroplasts parietal, discoid, numer-

ous.

Collected from padi swamps, Malacca.

Reported from S. China.

This is a slightly variable species, close to Ph. undulatus (Skv.)

Pochmann in its notched sides, but it can always be distinguished

by its much wider and trapezoidal shape.

(xxvii) Phacus onyx Pochmann (Fig. 31, a,, b,).

Cell round, trapezoidal, broadest and somewhat abrupt at the

posterior end, bearing a stout sharp tail, curved upwards from the

dorsal surface and slightly to one side, anterior end rounded; one

or both sides notched, 50—55y long « 35-37, wide, tail 14—

17 long. Flagellum over body length. Periplast longitudinally

striate. Paramylum one large central disc or ring, more rarely two

smaller ones. Chloroplasts parietal, discoid, numerous.

Collected from the lake of the Botanic Gardens, Singapore, and

from fish-ponds and padi swamps, Malacca.

Reported from Europe, U.S.A. and Indonesia.

This comes close to the preceding species, from which it differs

by the much larger size, and the very stout, outstanding, tailpiece.

(xxviii) Phacus longicauda (Ehr) Dujardin var. rotundus Poch-

mann) Hiiber-Pestalozzi (Fig. 4a).

Cell oval, slightly asymmetrical, ending posteriorly in a long

sharp tail, more or less curved; anterior end rounded; 35—140u

long (with tail) & 35-50, wide, tail 40-60, long. Flagellum about

body length. Periplast longitudinally striate. Paramylum one large

central disc. Chloroplasts parietal, discoid, numerous.

Collected from the lake in the Botanic Gardens, and from fish-

ponds and padi swamps, Malacca, common. Cosmopolitan in dis-

tribution, including Java.

Although a number of varieties of Ph. longicauda have been

described, this is the only form which has so far been reported

ae Malaya, oval in shape, and with the tail about as long as the

ody.

(xxix) Phacus tortus (Lemm.) Skvortzow (Figs. 4b, c).

Close to Ph. longicauda but twisted about the longitudinal axis

once through 180°, so that the outline may vary from ovate-fusi-

form to almost rectangular, with the anterior end appearing very

truncate; bearing posteriorly a long sharp tail; 80-85, long (with

171

Gardens Bulletin, S.

tail) & 27-36, wide, tail 30-33, long. Flagellum body length.

Periplast longitudinally striate, the striations following the twists.

Paramylum one large central disc. Chloroplasts, parietal, discoid,

numerous.

Collected from padi swamps, where it is very common, and

from fish-ponds, Malacca.

Reported also from Europe, Asia, Java and S. Africa.

(xxx) Phacus helikoides Pochmann (Fig. 4f).

Close to the preceding, but the cell body is twisted through 12 to

2 complete turns, so that the shape is usually broadly fusiform, with

usually 3 twists showing; bearing a long sharp tail at the poste-

rior end; 95—100» long (with tail) « 42, wide, tail 25» long. Fla-

gellum about 14 body length. Periplast striations longitudinal, but

following the twists. Paramylum one large central disc. Chloro-

plasts, parietal, discoid, numerous. |

Collected from padi swamps, Malacca, where it is common, and

less frequently from fish-ponds.

Reported from Europe, Asia, including Java, and N. & S.

America.

(xxxi) Phacus ranula Pochmann (Figs. 4g, h).

Closely related to Ph. longicauda but having numerous small

paramylum granules, and with a distinctly kinked tailpiece. Cell

elliptical in outline, flattened, slightly twisted, tapering to a long

sharp, distinctly kinked tail posteriorly, rounded anteriorly; 90u

long (including tail) & 37,» wide, tail 37 long. Flagellum about

body length. Periplast longitudinally striate. Paramylum several

small discs. Chloroplasts parietal, discoid, numerous.

Collected from padi swamps, Malacca, occasional.

Reported from Indochina and Java.

Section Pleuraspis

(xxxil) Phacus pyrum (Ehr.) Stein (Fig. 4k).

Cell ovoid, gradually narrowed posteriorly to form a long

straight pointed tail, rounded, or slightly narrowed at the anterior

end with an apical notch, 32» long (with tail) « 14, wide, tail

10 long. Flagellum about 14 body length. Periplast with marked

ribs running spirally to the right. Paramylum two large curved

plates placed just within the periplast, one on each side. Chloro-

plasts, discoid, numerous and small.

Common in fish-ponds and padi swamps, Malacca. Widespread

in tropical freshwaters.

172

Vol. XVI. (1958).

(xxxiii) Phacus pseudonordstedtii Pochmann (Figs. 4i, j).

Cell oval, flattened dorsi-ventrally, tapering posteriorly into a

long sharp tail, rounded to almost truncate at the anterior end,

37 long xX 17» wide, tail 11 long. Flagellum 14 times body

length. Periplast with prominent spiral ribs running to the right.

Paramylum two large lateral curved plates just within the periplast.

Chloroplasts numerous small discs.

Collected from padi swamps and a few fish ponds, fairly com-

mon.

Reported also from Europe, and Indonesia.

This species resembles Ph. nordstedtii Lemm. but it is less

rounded and much flatter.

Section Acanthopeltis

(xxxiv) Phacus suecicus Lemm. (Figs. 4n, 0, p).

Cell broadly ovoid to suborbicular, terminating in a stout,

sharp-pointed tail at the posterior end, truncate or slightly retuse

at the anterior end, with a central papilla through which the fla-

gellum passes; 25-36 long x 15-23 wide. Flagellum about

body length. Periplast bearing longitudinal rows of small wart-

like excrescences, or verrucae, the number of rows being variable.

Paramylum two large lateral curved discs lying just within the

periplast. Chloroplasts numerous small circular discs.

Collected from padi swamps and fish-ponds, fairly common.

Reported from Europe, Australia and Venezuela.

(xxxv) Phacus glaber (Défi.) Pochmann (Fig. 4d).

Closely related to the preceding species, but quite smooth. Cell

suborbicular with a short tail posteriorly, and a short papilla at the

anterior end, 26 long « 19, wide. Flagellum about body length.

Paramylum two large, laterally-placed curved discs just within the

periplast. Chloroplasts numerous small discs.

Fairly common in the lake of the Botanic Gardens, Singapore,

and in padi swamps and fish-ponds, Malacca.

Reported also from Germany, but probably much more wide-

spread.

(xxxvi) Phacus horridus Pochmann (Fig. 4e).

Very close to Ph. suecicus Lemm. but with distinct hooked

excrescences.

Cell broadly ovoid to subglobose, terminating posteriorly in a

sharp, slightly curved tail, truncate anteriorly with a central pa-

173

Gardens Bulletin, S.

pilla; 484 long * 33, wide. Flagellum about body length. Peri-

plast bearing longitudinal rows of small hooks which point poste-

riorly. Paramylum two large laterally-placed curved discs, just

within the periplast. Chloroplasts numerous small discs.

Fairly common in padi swamps and fish-ponds, Malacca.

Reported also from Australia and France.

Forms of several of the pigmented species of Phacus, but com-

pletely devoid of any chloroplasts, have occurred from time to

time, and they have usually been quite actively swimming, so that

they can hardly be regarded as senescent forms. On the other

hand they are so obviously related to the pigmented forms, that

there seems little justification in separating them off in the genus

Hyalophacus. In any case we know very little about the nutritional

requirements, and what causes the loss of chloroplasts.

Trachelomonas Ehrenberg 1833 emend. Déflandre 1930.

Cells solitary, free-swimming, somewhat metabolic, but enclosed

in a firm envelope or lorica, with a distinct opening or porus

through which the flagellum passes. The envelope may be spheri-

cal, oval, cylindrical or fusiform, sometimes with a distinct neck

markedly set off from the rest of the body at the porus. Colour

ranges from colourless to dark brown, and it may be punctate,

scrobiculate, verrucose or ornamented with spines, or quite smooth.

Flagellum one, usually long, projecting through the porus and

neck. Vacuolar system typical of family. Paramylum granules oval

or absent. Chloroplasts two to many, discoid, parietal. Pyrenoids

present or absent according to species. Lives holophytically, and

sometimes saprophytically.

Key to species

1. Lorica without. spines 952.1405... 6s. <2 cee oF

1. Lorica bearing spines... 5.45.5... '» oo hee 10.

2. Lorica spherical in outline .. i... ...... 7. 077 =

2. Lorica flattened, wider than long, smooth; porus with or

withont thickening 7 2305 3. (ii) Tr. curta.

2. Lorica oval to ellipsoidal .......:...s. sue 5.

2. Lorica cylindrical, with parallel sides ................ 8.

3. Lorica smooth . . «4. . 4 2-Sssi ob onsen d a ghee 4.

3. Lorica bearing irregular transverse thickenings, rugulose

(ii) Tr. rugulosa.

Lorica perfectly spherical, without any distinct neck, but

sometimes with a thickening round the porus

(i) Tr. volvocina var. minuta.

174

7“)

Vol.

a2.

12.

XVI. (1958).

Lorica only nearly spherical with a distinct inclined neck

(xxiv) JT. similis var. hyalina.

. Lorica without a distinct neck, only a mere thickening at the

Te 0 AE eS a re 6.

ee Wil we IEE ee ke ee Z-

Lorica ellipsoidal, equally rounded at both ends

(v) Tr. oblonga.

. Lorica cylindric-ellipsoidal, rounded at the posterior end,

somewhat flattened at the anterior end; porus bearing a

thickening, often with fine teeth

(vi) Tr. oblonga var. attenuata.

. Neck of lorica distinctly inclined, shape of lorica oval, finely

punctate, yellowish-brown ........ (xxii) Tr. similis.

. Neck of lorica straight, toothed; outline of lorica ellipsoidal,

membrane very rowph ..........-..-- (xxi) Tr. crebea.

. Neck of lorica divided into two lobes or lips; outline oval,

membrane rough ........ (xx) T. scabra var. labiata.

DeReen al WHMWIE GNIAR, CER Ls oe cs wee ee PD.

. Lorica bottle-shaped, with a distinct cylindrical neck, mem-

brane smooth ...... (xxii) Tr. volzii var. cylindracea.

. Lorica strictly cylindrical, with both ends rounded or the

anterior end flattened. Membrane punctate

(vii) JT. lacustris.

. Lorica shorter, tapering conically at the posterior end; mem-

Tisai? MM he SG So i. sk (vil) Tr. conica.

. Spimes short, conical, all the same size ............ 11.

. Spines stout, conical, much longer, all the same size, or

Le Ges 95 SSR ee ee 15.

. Spines variable in size, usually the posterior ones longer than

a eS i en 18.

a ee 12.

. Lorica with a distinct neck, toothed, oval in outline; spines

evenly dispersed over the body, short and uniform in

= Re ge Oy, ib, ear aie (xiv) Tr. mirabilis var. affinis.

. Lorica distinctly cylindrical, with sides parallel for most of

Me WE EE ce so eee See tee ee = a 13.

Lorica not so cylindrical, with sides not parallel ...... 14.

Lorica flattened, wider than long, with 3—4 rows of a few

oS ee (iv) Tr. lismorensis.

13.

14.

15;

16.

17;

18.

Gardens Bulletin, S.

Lorica cylindrical, with broadly rounded poles; spines

evenly distributed, short; porus often toothed

(xii) Tr. klebsii.

Lorica larger, cylindrical, sometimes flattened at the anterior

end, punctate, and set with larger, medium-sized spines,

evenly distributed; porus without teeth, but sometimes

thickened (25) i. pare eee tae oe see (xiii) Tr. zingeri.

Lorica ellipsoidal, thickly set with short, sharp, conical

spines, with or without a thickening round the porus

(ix) Tr. hispida.

Lorica ellipsoidal but much longer than the preceding,

thickly set with short spines; porus toothed

(x) Tr. hispida var. elongata.

Lorica cylindric-ellipsoidal, thickly set with short, conical

spines, porus without teeth ..... ye (xi) Tr. allia.

Spines evenly distributed over the whole surface of the lorica

16.

Spines mainly confined to the two ends .............. 17.

Lorica oval, with stout spines all over, porus without teeth

(xv) Tr. superba.

Lorica much longer, oblong ellipsoid in outline, with stout

spines all over; porus with fine teeth

(xvi) Tr. superba var. oblonga.

Lorica large, spines very long and stout (more than 10,)

POLUS PeOMEt eS en (xvii) Tr. megalacantha.

Lorica ellipsoidal, narrowing at both ends, with a distinct

neck bearing spines; one occasionally two rows of few

spines below neck; posterior end terminated in a single

spine surrounded by one row of similar ones

(xxv) Tr. hystrix var. paucispinosa.

Spines at posterior end straight, porus without teeth

(xix) Tr. dangeardina var. glabra.

Spines at posterior end curved (sometimes reduced), porus

toothed 7. Mace sen et ee crt ee (xviii) Tr. armata.

(i) Trachelomonas volvocina Ehr. var. minuta Fritsch (Fig. 6h).

Envelope perfectly spherical, smooth, 8-10, in diameter, vary-

ing in colour from colourless to clear yellowish-brown, or even

opaque brown. Porus sometimes surrounded by a thickened ring.

Flagellum 2-3 times body length. Chloroplasts two only, each

with a pyrenoid.

176

Vol. XVI. (1958).

Common in most stagnant waters in Malaya. Cosmopolitan.

The variety is distinguished only by its much smaller size and

may be a nutritional form.

(ii) Trachelomonas rugulosa Stein (Fig. 6m).

Envelope spherical, or nearly so, with thick, irregular sculptured

striations of the wall running transversely, dull yellow to reddish-

brown in colour; 17-20, in diameter. Porus with a slight thicken-

ing, but without teeth. Flagellum twice body length.

Collected from fish-ponds, Malacca. Probably widespread in

distribution, since reported from Europe, America and Africa.

(iii) Trachelomonas curta Da Cunha emend. Déflandre (Figs. 6b,

d).

Lorica spheroidal, compressed in the longitudinal direction, so

that it appears wider than long when viewed laterally, circular in

cross-section, smooth, 134 long & 15-16, wide. Porus some-

times with a thickened ring surrounding it. Flagellum twice body

length.

Collected from the lake in the Botanic Gardens, Singapore, fish-

ponds and padi swamps in Malacca, Seremban, and Port Dickson.

Probably widespread in distribution.

Reported from Europe, S. America and Australia.

(iv) Trachelomonas lismorensis Playfair (Figs. 60, p).

Envelope spheroidal, compressed in the longitudinal direction,

wider than long when viewed laterally, circular in cross-section,

hyaline to yellowish-brown, bearing 3—4 rows of a few short

conical spines, running transversely round the envelope; 6—9,

long <X 13, wide, spines 1-2 long, porus 2-2-5, in diameter.

Flagellum about twice body length.

Collected from fish-ponds and padi swamps, Malacca, occa-

sional.

Reported from Australia.

The Malayan forms are smaller than those from Australia, but

they obviously belong under this species.

(v) Trachelomonas oblonga Lemmermann (Fig. 6a).

Envelope ellipsoidal, slightly elongated, smooth, yellowish to

dark brown, rounded at both ends, 18 long & 14, wide; porus

with or without a ring-shaped thickening. Flagellum about twice

body length.

Fairly common in most bodies of standing water in Malaya.

Widespread in distribution.

177

és

(Y} ‘

Meat

rltatal) uy v\

Vv,

a 0

P ei

FIGURE 6

Malayan species of Trachelomonas Ehrenberg:—

a, T. oblonga Lemm.; b, d, T. curta Da Cunha emend. Défl.; c, 7.

oblonga var. attenuata Playf.; e-f, T. similis Stokes; g, T. similis var.

hyalina Skv.; h, T. volvocina Ehr. var. minuta Fritsch; i, T. conica Playf.;

j-k, T. volzii Lemm. var. cylindracea Playf.; 1, L. lacustris Drez.; m, L.

rugulosa Stein; n, T. crebea Kellicott; o-p, T. lismorensis Playf.; q, T.

superba Swir.; r, v, T. armata (Ehr.) Stein; s, 7. hispida (Perty) Stein;

t, T. dangeardiana Défl. var. glabra (Playf.) Défl.; u, 7. mirabilis Swir.

var. affinis Skv.; w, x, T. klebsii Défl.; y, T. allia Drez.; z, T. superba Swir.

var. oblonga Prowse; a1, T. hispida (Perty) Stein var. elongata Prowse;

bi-c1, T. hystrix Teiling var. paucispinosa Prowse; di, T. zingeri Roll,

€1, T. scahra Playf. var. labiata (Teiling) Hiiber-Pest.; f1, T. megalacantha

Da Cunha.

Vol. XVI. (1958).

(vi) Trachelomonas oblonga var. attenuata Playfair (Fig. 6c).

Differing from the type in being more cylindrical in shape, and

slightly flattened at the anterior end. The porus may be thickened

or may bear very fine teeth, 20u long & 14 wide. Less common

than the type.

Reported from Europe, Australia and Africa.

Trachelomonas oblonga is a somewhat variable species, and it

is not certain what justification there is to separate off so many

varieties, as has been done. The above variety is the only one

sufficiently distinct from the type, at least in the Malayan material,

to warrant separation.

(vii) Trachelomonas lacustris Drez. (non Tr. lacustris Skv.) (Fig

61).

Envelope cylindrical, finely punctate, clear yellowish-brown in

colour, broadly rounded at the posterior end, slightly flattened at

the anterior end, 18-20u long x 8—9u wide. Porus without

either teeth or a thickened ring. Flagellum about 14 times body

length.

Collected from the lake in the Botanic Gardens, Singapore, and

padi swamps, Malacca.

Reported from Europe, S. America and Australia.

The Malayan specimens are slightly smaller.

(viii) Trachelomonas conica Playfair (Fig. 6i).

_ Envelope cylindrical, smooth, clear yellowish-brown in colour,

in the anterior half with the walls strictly parallel, but sloping

conically at the posterior end, rounded at the apex of the cone, 20u

long < 14u wide. Porus without either thickening or teeth. Fla-

gellum about 12 body length.

Occasional in padi swamps, Malacca.

Reported from Australia and S. America.

(ix) Trachelomonas hispida (Perty) Stein emend. Défl. (Fig. 6s).

Envelope ellipsoidal, yellowish to reddish-brown, thickly covered

with short, sharp, conical spines 14-25, long « 12-20, wide.

Membrane sometimes finely punctate. Porus with or without a

ring-like thickening. Flagellum 13-2 times body length.

Common in most standing waters in Malaya, on the whole

smaller in size than reported elsewhere. Cosmopolitan in distri-

bution.

(x) Trachelomonas hispida var. elongata Prowse var. nov. (Fig.

6a ).

Differing from the type in being very much longer and narrower,

more than twice as long as broad. Lorica elongate oblong, narrow-

ing at both ends, and with the sides almost parallel in the middle

179

Gardens Bulletin, S.

half, thickly set with short spines, porus surrounded by a ring

of spines of the same length; 324 long x 13, wide without spines,

36u long 16, wide with spines, porus 3-5 wide.

A forma typica lorica valde longiore augustioreque, quam lati-

tudo ipsa plus duplo longiore haec varietas differt. Lorica elon-

gato-oblonga, utrinque paulatim angustata, spinis brevibus armata,

in ore 3-5u lato spinis aequialtis marginata, sine spinis 32, longa,

13 lata, cum spinis 36 longa, 16, lata.

Habitat: Malacca, in piscinis (Prowse 193 a).

(xi) Trachelomonas allia Drez. emend. Défl. (Fig. 6y).

Envelope cylindric-ellipsoidal, with broad, equally rounded ends,

sides more or less parallel in the median half, reddish-brown in

colour, thickly set with short, sharp, conical spines; 45-48, long

x 30, wide. Porus without thickening or teeth. Flagellum about

body length.

Collected from padi swamps and fish-ponds, Malacca, occa-

sional.

Reported from Europe, S. America and Indonesia.

(xii) Trachelomonas klebsii Déflandre (Figs. 6w, x).

Envelope distinctly cylindrical, with broadly rounded poles

which may be slightly flattened, light brown in colour, thickly

covered with short sharp conical spines; 25-30, long & 15—-16u

wide, porus 4-5, in diameter set with small teeth. Flagellum 14

times body length.

Collected from padi swamps, Malacca, occasional.

Reported from Europe, Venezuela, Java.

(xii) Trachelomonas zingeri Roll (Fig. 6d,).

Distinctly larger than the preceding. Envelope cylindrical,

rounded posteriorly, often slightly flattened at the anterior end,

55-60 long « 23-25, wide, brownish in colour, and thickly set

with medium-sized, sharp, conical spines. Porus sometimes with

a thickened ring. Flagellum about 2 body length.

Collected from padi swamps and one fish-pond, Malacca.

Reported from Russia.

Superficially the Malayan specimens look like Tr. australica

(Playf.) Défl. var. rectangularis Défl. bat the spines are distinctly

conical and sharp-pointed. For that reason it seems better to in-

clude them under this species.

(xiv) Trachelomonas mirabilis Swir. var. affinis Skv. (Fig. 6u).

Envelope ellipsoidal, rounded at the posterior end, and bearing

a cylindrical neck, distinctly toothed, at the anterior end; colour

brown; the whole body of the envelope is thickly covered with

180

Vol. XVI. (1958).

short, sharp, conical spines of uniform size; 37y long & 20, wide.

Neck 4u high « 4 wide. Flagellum about body length.

Collected from padi swamps, Malacca.

Reported from Manchuria.

The type species, with the spines very much longer at both ends,

has not been observed in Malaya so far, and the present variety

shows very little variation in the Malayan material.

(xv) Trachelomonas superba Swir. emend. Déflandre (Fig. 6q).

Envelope ellipsoidal, broadly rounded at both ends, brown in

colour, finely punctate with long, stout, sharp-pointed conical

spines of uniform length evenly distributed over the surface;

27-35 long & 22-30 wide (with spines), spines 3-5, long.

Porus without teeth. Flagellum about body length.

Common in most standing waters in Malaya. Cosmopolitan.

The Malayan specimens are slightly smaller than usual.

(xvi) Trachelomonas superba var, oblonga Prowse var nov. (Fig.

6z).

Differing from the type in the much longer proportions, about

twice as long as broad.

Envelope almost oblong in shape, rounded at both ends, densely

covered with stout conical spines, and with a comparatively wide

mouth bearing a ring of short teeth; dimensions 34 long & 17

wide without spines, 40u long x 25 wide with spines, spines

3—4 long, diameter of porus 4u.

A forma typica dimensione longiore, quam latitudo ipsa fere

duplo longiore haec varietas sat distincta.

Lorica ambitu oblonga vel fere, utrinqgue rotundata, spinis

conoideis, rigidis 3-4 longis dense vestita, cum ore pro rata lato,

4 in diam., secus marginem breviter dentato; sine spinis 34,

longa, 17 lata, cum spinis 40, longa, 25, lata.

Habitat: Malacca, in Jocis oryzalibus paludosis (Prowse 241a).

(xvii) Trachelomonas megalacantha Da Cunha (Fig. 6f, ).

Envelope large, oval, rounded at both poles, dark brown in

colour, bearing very long, stout, conical spines, evenly distributed

over the surface; 66y long & 60, wide (with spines), spines, 12

long. Porus bearing fine teeth. Flagellum body length.

Collected from a padi swamp, Malacca.

Reported from Brazil.

The extremely large spines make this an unmistakable species.

(xviii) Trachelomonas armata (Ehr.) Stein (Figs. 6r, v).

Envelope ellipsoidal to ovoid, sometimes slightly wider at the

posterior end, yellowish brown in colour, bearing long sharp,

181

Gardens Bulletin, S.

curved spines at the posterior end, and much shorter straight ones

at the anterior end; the length of the spines is variable, and in

some forms is very much reduced; 33—40p long & 26-28, wide.

Porus always toothed. Flagellum about twice body length.

Collected from fish-ponds and padi swamps in various localities

in Malaya. Cosmopolitan in distribution.

This is a decidedly variable species, and numerous forms have

been described. In the Malayan material there is so much grada-

tion between the various forms, differing in length of spines and

in shape, that there seems little justification in retaining the varietal

names.

(xix) Trachelomonas dangeardiana Défi. var. glabra (Playf.) Défl.

(Fig. 6t).

Envelope ellipsoidal, rarely ovoid, yellowish-brown in colour,

bearing at the posterior end strong, sharp, conical spines, always

straight and diverging from each other; the remainder of the lorica

is perfectly smooth, and the porus is without teeth, although there

may sometimes be a ring-shaped thickening; 40 long «K 30,

wide, spines 5—6y long. Flagellum about twice body length.

Collected from padi swamps, Malacca.

Reported from Australia and Venezuela.

The variety glabra differs from the type in the complete absence

of spines from the anterior half, and is the only variety yet ob-

served in Malaya, the type not having turned up so far.

Trachelomonas dangeardiana and its varieties can always be

distinguished from Tr. armata (E.) Stein by the straight posterior

spines, and the absence of teeth from the porus.

(xx) Trachelomonas scabra Playf. var. labiata (Teiling) Hiiber-

Pestalozzi (Fig. 6e, 1).

Envelope ovoid, slightly narrowed at both ends, slightly com-

pressed dorsiventrally, brown in colour, with a very rough surface;

28u long & 17, wide. Neck divided into two very distinct lobes.

Flagellum twice body length.

Collected from padi swamp in Malacca.

Reported from Sweden.

This variety is unmistakable in its bilabiate condition.

(xx1) Trachelomonas crebea Kellicott emend. Défl. (Fig. 6m).

Envelope regularly ellipsoidal, irreguiarly punctate and decidedly

rough, golden to reddish-brown in colour; porus with a widish

cylindrical neck, irregularly toothed; 39-40, long & 25, wide,

neck 5, long & 6, wide. Flagellum about body length.

182

Vol. XVI. (1958).

Collected from a padi swamp, Malacca.

Reported from Europe, N. America, S. America.

The Malayan specimens are distinctly larger in size.

(xxii) Trachelomonas volzii Lemm. var. cylindracea Playf. (Figs.

6j, k).

Envelope distinctly cylindrical, sides parallel, rounded or almost

straight at the posterior end, conical at the anterior end, bearing

a straight cylindrical neck thickened at the base, completely

smooth, brown in colour; the whole is shaped like a bottle; 32-42

long X 16—17, wide, neck 5—7y long X 4u wide. Flagellum about

body length.

Collected from fish-ponds, padi swamps, drains, Malacca, Se-

remban, Singapore, Port Dickson.

Reported from Australia.

The characteristic bottle shape makes this variety very easy to

recognise.

(xxiii) Trachelomonas similis Stokes (Figs. 6e, f).

Envelope ellipsoidal, yellowish-brown in colour, regularly punc-

tate, rounded at both ends and bearing at the anterior end a

curved, inclined neck, often slightly swollen at the base, lightly

toothed; 27% long X 16 wide, neck 5y long. Flagellum about

body length.

Collected from the lake in the Botanic Gardens, Singapore, and

from fish-ponds and padi swamps, Malacca, common.

Reported from Europe, America, Asia and Java.

(xxiv) Tr. similis var. hyalina Skvortzow (Fig. 6g).

Differing from the type in being nearly spherical, and perfectly

smooth without punctae; 25, long & 22-5, wide, neck 3-4: long.

Flagellum about body length.

Collected from the lake in the Botanic Gardens, Singapore, and

from padi swamps, Malacca.

Reported from Burma.

Skvortzow describes this as being hyaline, brown, but the Mala-

yan specimens range from completely colourless to opaque brown.

On the other hand the envelope is entirely without punctae, and

the edge of the neck always quite smooth.

(xxv) Trachelomonas hystrix Teiling var. pi ane Prowse

var. nov. (Figs. 6b;, c;).

Differs from the type in the considerable reduction of the num-

ber of spines, particularly in the median part of the envelope.

Envelope yellowish, ellipsoidal, narrowing equally at both ends;

neck short, wide, ornamented by a ring of 5—6 stout spines, and

183

Gardens Bulletin, S.

about a fifth of the way behind the neck bearing another ring of

stout conical spines, while at the posterior end there is a single

stout spine surrounded by a ring of 3—4 others, the tail spine

usually being longer; occasionally 1—2 very short almost wart-like

spines are borne in the median region; 33 long x 16, wide,

neck 4—5u wide X 2-3, high. Flagellum about half body length.

Chloroplasts parietal, discoid, 4, in diameter.

A forma typica haec varietas spinis perpaucissimis praecipue in

loricae medio differt. =

Lorica luteola, ellipsoidea, anteriore abrupte constricta, in ore

5—6 spines, rigidis ornata, infra orem spinis consimilibus in unam

seriem dispositis, prope basin 3—4 spinis similiter dispositis, et imo

posteriore spina unica armata, in altera parte inermis vel interdum

medio 1—2 spinis brevissimis vel verrucoideus praedita, proto-

plastum fere complexa, 33 longa, 16, lata, in ore 4—5p lata, 2-3

alta. Flagellum dimidio longum. Chlorophora parietalia, discoidea,

4u in diam.

Habitat: Malacca in locis oryzalibus paludosis et stagnis pisca-

toriis (Prowse 186a).

Strombomonas Déflandre 1930

(Trachelomonas Ehrenberg pro parte)

Separated off from Trachelomonas on the basis that there is no

porus, as in that genus, and the neck is wider, and generally longer,

tapering gradually into the body of the lorica, and not set off

sharply. The wall is often variable in thickness, but is very rarely

punctate, and rarely ornamented by scrobiculations, perforations or

spines. The stigma is large, and the flagellum relatively shorter

than in Trachelomonas, rarely exceeding body length. There is also

a much greater tendency for the protoplast to occupy the whole

envelope.

Key to the species

1. Lorica nearly spherical in outline, with short neck and a sharp

poimted taila; sae (ii) Str. praeliaris var. brevicollaris.

1. Lorica oval, sides nearly parallel, neck short and wide, tail

short and. bhatt! ay au oes + inet lonis (ii) Str. deflandrei.

1. Lorica. fusiform... :4¢.0:2den so idle Sunk «0c 2.

1. Lorica almost cylindrical, sides straight or slightly concave,

and slightly narrower at the posterior end. Outline hexa-

gonal. Shortish neck and sharp tail. . . (i) Str. girardiana.

184 f

Vol. XVI. (1958).

ESS et or 3.

. Lorica long, fusiform, relatively narrow, neck toothed

(iv) Str. australica.

2. Lorica relatively short. Neck short, not toothed

(v) Str. fluviatilis.

3. Lorica smaller, 30. long, tapering rather sharply to neck and

SRN ae ee ES oak Sue ge i wm (vi) Str. schaunslandit.

3. Lorica larger, 70» or over, tapering less sharply and more

eomcally-to tie tail ne. nn ie et (vil) Str. gibberosa.

(i) Strombomonas girardiana (Playf.) Défl. (Fig. 51).

Envelope yellowish, rough, circular in cross-section, almost

hexagonal in outline, with sides straight or slightly concave, or

narrowing slightly in the posterior direction, sloping abruptly at

the posterior end to the sharp, conical, tail, and similarly at the

anterior end to the cylindrical neck; sometimes extra thickenings

on the wall increase the angular appearance of the outline; 40,

long < 23, wide, neck 6 long X 6» wide, tail 11, long. Flagel-

lum about body length.

Collected from padi swamps and fish-ponds, Malacca, occa-

sional.

Reported from Australia, Egypt and Venezuela.

(ii) Strombomonas praeliaris (Palmer) Défi. var. brevicollaris

Prowse var. nov. (Fig. 5k).

Differing from the type by its very short neck.

Envelope hyaline to pale yellow, nearly spherical, coarsely

roughened, and tapering sharply to a tail posteriorly; bearing a

low wide neck at the anterior end; 32 long (including neck and

tail) & 24» wide; neck 7-5 wide 2, high; tail 7-5 long. Fla-

gellum about body length. Protoplast filling most of the envelope.

Chloroplasts parietal, discoid, 3-4-5 in diameter.

A forma typica haec varietas ore breviori haud constricto differt.

Lorica hyalina vel lutescens, globosa, asperrima, apice collo 7:5u

lato, 2u alto praedita, basi in caudum 7-5, longam abrupte pro-

ducta, cum cauda colloque 32 longa, 24, lata. Flagellum loricae

aequilongum vel. fere. Chlorophora parietalia, discoidea, 3—4:5u

in diam.

Habitat: Malacca in locis oryzalibus paludosis (Prowse 240 a).

(iii) Strombomonas deflandrei (Roll) Défl. (Fig. 5r).

Envelope brown, rough, circular in cross-section, broadly oval

in outline with the sides sometimes nearly parallel, rounded at the

poles, bearing a short, conical, blunt or pointed tail at the poste-

185

Gardens Bulletin, S.

rior end, and a short, wide neck, often obliquely cut, at the ante-

rior end; 33y long X 22 wide, neck 3, high 8» wide. Flagel-

lum about 14 body length.

Collected from padi swamps and fish-ponds, Malacca, occa-

sional. |

Reported from Russia.

(iv) Strombomonas australica (Playf.) Défl. (Figs. 5m—p).

Envelope hyaline to clear yellow, rough to almost smooth, cir-

cular in cross-section, elongate fusiform in outline, tapering to a

long, sharp tail-piece (occasionally short and blunt) at the poste-

rior end, and terminating in a long cylindrical neck, usually

toothed, at the anterior end; 40-66u long « 17-20, wide, neck

8-12» long & 3—Syu wide, tail 8-13 long. Flagellum about 14

body length.

Collected from fish-ponds and padi swamps, Malacca.

Reported from Australia and China.

The Malayan specimens showed a wide degree of variation,

some forms coming near to forms of the succeeding species, Str.

fluviatilis (Lemm.) Défi.

(v) Strombomonas fluviatilis (Lemm.) Défl. (Fig. 5q).

Envelope hyaline to clear brown, rough, circular in cross-sec-

tion, ellipsoidal-fusiform in outline, narrowing posteriorly to a

short sharp tail, and anteriorly to a comparatively short, straight

neck, not toothed; 27-28, long * 12-13, wide, neck 3y long

3. wide, tail 3 long. Flagellum about body length.

Collected from fish-ponds and padi swamps, Malacca.

Reported from Europe, Asia, S. America and Java.

This may be distinguished from shorter forms of the preceding

species by its smaller size, and distinctly shorter neck and tail.

(vi) Strombomonas schaunslandii (Lemm.) Défl. (Fig. Sh).

Envelope hyaline to yellowish-brown, rough, circular in cross-

section, broadly rounded and slightly rhombic in the median part,

tapering fairly sharply to a pointed tail at the posterior end, and

bearing a relatively long, smooth, cylindrical neck, widening

slightly at the opening, but not toothed, at the anterior end; 32,

long & 22 wide, neck 124 long * 6, wide, tail 11» long. Fla-

gellum about body length.

Collected from fish-ponds and padi swamps, Malacca.

Reported from Siam, Venezuela, Manchuria, Australia and

Java.

186

Vol. XVI. (1958).

This may be distinguished from the next species by its smaller

size, less rhombic outline, relatively longer neck, and the sharp-

ness with which it tapers into the tail-piece.

(vii) Strombomonas gibberosa (Playf.) Défl. (Figs. i, j).

Envelope hyaline to light brown, smooth or rough, circular in

cross-section, distinctly broadly rhombic in outline, tapering to a

sharp tail posteriorly, and to a wide neck at the anterior end;

larger than the preceding, 70-75 long & 33-36 wide, neck 12-

16u long * 7-8. wide. Flagellum about body length.

Collected from fish-ponds, padi swamps, drains, Malacca.

Reported from Europe, Manchuria, Australia and Venezuela.

This and the previous species are evidently related, and should

intermediate forms turn up, it may be necessary to combine the

two species.

Colourless Eugleninae

Cells completely devoid of chloroplasts, but possessing a definite

cytostome and reservoir, and storing paramylum.

Key to the families

A. Possessing a stigma. Cells elongate fusiform, almost needie

MUNGRREN eee 6 ole, w Kern oS Cyclidiopsidaceae.

B. Completely without a stigma.

I. Cells without the special rod-shaped bodies or siphon

(“staborgan”) near the reservoir; more frequently

rounded in cross-section, usually free-swimming,

with 1-2 flagella, metabolic or not

Astasiaceae.

II. Cells bearing a specialised rod-shaped body or siphon

near the cytostome and reservoir. Generally showing

dorsiventral organisation, frequently crawling, some-

times swimming, with 1-2 flagella, metabolic or

non-metabolic, often holozoic in nutrition

Peranemaceae.

Cyclidiopsidaceae

Cells colourless, completely without plastids, very long fusiform

or needle shaped, rigid and not metabolic. Cytostome centrally

placed at the anterior end, well-marked, leading to an elongate

ellipsoidal reservoir. Stigma distinct, next to the reservoir. Nucleus

central, cylindrical to elongate oval. Paramylum long rods or

needles.

oO PRES TRE Ss ee. a's 2. Cycliodiopsis Korschikow.

Gardens Bulletin, S.

Cyclidiopsis Korschikow 1917.

Characters of the family.

One species recorded for Malaya ; .. 2.72 uae Cycl. acus.

Cyclidiopsis acus Korschikow (Fig. 7a).

Cell very long, fusiform to needle shaped, rigid, tapering poste-

riorly to a long thin tail-piece, truncate at the anterior end, with a

centrally placed cytostome leading to a long oval reservoir; 125y

long x 5, wide. Flagellum short, less than 4 body length. Peri-

plast apparently smooth. Eye-spot oval, crimson, next to the reser-

voir, distinct. Cytoplasm hyaline, with 5—8 long thin cylindrical

paramylum granules. Nucleus central, long cylindrical.

Collected from the Malacca river in a plankton net haul, and

once from a padi swamp.

Reported from Russia, Sweden, Germany and Australia.

There is a good deal of confusion about this species, and many

authors would include it under Euglena acus as a colourless form.

Hyaline forms of the latter have occurred in Malayan collections,

and I have been fortunate to be able to compare the two. The

forms of Euglena acus differ from the type only in the absence of

the chloroplasts. The cytostome is slightly one-sided, and the nuc-

leus is shorter and rounder than in Cyclidiopsis, while the reservoir

of the latter is much more prominent. No pigmented forms of

Euglena acus having quite the same characteristics as Cyclidopsis

acus have occurred in the Malayan material, so for the time being

it seems better to retain the separate generic name. Further re-

search may of course reveal that such a separation is not justified.

Astasiaceae

Cells solitary, free-swimming, metabolic or rigid, fusiform, cy-

lindrical, ovoid or ellipsoidal, completely without chloroplasts;

stigma absent (except in Khawkinea Jahn & McKibben, which

has not been reported from Malaya); flagella 1-2. Vacuolar sys-

tem a typical cytostome and reservoir without any rod-shaped or

tubular siphon. Storage material paramylum and sometimes fat.

Nutrition generally saprophytic, occasionally holozoic.

Key to the genera

1. Cell with one flagellum... 3.0.5.6. »4 Os) <. 020 2:

1. Cell with two unequal flagella, distinctly metabolic

Distigma.

2. Cell distinctly metabolic, oval to fusiform, roundish in cross-

SECLION «po inca sess fu pieeel Bollea alee Astasia.

2. Cell more rigid, distinctly curved to one side .. Menoidium.

188

ee,

50p

FIGURE 7

Colourless Eugleninae in Malaya:—

a, Cyclidiopsis acus Korschikow; b, Astasia variabilis Skv.; c, Menoidium

obtusum E.G. Pringsheim; d, Menoidium pellucidum Perty; e-f, Distigma

proteus Ehr.; g, Menoidium obtusum E.G. Pringsheim, very small form;

h, Peranema cuneatum Playf.; i, Distigma curvatum E.G. Pringsheim; j,

Peranema curvicauda Skuja; k, Petalomonas mediocanellata Stein; 1, Pera-

nema kupfferi Skuja; m, Petalomonas asymmetrica Shawhan & Jahn; n-o,

Petalomonas abcissa (Duj.) Stein var. pellucida Skuja; p-q, Petalomonas

platyrrhyncha Skuja; r-u, Petalomonas heptaptera Prowse.

Gardens Bulletin, S.

Astasia Dujardin 1841 |

Cell distinctly metabolic, changing shape quite rapidly, but

usually cylindrical or fusiform when swimming. Eye-spot absent.

Flagellum 1, forking typically where it enters the reservoir. Va-

cuolar system as in Euglena with a typical cytostome and reservoir.

Paramylum granules generally round or ovoid. The species of this

genus come very close to being colourless forms of Euglena, and

further research may result in the transfer of many of them to that

genus.

One species reported for Malaya ............ As. variabilis.

Astasia variabilis Skvortzow (Fig. 7b).

Cell small, oval, rounded at the posterior end, rounded or

slightly truncate at the anterior end, but drawn out a little at the

anterior end when swimming; distinctly metabolic when not swim-

ming; 18—20u long « 7» wide. Flagellum about body length. Pe-

riplast apparently smooth. Cytoplasm clear, containing 10-15 oval

paramylum granules.

Collected from a drainage channel in Malacca, occasional, but

more frequent in putrefaction cultures from the same source.

Reported from China.

I have tentatively placed the Malayan specimens under this spe-

cies, since they seem to agree most closely with Skvortzow’s des-

cription, although the latter is lacking in one or two details.

Menoidium Perty 1852

Cells single, free-swimming, rigid or hardly metabolic, more or

less flattened dorsiventrally, and distinctly curved to one side late-

rally, rounded or slightly pointed at the posterior end, truncate or

oblique at the anterior end. Flagellum 1, typical for Euglena.

Periplast longitudinally striate. Paramylum granules rod or ring-

shaped, often with 1—2 large ones.

Key to the species

1. Cell narrowed to a neck at the anterior end, with two pointed

lip-like projections, slightly narrowed but rounded at the

pOSstenpr eH EOS oth ae (i) M. pellucidum.

1. Cell only very slightly narrowed at the anterior end, truncate,

broadly rounded at the posterior end .. (ii) M. obtusum.

(i) Menoidium pellucidum Perty (Fig. 7d).

Cell very flat, distinctly curved to one side, convex side more

curved than the concave side, narrowed to a neck-like end, often |

190

Vol. XVI. (1958).

terminating obliquely in two pointed lips; at the posterior end

slightly narrowed but rounded; 70-80 long x 10-12, wide.

Flagellum up to about 4 body length. Periplast longitudinally

striate, but these are not always visible. Cytoplasm faintly granular

to hyaline, containing 1—2 large cylinders and several small rod-

shaped paramylum granules.

Collected from fish-ponds, Malacca. Widespread in distribution.

(ii) Menoidium obtusum E. G. Pringsheim (Figs. 7b, g).

Cell only somewhat flattened, curved to one side, broadly

rounded at the posterior end, very slightly narrowed at the anterior

end, rather truncate, without any pointed lips; 40-45, long x

10u wide. Flagellum about 4 body length. Cytoplasm slightly gra-

nular to hyaline; paramylum two large rings and several smaller

scattered rods. Striation of the periplast not visible.

Collected from fish-ponds, Malacca.

Reported from Central Europe and Brazil.

The very small form shown in fig. 7g, 10x long 5, wide, is so

obviously related to this species that it is ata only a juvenile

form. I have therefore included it here.

Distigma Ehrenberg 1838

Cell usually distinctly metabolic, with two flagella, a long swim-

ming flagellum and a shorter trailing one. Periplast spirally striate.

Vacuolar system similar to that in Astasia without any rod-shaped

bodies or siphon, although some specimens show a superficial

resemblance to species of Peranema. Paramylum usually medium

to small oval granules, often densely packed.

Key to the species

1. Cell, long, fusiform, tapering almost to a point at the posterior

end, narrowed at the anterior end, with a distinct notch

48-65. long K 8-10 wide ......... (i) D. proteus.

1. Cell smaller, shorter, varying from cylindrical to pear-shaped,

rounded anteriorly, slightly narrower, or broadly rounded

at the posterior end, metabolic. 12—16n long & 8p wide

(ii) D. curvatum f. minor.

(i) Distigma proteus Ehrenberg em. Pringsheim (Figs. 7e, f).

Cell elongate-fusiform, tapering almost to a point at the pos-

terior end, broadest nearer the anterior end, but tapering towards

the cytostome, where there is a distinct notch; 48-65, long

191

Gardens Bulletin, S.

8-10, wide. Longer flagellum about body length, shorter flagel-

lum about 4 body length. Distinctly metabolic, even when swim-

ming. Periplast spirally striate, but the cell contents are often so

dense as to render the striations practically invisible. Paramylum

abundant small oval rods or cylinders, nearly filling the cell.

Collected from padi swamps, Malacca. Widespread in distribu-

tion.

(ii) Distigma curvatum E. G. Pringsheim fa. minor Pringsheim

(Fig. 7i).

Cell almost cylindrical to. pear-shaped, rounded anteriorly,

rounded or slightly narrowed at the posterior end, more or less

slightly curved, very metabolic; 12—16u long & 8, wide. Long

flagellum about 1-14 body length, shorter flagellum 4 body length

or less. Periplast finely, spirally striate. Paramylum ranging from

large rod-shaped granules to small oval granules.

Collected from padi swamps, Malacca.

Reported from Czechoslovakia.

The Malayan specimens show minor differences from the type,

being more usually nearly cylindrical. It seems best however, to

include them under this species.

Peranemaceae_

Cells metabolic or rigid, often crawling but also free-swimming,

usually bilaterally asymmetrical, and dorsiventrally organised. Fla-

gella 1-2. Vacuolar system as in Euglena, but usually with two

rod-shaped, or a tubular, siphon, the former closely associated

with the reservoir, while the latter may reach the full length of the

cell. In some cases the siphon can be extruded at the anterior end.

Storage material paramylum granules, and sometimes fat. The

cells are completely without stigma and chloroplast, but occa-

sionally partly digested spheres of chloroplast material may be

seen inside the cell, giving the superficial appearance of chloro-

plasts. Nutrition usually holozoic.

Key to the genera

1. Cell with lflageiipm: vee aah: me op sida ae 2

1. Cell with 2 flagella s...2...43.....4 2, Se a

2. Cell distinctly metabolic with rod-shaped siphon organs in the

FOSELVOIN | 5 o's Se RT en Peranema.

2. Cell rigid, periplast firm, generally somewhat flattened

Petalomonas.

192

Vol. XVI. (1958).

3. Cell not flattened, usually somewhat metabolic

Heteronema

ee 4.

4. Swimming flagellum projecting forwards, much longer than

ES Notosolenus.

4. Swimming flagellum hardly as long as, or much shorter than

the trailing flagellum. Siphon a long tube reaching nearly

OB SE | a Entosiphon.

Peranema Dujardin 1841

Cell very metabolic, with soft periplast. Flagella 1 or occasion-

ally 2. Cytostome and reservoir as in Euglena but a distinct pair

of rod-shaped bodies can be seen against the reservoir. Storage

products paramylum granules, usually round or oval, and often

fat. In many species nutrition distinctly holozoic.

Key to the species

1. Cell elongate wedge-shaped, narrowing towards the anterior,

broadly rounded at the posterior end, with a short, sharp,

laterally-placed tail-piece ............- (i) P. cuneatum.

1. Cell fusiform, tapering posteriorly to a sharp tail, narrowing

slightly towards the anterior end ...... (ii) P. kupfferi.

1. Cell elongate, to almost elongate-cylindrical, slightly narrowed

at the anterior end, bearing at the posterior end a one-sided.

curved, sharp-pointed tail ......... (mi) P. curvicauda.

(i) Peranema cuneatum Playf. (Fig. 7h).

Cell elongate wedge-shaped, narrowing slightly towards the an-

terior end, broadly rounded at the posterior end, with a short,

sharp, pointed tail-piece lying to one side; very metabolic; 66—70«

long X 15—20xu wide. Flagellum about body length. Cytoplasm

practically hyaline, with a few small scattered granules of para-

mylum. The rod-shaped siphon bodies prominent below the reser-

voir. Periplast apparently smooth.

Collected from a fish-pond at Alor Gajah, Malacca, not com-

mon.

Reported from Australia.

Playfair’s description is incomplete, but the shape of the organ-

ism is so characteristic that it seems certain that the Malayan

specimens belong here.

(ii) Peranema kupfferi Skuja (Fig. 71).

Cell fusiform, narrowing at both ends, posteriorly tapering to a

pointed tail, very metabolic, 80u long « 14x wide extended, 45a

193

Gardens Bulletin, S.

long < 33, wide contracted. Flagellum about body length. Peri-

plast distinctly spirally striate. Cytoplasm granular, with numerous

round to oval paramylum granules of varying size almost filling

the cell. Rod-shaped siphon bodies reaching the full length of the

cytostome.

Collected from padi swamps, Malacca, fairly common.

Reported from Sweden.

The Malayan specimens are only half the size of the type, but

agree in nearly every other way, particularly as to shape and

movement. Separation on the basis of size alone is a very dubious

practice, especially as we know so little about the nutrition and

growth of these organisms. For that reason the Malayan specimens

have not been given a varietal name.

(111) Peranema curvicauda Skuja (Fig. 7}).

Cell elongate, lanceolate to almost cylindrical, narrowed slightly

at the anterior end, tapering to a sharp-pointed tail-piece, set to

one side at the posterior end, tail-piece distinctly curved; 50-54

long & 8-10 wide. Cell very metabolic. Flagellum about .body

length or longer. Reservoir long, oval-shaped with well marked

rod-shaped siphon bodies. Periplast firm, longitudinally striate, but

the striations often faint. Cytoplasm clear, containing a number

of loosely arranged oval paramylum granules.

Collected from padi swamps, Malacca, occasional.

Reported from Sweden.

Petalomonas Stein 1859

Cell ovoid, fusiform, oval or triangular, with a firm periplast,

often ribbed; dorsiventrally flattened. Cytostome groove usually

asymmetrical at the anterior end. Flagellum one, thick, spiralling

at the distal end in swimming. Rod-shaped siphon bodies often

difficult to see, although present. Nucleus central, relatively large.

Cell only very slightly metabolic. Nutrition saprophytic to holozoic.

Key to the species

1. Periplast without TDS Oo Pees. sa tse veka ieee calle

1. Peripiast: with a8 atte ss 3+ ss 'kpi ha. acySilie Sgn a

2. A median longitudinal furrow present on both the dorsal and

ventral iface (i012. Aire (i) P. mediocanellata.

2. A furrow present at one side, not median

(ii) P. asymmetrica.

3. Dorsal face with two longitudinal ribs, ventral face without.

Cell oval in outline ...... (iii) P. abcissa var. pellucida.

Vol. XVI. (1958).

3. Cell with 3 well-marked ribs on the dorsal face, the ventral

side slightly concave. Bearing a short sharp tail at the pos-

ONE NNR ee ora Sh tras ce ww ~ os © (iv) P. platyrrhynca.

3. Cell with 5 well marked ribs on the dorsal face, and 2 on the

ventral face, with sometimes a median one as well. Poste-

rior end bearing a short wart-like outgrowth

(v) P. heptaptera.

(i) Petalomonas mediocanellata Stein (Fig. 7k).

Cell ovoid, flattened, rounded at the posterior end, narrowed

and rounded at the anterior end, bearing a median longitudinal

furrow on both the dorsal and ventral faces 18-20u long

13-15 wide. Flagellum about body length. Cytoplasm slightly

granular, with many medium sized round paramylum granules

crowded in the posterior half.

Collected from fish-ponds, Malacca.

Reported from Sweden and U.S.A.

The Malayan specimens are slightly smaller.

(ii) Petalomonas asymmetrica Shawhan and Jahn (Fig. 7m).

Cell ovoid, flattened, broadly rounded at the posterior end, nar-

rowing at the anterior end; on the left side (viewed.from the dorsal

face) there is a deep furrow, with the dorsal edge projecting further

over than the ventral, the furrow curving slightly towards the hol-

lowed ventral surface; 27—30n long & 22—25u wide, 8 thick. Fla-

gellum about body length. Cytoplasm clear at the anterior end,

slightly granular in the posterior half, containing scattered round

granules of paramylum.

Collected from fish-ponds, Malacca.

Reported from U.S.A.

(iii) Petalomonas abcissa (Duj.) Stein var. pellucida Skuja (Figs.

7n, 0).

Cell oval, with sides sometimes nearly parallel, flattened, slightly

narrowed) at the anterior end, broadly rounded to almost truncate

at the posterior end, dorsal surface slightly convex, ventral surface

concave; two prominent ribs run longitudinally the full length of

the dorsal surface; 254 long & 14, wide. Flagellum 1-14 body

length. Periplast firm, colourless smooth. Cytoplasm clear to very

slightly granular, containing several scattered, oval, paramylum

granules.

Collected from padi swamps, Malacca.

Reported from Sweden.

The slightly larger size of the Malayan specimens is of unim-

portance taxonomically with such small organisms.

195

Gardens Bulletin, S.

(iv) Petalomonas platyrrhyncha Skuja (Figs. 7p, q).

Cell oval, flattened, broadly rounded to truncate at the posterior

end, with a sharp tail piece, narrowing anteriorly; 39—40y long X

22-25, wide; there are three prominent longitudinal ribs on the

dorsal face, which is slightly convex, the ribs converging at the

ends; ventral face hollowed. Flagellum about 2 body length. Peri-

plast firm, smooth. Cytoplasm clear, or slightly granular, with

several medium sized round paramylum granules.

Collected from fish-ponds, Malacca.

Reported from Sweden.

(v) Petalomonas heptaptera Prowse sp. nov. (Figs. 7r—u).

Allied to P. platyrrhyncha Skuja, but broader and with seven

ridges instead of five.

Cell broadly oval to almost rectangular in outline, bearing a short

blunt wart-like tail posteriorly, broadly rounded at the anterior

end, dorsi-ventrally flattened, 30u long & 25, wide, 15-18» thick.

On the dorsal surface bearing five prominent wing-like ridges, and

with two such ridges on the ventral surface, with sometimes a third

central one, the whole spiralling to the left. Flagellum about 12

body length. Cytoplasm slightly granular. Paramylum densely

packed, medium-sized granules.

A P. platyrrhyncha Skuja, cui affinissima, carinis 7 (non 5)

latioribus differt.

Cellula ambitu late ovata vel fere rectangularis, posteriore cum

cauda verrucosa obtusa, anteriore late rotundata, utrinque compla-

nata, 30yn longa, 24 lata, 15-18, crassa, dorso 5-alata, ventre

2-, rarissima 3-alata, omnibus alis in laevum spiraliter tortis.

Flagellum sesquialtum. Cytoplasma parce granulosum. Paramylum

dimensione mediocre cum granulis globosis dense dispositis.

Habitat: Malacca, in piscinis (Prowse 227 a).

Heteronema Dujardin 1841 emend. Stein 1878

Cells fusiform to nearly cylindrical, not flattened, but nearly cir-

cular in cross-section, very metabolic. Flagella 2, the longer pro-

jecting forward, the shorter one trailing. Reservoir usually well

marked, with the rod-shaped siphon bodies extending along the

side. Saprophytic to holozoic in nutrition.

Key to the species

1. Cell long and slender, ending in a sharp point at the posterior

end .. .... Uy Rais ea oe ee (i) H. leptosomum.

1. Cell almost cylindrical (when swimming) to slightly fusiform,

slightly narrowed but rounded at both ends

(Gi) H. polymorphum.

196

Vol. XVI. (1958).

1. Cell almost cylindrical, narrowed slightly and truncate at each

end (when swimming) with a distinct invagination at the

apHTIOT CBs oes eee se we es 8s (iii) H. invaginatum.

(i) Heteronema leptosomum Skuja (Fig. 8c).

Cell long and slender, sometimes almost needle-shaped, slightly

truncate at the anterior end, tapering to a point at the posterior

end; 40-55 long « 3-7, wide. The long flagellum about 7 body

length, the shorter about 4 body length. Periplast apparently

smooth. Cytoplasm hyaline, with scattered grains of paramylum.

In the swimming stage the cell is extended and long, and very

thin, but in the metabolic stage a large bulge forms at the poste-

rior end and rapidly progresses forwards.

Collected from padi swamps, Malacca.

Reported from Lithuania.

The very characteristic long slender shape of this species makes

it unmistakable.

(ii) Heteronema polymorphum Defi. (Fig. 8b).

Cell cylindrical to slightly fusiform when swimming, slightly nar-

rowed and rounded at both ends; 80-100, long « 10—20n wide,

in the metabolic stage often quite short and broad. Long forward

projecting flagellum nearly body length, shorter trailing flagellum,

1/3-1/2 body length. Periplast apparently smooth. Cytoplasm

packed with oval paramylum granules.

Collected from padi swamps, Malacca.

Reported from France.

(iii) Heteronema invaginatum Prowse nov. sp. (Fig. 8a)

The most characteristic feature of this species is the invagination

of the posterior end, a feature found in no other species. Cell trun-

cate fusiform in outline, round in section, narrowing slightly at

the anterior end; posterior end markedly invaginate for about 4

length of the cell, narrowing slightly when the cell is swimming,

but opening out when the cell is stationary; cell 48-50» long x

8-9. wide. Cytostome and reservoir well marked at anterior end,

reaching nearly + way along cell. Storage products densely packed

granules of paramylum. Longer flagellum pointed forward during

swimming, about body length, shorter trailing, about + body length.

Inter Heteronema spp. forma ambitu fusiformi utrinque truncata,

transverse orbiculari, posteriore quarta parte invaginata, haec spe-

cies sat distincta.

Cellula 48-50 longa, 8-9» lata, parum metabolica, ambitu

truncato-fusiformis, transverse orbicularis; anteriorem versus paulo

197

Gardens Bulletin, S.

angustata, cum cytostomate conspicuo quarta parte ex ejus apice

attingenti; posteriore quarta parte invaginata, movente angustata,

immovente dilatata. Granula paramylonica copiosa, dense cumu-

lata. Flagella duo: longius cellulae aequilongum, motionis direc-

tionem indicans, alterum brevius priori circa aequilongum.

Habitat: Malacca in locis oryzalibus paludosis (Prowse 240 b).

XS o>

FIGURE 8°

Colourless Eugleninae in Malaya:—

a, Heteronema invaginatum Prowse; b, Heteronema polymorphum Défl.;

c, Heteronema leptosomum Skuja; d, Notosolenus stenochismos Skuja; e,

Notosolenus orbicularis Stokes; f-g, Notosolenus similis Skuja; h-i, Noto-

solenus lens Skuja; j, Entosiphon sulcatum (Duj.) Stein; k, Entosiphon

obliquum Klebs.

Figure b is at a lower magnification than the others.

198

Vol. XVI. (1958).

Notosolenus Stokes 1884 emend. Skuja 1939

Cells ovoid, elliptical or almost round, distinctly flattened, with a

distinct ventral groove which may reach the full length of the cell.

Periplast delicate, with or without striations. Cytostome opening

curved. Swimming flagellum pointing forward, long, trailing flagel-

lum short. Some forms have a superficial resemblance to Aniso-

nema Dujardin emend. Stein, but it is always the long flagellum

which points forward for swimming, and not the shorter one. Nut-

rition saprophytic or holozoic (specimens have been seen in

which chloroplast material from Spirogyra has been visibly in-

gested). |

Key to the species

1. Cell ovoid, narrowed towards the anterior ............ pig

1. Cell broadly ellipsoidal, equally rounded at both ends ... 3.

2. Cell only slightly flattened, narrowing at the anterior end to

a more or less blunt point; dorsal surface convex, the vent-

ral surface bearing a shallow lateral groove

(i) N. similis.

2. Cell slightly flattened, narrowing at the anterior end, rounded

or conical at the posterior end, dorsal face convex, ventral

face concave with a deep median groove

(ii) N. stenochismos.

3. Cell small, markedly flattened, with a prominent wide, median

MMAR Deer te a's bole eget e ns (iii) N. orbicularis.

3. Cell much less flattened, lens-shaped, almost round in outline,

SRONVS WET 5 Pe a rsa ece'e a so 2 (iv) N. lens.

(i) Notosolenus similis Skuja (Figs. 8f, g).

Cell ovoid, slightly flattened, narrowed towards the anterior end

to form a blunt point, broadly rounded at the posterior end;

24-28 long * 15, wide. Dorsal face convex, ventral face slightly

concave, with a very shallow furrow to one side caused by a

slight inrolling of the periplast. Longer swimming flagellum point-

ing forwards, 14 body length, shorter trailing flagellum about 4

body length. Cytoplasm clear, with scattered granules of para-

mylum.

Collected from padi swamps, Malacca.

Reported from Lithuania.

The Malayan specimens are twice the size of the Lithuanian

specimens, and the paramylum granules are appreciably larger

199

Gardens Bulletin, S.

than as described by Skuja. Nevertheless they come very close to

this species, and it seems better to include them here, rather than

erect a new species.

(ii) Notosolenus stenochismos Skuja (Fig. 8d).

Cell ovoid, slightly flattened, narrowed and slightly drawn out at

the anterior end, notched at one side, rounded to slightly conical

at the posterior end; 27 long « 14 wide. Dorsal face very con-

vex, ventral face much less convex with a deep narrow groove,

circular in cross-section, and running the full length of the cell,

slightly inclined to one side. Swimming flagellum about body

length, trailing flagellum about 1/3 body length. Cytoplasm hya-

line, with paramylum granules of varying'size, from very small to

quite large. Reservoir distinct, showing the two rod-shaped siphon

bodies.

Collected from the padi swamps, Malacca.

Reported from Lithuania.

(iii) Notosolenus orbicularis Stokes (Fig. 8e).

Cell small, almost circular in outline, broadly rounded at both

ends, distinctly flattened, 14u long « 12 wide, slightly convex

on the dorsal face, flattened on the ventral face, with a wide

median groove running the full length of the cell. Swimming flagel-

lum about body length, trailing flagellum about 4 body length.

Cytoplasm faintly granular, with scattered paramylum grains.

Collected from padi swamps, Malacca.

Reported from U.S.A.

(iv) Notosolenus lens Skuja (Figs. 8h, i).

Cell small, round ellipsoidal to almost circular in outline, less

flattened than the preceding species; 18-20 long x 16y wide,

flattened on the ventral face, convex on the dorsal face; ventral

furrow very short to almost absent. Swimming flagellum 1-12

body length, trailing flagellum 1/2—1/3 body length. Periplast

aparently smooth. Cytoplasm clear, to faintly granular, with scat-

tered paramylum grains.

Collected from padi swamps, Malacca.

Reported from Sweden.

Entosiphon Stein 1878

Cells ovoid or ellipsoidal, more or less flattened. Periplast firm,

with longitudinal striae, ribs or furrows. Flagella 2, the swimming

flagellum being shorter than the trailing flagellum. Possessing a

distinctly tubular siphon, which can be extruded at the anterior

end.

200

Vol. XVI. (1958).

Key to the species

1. Cell oval, periplast with 4-8 longitudinal ridges

(i) E. sulcatum.

1. Cell ovoid, often terminating in a blunt point posteriorly,

periplast very delicately striated, longitudinally. Siphon fre-

qucmtly-protreades =o. 0 eT. (ii) E. obliquum.

(i) Entosiphon sulcatum (Duj.) Stein (Fig. 8)).

Cell oval, rounded at both ends, 20u long « 14, wide. Peri-

plast with 4-8 longitudinal ridges. Swimming flagellum less than

body length, trailing flagellum up to twice body length. Siphon

reaching nearly the full length of the cell. Cytoplasm clear, with

scattered small granules of paramylum. Saprophytic or holozoic,

often containing ingested particles of chloroplasts.

Collected from fish-ponds, Malacca.

Reported from Europe and America

(11) Entosiphon obliquum Klebs (Fig. 8k).

Cell (in swimming state) ovoid, somewhat truncate at the ante-

rior end, narrowly rounded at the posterior end, usually with a

blunt point; 24u long x 18, wide. More frequently the cell be-

haves metabolically, protruding the siphon (see fig. 8k) and the

characteristic shape of the swimming state is obscured. Swimming

flagellum about 4 body length, trailing flagellum body length or

over. Periplast so very faintly striate longitudinally, that the stria-

tions are often not visible. Cytoplasm clear, but the cell is usually

packed with oval paramylum granules. Siphon reaching the full

length of the cell.

Collected from padi swamps, Malacca. Widespread in distri-

bution.

The Malayan specimens are larger, and the frequent metabolic

state obscures any marked narrowing of the posterior end. Never-

theless, this is a very common species, which must inevitably show

some variation in behaviour. The extremely fine striations of the

periplast make it certain that the Malayan specimens belong here.

I would like to express my gratitude to Mr. Purseglove, lately

Director of the Botanic Gardens, Singapore, and to his staff, for

the generous way in which they have provided facilities for my

work prior to my moving up to Malacca, and for the frequent help

they have given me subsequently. I am particularly grateful to the

Director for allowing me to publish this article through the medium

of the Gardens Bulletin. To Mr. H. M. Burkill, at present Acting

Director of the Botanic Gardens, I must express my thanks for the

201

Gardens Bulletin, S.

extremely interesting samples of pond water he sent me from time

to time. I would also like to thank Professor Gilliland, and his

colleagues of the Department of Botany, University of Malaya, and

to the University in general, for all the help and encouragement

they have given me in my work, and for so generously allowing

me to use the library. To Mr. M. W. F. Tweedie, Director of

Raffles Museum, I am particularly grateful, because many of the

illustrations to this paper were drawn with the aid of the camera

lucida which he so kindly lent me, while I was awaiting my own

from Britain. To Dr. Hickling, Fisheries Adviser to the Colonial

Office, and so very much responsible for the establishment of the

project at Batu Berendam where so much of this work has been

done, I owe special thanks for being allowed to carry out this re-

search, and for permission to publish the material in this paper.

To Mr. S. F. Owen, Resident Engineer, P.W.D., Batu Berendam,

I am particularly grateful for being allowed to come up and work

long before the completion of the constructional work at the pro-

ject, and for so generously providing facilities for me to carry out

my research. Finally, to Dr. Furtado of the Botanic Gardens,

Singapore, I am deeply grateful for his help in Latin diagnosis,

and for his advice in suggesting names for new species.

Bibliography

ALLERGE, C. & JAHN, T. L. (1943). A survey of the genus Phacus

Dujardin (Protozoa): Euglenoidinae. Trans. Amer.

Micr. Soc. 62: 233-244, pls. 1-3.

BERNARD, C. (1908). Protococcacées et desmidiées d’eau douce

recoltées a Java. Dept. Agric. Indes Néer-Batavia

Lands.: 204—207, pl. XVI. .

BERNARD, C. (1909). Sur quelques algues unicellulaires recoltées

dans le domaine Malais. Buitenzorg: 81, pl. 6.

ConraAD, W. (1935). Etude systematique du genre Lepocinclis

Perty. Mem. Mus. Roy. d’Hist. Nat. Belg. 1: 1-85.

CONRAD, W. & VAN MEEL, L. (1951). Materiaux pour une mono-

graphie de Trachelomonas Ehrenberg 1834, Strom-

bomonas Déflandre 1930 et Euglena Ehrenberg

1832, genres d’Euglenacées. Mem. Inst. Roy. Sc.

Nat. Belg. 124: 1-124, pls. 1-19.

DEFLANDRE, G. (1926). Monographie du genre Trachelomonas,

Nemours.

DEFLANDRE, G. (1928). Algue d’eau douce du Venezuela (Fla-

gellés et Chlorophycées) Rev. Algol. 3: 213:224.

202

Vol. XVI. (1958).

DEFLANDRE, G. (1930). Strombomonas nouveau genre d’Eugle-

nacées (Trachelomonas Ehr. pro parte). Arch. Pro-

tistenk 69: 551-614

DEFLANDRE, G. (1932). Contributions 4 la connaissance de fla-

gellés librés I. Ann. Protist. III: 219-239, pls. 21—

2s.

FritscuH, F. E. (1935). The structure and reproduction of the

algae I Cambridge: 724-744.

Gospics, M. (1935). The genus Euglena. Wisconsin.

HUBER-PESTALOZzI, G. (1936). Phytoplankton aus seen und

sumpfen Javas gesammelt von Prof. C. Schroeter.

Ber. Schweiz. Bot. Ges. 46: 131-168.

HUBER-PESTALOZzI, G. (1955). Das Phytoplankton des Susswas-

sers. Stuttgart. 4—Euglenophyceen.

LEFEVRE, M. (1933). Contribution a la connaissance des flagellés

d’Indochine. Ann. Crypt. Exot. 6: 258-264.

NYGAARD, G. (1932). Contributions to our knowledge of the

freshwater algae of Africa 9—Freshwater algae and

phytoplankton from Transvaal. Trans. Roy. Soc. S.

Africa 25: 18, fig. 19.

Oye, P. VAN (1925). Flagellates du Congo Belge. Bull. Soc. Roy.

Bot. de Belg. 58 (1): 11-19.

PLAYFAIR, G. I. (1921). Australian freshwater flagellates. Proc.

Linn. Soc. N.S.W.: 99-146, pls. 1-9.

POCHMANN, A. (1942). Synopsis der gattung Phacus Arch. Pro-

tistenk. 95 (2): 8-252, figs. 1-170.

Prescott, G. W. (1951). Algae of the Western Great Lakes

Area. Michigan: 387—420, pls. 83-89.

PRINGSHEIM, E. G. (1942). Contributions to our knowledge of

saprophytic algae and flagellata MJ—Astasia Dis-

tigma, Menoidium and Rhabdomonas. New Phytol.

41: 171-205.

PRINGSHEIM, E. G. (1948). Taxonomic problems in the Eugleni-

nae. Biol. Rev. Camb. Phil. Soc. 23: 46-61.

SkUJA, H. (1949). Zur siisswasseralgenflora Burmas. Nova acta

reginae soc. scientiarum upsaliensis 14 (5): 162-—

165, pl. 36.

SKVvoRTzOW, B. V. (1927). Die Euglenaceengattung Phacus Du-

jardin. Ber. d’Deutsch. Bot. Gesell. 46: 105-125.

203

Gardens Bulletin, S.

SKVORTzOW, B. V. (1937). Contributions to our knowledge of the

freshwater algae of Rangoon, Burma, India. I—

Euglenaceae from Rangoon. Arch. Protistenk. 90:

69-87, pls. 9-12.

SUXENA, M. R. (1955). Freshwater Eugleninae from Hyderabad,

India I. Journ. Ind. Bot. Soc. 34 (4): 429-450.

TIFFANY, L. H. & BRITTON, M. E. (1952. The algae of Illinois.

Chicago.: 315-327.

Wana, C. & Nig, D. (1934). Notes on Trachelomonas in Nan-

king. Sinensis 5: 122-146.

204

A Revision of the Malayan Myristicaceae

By JAMES SINCLAIR, B.SC.

GENERAL PART

Introductory Remarks

THIS ACCOUNT was originally prepared and drafted at Singapore,

but later corrected after a visit to Leiden in April and May 1956,

where many valuable Malaysian collections are housed. After

seeing these collections and others at Florence, Geneva, Munich,

Kew, the British Museum and Edinburgh, the author is firmly

convinced that a ‘regional flora’ not only of the Myristicaceae

but of all groups cannot be satisfactory, unless the species of the

whole adjacent Malaysian region are also examined. Much was

learnt about geographical distribution and many species thought

to be confined to Malaya were found to occur in Sumatra and

Borneo. Several of the specific names in the original draft had to

be altered as a result of these visits, and some more may yet have

to be changed when specimens from the whole Malaysian region

are examined carefully. These name changes will affect poly-

morphic species with a wide distribution range, which have been

given many different names throughout the region. Thus, while

realizing that it would have been better to have revised Myristica-

ceae for the whole of the Malaysian region before attempting the

local account, the latter is now published, as the full revision will

take some time to prepare.

The family Myristicaceae is a difficult one and many mistakes

have been made in identifying species. This is due to the following

reasons:— The trees are dioecious and female flowers are often

scarcer than the male. The leaves are often similar in form and

texture, especially in Myristica, so that one has to be able to co-

relate male flowers, female flowers and fruit of a single species.

The keys are lengthy as there are many species in each genus and

if not well constructed, they may be misleading. The brief ac-

counts of the early authors such as Miquel and Roxburgh, founded

often on insufficient material, have caused much trouble. The

types of the New Guinea material, housed at Berlin, were destroyed

by bombs during the war in March 1943. A few duplicates of these

types are to be found in Breslau and elsewhere and it is hoped

205

Gardens Bulletin, S-

that others can still be located. This deplorable loss makes the

work of identifying new material from New Guinea extremely

difficult.

Brief History

The name Myristica first appears in the second edition of the

Genera Plantarum, published in 1742, where Linnaeus mentions a

single species, the nutmeg of commerce. The genus Myristica, ac-

cording to the Rules of Botanical Nomenclature, is not valid until

1760 in Ludwig, Definitiones Generum Plantarum where Boehmer

(page 513) selects M. fragrans as the type of the genus. Much has

been written about M. fragrans, the nutmeg of commerce, and the

longest account is to be found in Warburg’s book Die Muskatnuss.

His classic systematic publication, giving it the full title, Monogra-

phie der Myristicaceen in Nova Acta Academiae Caesareae Leo-

poldino-Carolinae Germanicae Naturae Curiosorum, Abhandlungen

der Kaiserlichen Akademie der Naturforscher 68 (1897) is by far

the most important treatment of the family. It will be referred to

here in citations simply as Warb., Monog. Myrist. It gives an ac-

count of the history of the family, so only a few of the more im-

portant works need be mentioned here, not forgetting the account

in De Candolle’s Prodromus 14 (1856), the only complete and

best monograph before Warburg’s time. King’s Species of Myristica

of British India was published in 1891, a few years before War-

burg’s. His plates are much better than Warburg’s, but he deals

only with the Indo-Malayan species and rather surprisingly

recognized only one genus Myristica, but divided it into sections.

For the Malayan flora, the only other two important contri-

butions are Gamble (1912) in Materials for a Flora of the Malayan

Peninsula and Ridley’s The Flora of the Malay Penisula volume 3

(1924). Gamble’s is useful in that he quotes collector’s numbers,

but several of these are wrongly identified and should be checked

with the list at the end of the present revision. The differences

between Ridley’s account and mine will be pointed out presently.

Since Warburg’s time several new species have been added to the

Philippine flora by Elmer and Merrill and these are to be found in

scattered papers in the Philippine Journal of Science and in Leaf-

lets of Philippine Botany. Markgraf in 1935 published a short but

important paper on the Myristicaceae of New Guinea in Engler’s

Botanische Jahrbucher. In it, several of Warburg’s species are re-

duced to synonyms and some of his doubtful species are given more

complete descriptions. A short account of some of the Pacific

islands’ species is given by A. C. Smith in Bull. Torr. Bot. Club

(1941) 66.

206

Vol. XVI. (1958).

With regard to the African species most of Warburg’s names

remain unaltered, but a few new species have been added and the

genus Mauloutchia included in Brochoneura. Such changes as there

are, will be found in the following floras: ——G. Gilbert et G. Trou-

pin, Flore de Congo Belge et Du Ruanda-Urundi 2 (1951); Eg-

geling and Dale, The Indigenous Trees of the Uganda Protectorate

(1951); H. Perrier de la Bathie, Flore de Madagascar et des

Comores (1952); Hutchinson and Dalziel, Flora of West Tropical

Africa, second edition, revised by Keay (1954).

For the American Myristicaceae we have van Ooststroom’s re-

vision of the Surinam species in Flora of Suriname vol. 2 (1934)

113 and in Additions (1939) 473. A. Ducke published “Notes on

the Myristicaceae of Amazonian Brazil, with descriptions of New

Species” in Journ. Wash. Acad. 26 (1936) 213-222 and 253-264.

A. C. Smith published a comprehensive revision “The American

Species of Myristicaceae” in Brittonia vol. 2 No. 5 in December

1937. In it many new species have been added. Of the five Ame-

rican genera Compsoneura, Dialyanthera, Iryanthera, Osteoph-

loeum and Virola, Warburg recognized 4, 2, 4, 1 and 27 species

respectively, while Smith gives 8, 6, 20, 1, and 38 species, a total

of 73. In a very recent paper, just received, entitled “Studies of

South American Plants XV”, in American Journ. Bot. 43, 8

(1956) 573, A. C. Smith lists ten species of Myristicaceae as new

records to Amazonian Colombia, including one new species of

Virola new to science and one new combination in Jryanthera.

Summary of the Main Changes made in the Present Revision

In the present revision 53 species, 5 varieties and 1 form are

described as against 45 species and 4 varieties in Ridley’s Flora.

Breaking these up into individual genera we have:—Gymnacran-

thera 4 species + 1 variety, Ridley 3 species + 2 vars.; Hors-

fieldia 19 species + 1 var., Ridley 18 species; Knema 19 species

+ 3 vars. and 1 form, Ridley 14 species + 2 vars.; Myristica 11,

Ridley 10 but two others M. fragrans and M. guatteriifolia are

mentioned by him.

New Species

The following 6 are new species: —

Horsfieldia penangiana, H. punctatifolia, H. subalpina, Knema

communis, K. plumulosa and K. rigidifolia.

207

Gardens Bulletin, S.

New Varieties

The following 2 are new varieties: —

Knema glaucescens var. patentinervia and var. cordata. One new

variety for Borneo, namely Horsfieldia macrocoma var. rufirachis

is also described here.

New Forms

The following is a new form:—

K. glaucescens f. rubens.

New Combinations

The following 7 are new combinations:—

Gymnacranthera bancana (Miq.) and var. borneensis (Warb.),

G. eugeniifolia (A. DC.) and var. griffithii (Warb.), Knema scorte-

chinii (King), K. stenophylla (Warb.) and H. polyspherula (Hk.

f. emend. King).

New Status

The following receive new status:—

Horsfieldia subglobosa var. brachiata (King), H. macrocoma

var. canarioides (King), Myristica elliptica var. simiarum (A.

DC.) and var. celebica (Miq.).

New Records

The following 3 are new records for the Malay Peninsula: —

Gymnacranthera contracta, Horsfieldia glabra and Knema man-

daharan.

Differences between Ridley’s Flora and the Present Revision

RIDLEY’S FLORA PRESENT REVISION

Gymnacranthera .. G. murtonii made a synonym of G. bancana (M. -°

bancana). G. eugeniifolia used instead of G.

farquhariana, the latter name retained for the

Indian plant. G. farquhariana var. major and

var. griffithii are similar and placed in var.

griffithii under G. eugeniifolia. One additional

species G. contracta, a new record is added.

Horsfieldia .. H. bivalvis used instead of H. globularia. The

variety paludicola of H. fulva is only H. crassi-

folia. H. racemosa reduced to H. macrocoma

var. canarioides. H. bracteosa used instead of H.

amygdalina for Malayan material; there being

no true H. amygdalina in Malaya. H. majuscula

reduced to a synonym of H. subglobosa. H. bra-

chiata reduced to a variety of H. subglobosa.

Three new species added, H. punctatifolia, H-.

208

Vol. XVI. (1958).

RIDLEY’S FLORA PRESENT REVISION

subalpina and H. penangiana (the latter includ-

ed a part of the specimens quoted under G.

farquhariana var. griffithii). One new record A.

glabra added. H. lemanniana becomes H. polys-

pherula except for the type which is H. irya.

Knema ... -.. K. meridionalis reduced to K. latericia (not in

Ridley’s Flora). K. conferta var. scortechinii

raised to specific rank. K. wrayi reduced to a

synonym of K. glaucescens. K. plumulosa sub-

stituted for K. cantleyi sensu auct. non Hk. f. =

K. laurina. K. missionis made a synonym of K.

globularia. The following new species added, K.

communis and K. rigidifolia. The following new

record added, K. mandaharan. K. stenophylla

= (G. stenophylla Warb.) substituted for K.

geminata. The geminata found in Sumatra and

not in Malaya is only K. glaucescens.

Myristica .. .. M. suavis reduced to a synonym of M. crassa. M.

frangrans is fully described. It was only men-

tion briefly in Ridley’s Flora. M. guatteriifolia is.

added, but it is not a new record. Ridley merely

mentioned it, but did not know it was native.

The reasons for these changes and other relevant details are

discussed under the species concerned in the systematic part.

Scope for future work

The descriptions of the Malayan Myristicaceae are not yet

absolutely complete. In the present account, certain descriptions

of male or female flowers and other missing characters have been

added. These were previously unknown and I have been fortunate

to see some of them in the field in living specimens. A list of

wanted organs or characters is now appended in the hope that

future students and collectors will know what is required.

Name Parts unknown or wanted

Gymnacranthera bancana -- Colour of the fresh, mature aril.

G. contracta his es .. Colour of the fresh, mature aril.

Horsfieldia flocculosa ee .. Female flowers and mature fruit

unknown. Bark characters.

H. glabra oa at .. Mature fruit of Malayan material.

Bark characters.

H. grandis es ae .. Mature fruit.

H. macrocoma var. canarioides .. Colour of the fresh, mature aril.

More information about the

monoecious condition of the

flowers and if they are ever

dioecious.

H. penangiana .. * .. Female flowers and fruit unknown.

Bark characters.

209

Name

H. ridleyana

H. subalpina

H. superba

H. tomentosa

Knema curtisii

Gardens Bulletin, S.

Parts unknown or wanted

Mature fruit. Bark characters.

Female flowers unknown. Bark

characters. More material and

notes.

Colour of the fresh mature aril.

Bark characters.

K. glaucescens var. cordata Female flowers unknown.

K. glaucescens var. patentinervia .. Bark characters.

K. oblongifolia .. Bark characters.

. retusa i aS .. Male and female flowers unknown.

Bark characters.

Female flowers unknown. Mature

male flowers also wanted. Colour

of the fresh mature aril.

Female flowers only known from

the remains of young ovaries.

Colour of the fresh mature aril.

Female flowers unknown.

Female flowers unknown. Bark

characters wanted. Can the bark

be distinguished from that of

M. lowiana?

Female flowers unknown.

Bark characters.

. rigidifolia

K.. scortechinii

K. stenophylla

Myristica cinnamomea

M. gigantea

M. lowiana

M. malaccensis ..

(1) These are the chief points required for further and more

adequate descriptions. Other minor ones such as the presence of

stilt roots and the colours of fresh leaves, flowers and fruit should

be noted, as I have not seen all the species in the fresh condition.

(2) Further records on distribution of individual species in the

various Malayan States are wanted, especially with regard to rare

species, e.g. Knema retusa. New records would also be welcome

for common species if they are absent from any state in which they

might reasonably be expected to occur. For example in my three

visits to Trengganu in 1953, 1954 and 1955, several common

species were obtained which were new records for that state. It will

be noted that there is only one record for Perlis, namely Knema

globularia.

(3) It is regrettable, that no chromosome count has yet been

made of the Myristicaceae as is the case in many tropical plants.

Such a study is very much overdue, and doubtless, it would pro-

vide significant results. It would be interesting to see if the chro-

mosome counts vary in the case of certain closely related groups

and species such as Myristica crassa and M. teijsmannii, the M.

210

Vol. XVI. (1958).

guatteriifolia complex, the closely related M. elliptica, celebica and

simiarum (see my treatment of these three, page 356) or Horsfieldia

macrocoma and its varieties and allies.

(4) The presence of the red sap usually called ‘kino’ seems to

vary in amount in different species. Sometimes it is described as

copious and gushing out or at other times meagre. This is a physio-

logical problem and it would be interesting to see how the amount

varies in any one species over a yearly period, if it is always copi-

ous or otherwise and if it is related to rainfall, amount of nutritive

salts available, edaphic factors or the age of the tree, since the

quantity appears to be scanty in very young trees.

(5) Very little is known about the general biology of the family.

Many problems remain to be solved here. Most of what we do

know has been gained from observations on Myristica fragrans

under cultivation. If other species proved to be of economic value,

we should, doubtless, know more of their special requirements and

life history. A study of germination would make an interesting re-

search problem. We know that the seeds of M. fragrans take from

one_month to six weeks to germinate, but little is known about

germination in the wild species. Fresh seeds of M. maingayi and

Horsfieldia wallichii sink in water and thus would not be distri-

buted easily in nature if they fell into streams. The hollow seeds

of H. irya are said to float. This species certainly has a wide distri-

bution. The seeds of M. elliptica float even if the arils are removed,

and so do its fruits. The dry seeds of M. fragrans (as sold in the

market for culinary purposes) float, but if the testa is removed

they sink.

GENERAL CHARACTERS

Although this account primarily deals with the four genera Gym-

nacranthera, Horsfieldia, Knema and Myristica in Malaya, some

general reference is necessary to certain Malay Islands’ species and

to the African and American genera as well, otherwise a very one-

sided conception of the family will result. A complete discussion

of the African and American genera is outside the scope of the

present work. It is well dealt with in Warburg’s monograph and

Smith’s revision.

Habitat and Distribution

The majority of Myristicaceae are confined to lowland tropical

rain forests of the Old and New Worlds. There are more species in

the Malaysian region than elsewhere. Two species Virola sessilis

211

Gardens Bulletin, S.

and subsessilis are found in the dry savannah regions of Mid-

Brazil, but the majority are shade-loving and do not like exposed

places. Myristica fragrans, when cultivated on Bukit Mertajam, an

exposed hill in Province Wellesley, did not grow above 300 m., but

had there been more shade, it would probably have ascended

higher. Although the majority are confined to lowland forests, the

following Malayan species are mountain plants and only grow at

altitudes of 1,000—1,150 metres or more—Horsfieldia glabra, H.

subalpina, Knema rigidifolia and K. oblongifolia var. monticola.

On Mount Kinabalu in Borneo several species have been collected

at 1,860 m. while others occur in New Guinea at 2,000 m. or

slightly over.

The four Malaysian genera, Gymnacranthera, Horsfieldia, Kne-

ma and Myristica are not found in Africa or America. Myristica

has more species than any other genus and has its chief centre of

distribution in New Guinea, where there are still undescribed spe-

cies. There is a second centre of distribution in the Malay Penin-

sula (see further notes under Myristica in the systematic part).

Its range of distribution covers Ceylon, South India to the Concan,

Burma, the Nicobars and Andamans, Indo-China, Siam, Malaya,

the Malay Islands, North Australia, Polynesia and Micronesia.

According to A. C. Smith in Bull. Torr. Bot. Club. 66 (1941)

397, there are 2 species in Samoa (inutilis and hypargyraea), 1 or

possibly 2 in Tonga (inutilis and hypargyraea), 4 or possibly 5 in

Fiji (macrantha, castaneifolia, chartacea, gillespieana and grandi-

folia, which is sterile and which may be macrantha or castaneifolia),

2 in the New Hebrides (inutilis and guillauminiana), 1 in the Caro-

lines (insularis) and 8 in the Solomons including inutilis. Some of

the species from the Solomons are found in New Guinea.

Horsfieldia is the most abundant in species after Myristica, and

Knema comes third in order. Horsfieldia has much the same distri-

bution as Myristica and has recently been found in North Australia.

The two chief centres of its distribution are the Malay Peninsula

and New Guinea, but allowing for undescribed species, New Guinea

will have the majority. There is also a fair number of species in

Borneo, and in the Moluccas, some undescribed. Knema is absent

from Ceylon and Australia, the two chief centres being the Malay

Peninsula and Borneo. The species diminish in number towards the

east part of Malaysia, there being only one species in New Guinea.

Gymnacranthera has about six species, some of them very close to

each other. It is absent in Ceylon, the Himalaya-Khasia-Silhet re-

gion, Burma, the Andamans and Nicobars, Indo-China and

Australia.

242

Vol. XVI. (1958).

The American genera are Compsoneura, Dialyanthera, Iryan-

thera, Osteophloeum and Virola. The latter has the largest number

of species after Knema, now some 38 in number. Next is /ryan-

thera with 20 species. The rest are small genera with 1—8 species.

The following is the detailed distribution for the American genera:—

Compsoneura, South Mexico, Guatemala, British Honduras, Costa

Rica, East Peru, Colombia and North Brazil; Dialyanthera, Pana-

ma, Peru, Ecuador and east Colombia; Jryanthera, Peru, Colombia,

British and French Guiana, Surinam and Brazil; Osteophloeum,

Peru, Colombia, British Guiana and North Brazil; Virola, Central

America, West Indies, Peru, Colombia, Venezuela, British and

French Guiana, Brazil and Bolivia.

The African genera are Brochoneura, Madagascar; Cephalos-

phaera, (formerly Brochoneura usambarensis) Tanganyika; Coelo-

caryon, the Belgian Congo and the Cameroons; Pycnanthus, Libe-

ria, Gold Coast, Cameroons, Fernando Po, the Belgian Congo and

Uganda; Scyphocephalium, Nigeria and the Cameroons; Staudtia,

Cameroons, the Beigian Congo and Uganda. Mauloutchia has been

sunk into Brochoneura by H. Perrier de la Bathie in Flore de

Madagascar and he says it is not distinguishable from the latter

genus. Pycnanthus has the largest number of species in Africa.

The seeds are distributed chiefly by Fruit Pigeons, Carpophaga

and Myristicivora, Hornbills and Birds of Paradise which are at-

tracted by the brightly coloured arils. Pigeons and doves are the

most important dispersers since they fly long distances over land

and sea; they can fly fast and are found all over the world in hot

and cold climates. The Dutch formerly tried to control the market

of nutmegs by confining the plantations to Banda and Amboina,

but pigeons carried the seeds to other islands. A Hornbill, Buceras

ruficollis is said to devour nutmegs in the plantations of Ceram.

Warburg described Myristica avis-paradisiacae from the seeds

found in the stomach of a Bird of Paradise. He states that the seed

is the smallest in the genus. Probably other birds besides pigeons

and hornbills eat the seeds and arils of Myristicaceae, but small

birds certainly-will not be able to swallow the larger seeds such as

those of Myristica maingayi and Horsfieldia wallichii. They prob-

ably peck off the aril and leave the seed intact. Smaller seeds are

certainly swallowed. They may be disgorged later or may pass

through the bird’s intestine.

It is possible that water dispersal is more important than was

formerly imagined. I have found that the ripe fruit of Horsfieldia

wallichii floats in water and so does the seed while surrounded by

the aril. Between the aril and the testa there is a considerable

amount of trapped air. If the aril is removed the seed sinks at once.

213

Gardens Bulletin, S.

The seeds of Myristica maingayi also sink in water, but I have not

tried them with the arils attached. It is probable, therefore, that the

fruits of most species float in water, but that the seeds alone sink

except in the case of M. elliptica where the fruits and the seeds,

even without the aril, float. Probably the seeds of some smaller

species may float. Those of Horsfieldia irya have a hollow in the

endosperm and there is evidence that they are distributed by water.

This species is common by the banks of streams and has a wide dis-

tribution. Seeds, even if they do not float, could easily be carried

about in rapid currents or even in rushes of rain water which are

often forceful and sudden in Malayan forests. Since the seeds con-

tain a lot of fat in the endosperm, it is to be expected that they

will not survive long in sea-water. No fossil seéds have been found.

Squirrels and other rodents, no doubt, are agents in carrying

seeds to their nests.

General Appearance and Growth Habit

Myristicaceae may be recognized in the forest, often even at a

distance, by their straight, mast-like trunks, the Garcinia or Anno-

naceous style of branching and the crown of foliage at the top. If

there is plenty of light, the foliage may be borne to within a few

feet from the ground. The branches are slender and more or less

whorled. They may leave the tree at an angle of 50°, but are often

horizontal, sometimes even at angles greater than 90°, where, in

such cases their weight drags them down. The twigs are more or

less distichous, as are the leaves. The shape of the crown is often

pyramidal, the best examples of which are seen in Knema furfu-

racea, K. hookeriana and K. latericia. The crown of Myristica

fragrans is more rounded, but often trimmed artificially so that the

nuts can be easily and conveniently harvested.

The majority are small trees 5—12 m. high. A few reach heights

from 30-40 m. In Malaya, Myristica gigantea is probably the tal-

lest. Then come M. iners, M. maingayi and Horsfieldia wallichii.

The American Osteophloeum may reach 40 m. and Virola bicuhyba

36 m. Some are shrubs less than 5 m. high. Compsoneura debilis is

1—3 m. high, with a stem no thicker than the finger. Virola sessilis,

a plant of the dry Brazilian campos, is the smallest of them all,

being 30—60 cm. high.

Roots

Little is known about roots. Those of Myristica fragrans are

aromatic and lie near the surface of the ground. It has been found

that when trenches are dug round the tree for manuring purposes,

the roots are often injured. Again, care must be taken when plant-

ing out seedlings of this species, to see that the tap root is not bent,

214

Vol. XVI. (1958).

otherwise it never straightens properly and the tree will remain

stunted. The root system of other species is probably also super-

ficial, especially in those which grow in the swamp forest. Stilt

roots are often present in these, but if the same species grows in

dry ground, these roots are very rudimentary or absent. Malayan

species with stilt roots are:—-Gymnacranthera eugeniifolia var.

major, Horsfieldia crassifolia, Knema glaucescens var. cordata (if

in wet ground), K. glaucescens, K. intermedia (if in wet ground),

K. plumulosa, Myristica crassa (a few weak ones), M. elliptica,

M. iners and M. lowiana. Some Malay Islands’ species with stilt

roots are M. argentea, M. fatua and M. speciosa.

Bark Characters

These are often distinct and help considerably in field identi-

fication if one is familiar with them and can remember them. They

are difficult to express in a key or in a drawing as there is so

much detail. A photograph or a specimen is the best means of

reference. All bark characters in this account apply to mature or

well developed trees, since the bark in young trees and saplings

is smooth and under-developed. Young bark will be misleading

as to correct identity since it does not show the fissures and flaking

which are often diagnostic.

The bark of Myristica is usually brittle, blackish, coal black or

greyish black but there are some with a brownish bark, e.g. M.

crassa, M. elliptica, M. guatteriifolia and M. maxima. That of Hors-

fieldia is brown or greyish brown, Knema greyish, greenish grey or

less often a brownish grey while in Gymnacranthera it is reddish

brown. Certain species have a flaky bark. These are Horsfieldia

macrocoma vat. canarioides, H. sucosa (slightly), the three Knema

species furfuracea, hookeriana and latericia (all of which have a

similar bark), K. communis which has dents here and there where

patches of bark have fallen out, K. globularia (slightly flaky), K.

glaucescens, K. mandaharan, K. oblongifolia var. monticola

(slightly), K. stenophylla, Myristica guatteriifolia and M. maxima

(slightly). The bark of Myristica is usually longitudinally furrowed

except in M. elliptica, M. guatteriifolia and M. maxima. It is also

furrowed in Horsfieldia where the furrows or striations may be fine

or coarse. In Gymnacranthera it may be smooth or with a few

faint or shallow furrows. It is not furrowed in Knema.

Kino and Malay Names

This substance, already mentioned, oozes from freshly cut

bark in greater or less quantities according to the individual

species. It is also present in the wood, in the twigs and to a lesser

215

Gardens Bulletin, S.

extent in the petioles and the inflorescence axis. Its presence is

an excellent field test when one suspects that a tree belongs to the

Myristicaceae. The colour varies from dark red to pink; less often

it has an orange tint. It is not so obvious in very young trees.

The amount probably varies in the same species from time to

time. It contains tannin and gum and has left many an indelible

stain on the clothes of plant collectors. Warburg states that the

kino of one species has been used in America as a styptic. Its func-

tion is not known, but it may be that it helps the wounds of a

damaged tree to heal. It has been described as bloody and grue-

some and it is from its very appearance that Malays have aptly

named the Myristicaceae pendarah, chendarah, penarah, darahan,

chendarahan, pendaharan and penarahan, darah being the Malay

word for blood. Other Malay names for Myristicaceae may be

conveniently mentioned here. The nutmeg Myristica fragrans is

called buah pala, while wild nutmegs are pala bukit or pala hutan.

Horsfieldia irya is pianggu and H. crassifolia jangkang paya. Jang-

kang means stilt roots and paya is swamp. In Sarawak and Brunei,

Myristicaceae are generally called kKumpang. Other names balun

ijok, pianggu, penaharan and darah-darah are also used in

Sarawak.

Wood

Wood characters in Myristicaceae are of little importance to the

systematist for distinguishing between genera, but they do show

features peculiar to the family and indicate its relation to other

families. Garratt in a paper, “Systematic Anatomy of the Wood of

the Myristicaceae” in Tropical Woods 35 (1933) 6—48, states that

the wood anatomy supports Warburg’s grounds for dividing the

family into a number of genera, as against the former monogeneric

view of various authors and that, while there is a definite similarity

in the structure of the wood in a number of cases, the distinctions

between most of the genera become clear when the natural geogra-

phical grouping is considered. This is especially true with the Af-

rican and American Myristicaceae, the woods of which appear to

be readily separable into the genera recognized by Warburg. Only

among the Asiatic genera is there any constant overlapping in ana-

tomical features and, even there, Horsfieldia stands apart from the

other three genera in the matter of the simple perforations of the

vessels.

The chief features of the wood are (1) the narrowness of the rays

and (2) the presence in them of tanniniferous tubes which secrete

the red sap or kino. These tubes are not found in any other family.

Freshly cut sap-wood is usually white but soon turns yellowish

216

Vol. XVI. (1958).

brown, stained by the kino which issues from the ruptured tissues;

it is soft and light to moderately hard and heavy. The heart-wood is

darker in colour, often dark red or purplish brown and is heavier,

harder and more durable. Growth rings are usually distinct and

visible to the naked eye. The vessels are mostly medium-sized, few,

solitary or in radial pairs or groups of 3, rarely 5, mostly open, but

sometimes with solid deposits; the perforations are scalariform, ex-

cept in Horsfieldia, where they are simple, as is pointed out above.

The wood parenchyma is moderately abundant and visible to the

naked eye on all surfaces. The rays are 2—3-cells wide, but occa-

sionally 1-cell wide in parts; oil and mucilage cells, as well as the

above mentioned tanniniferous cells, are present in them.

The timbers are of no more than local importance. For several

reasons they are little known in commerce. The trees are not abun-

dant in any one area of forest. The wood is not durable and is soon

subject to attack by fungi, termites and powder-post beetles.

Though easy to saw, the wood splits when nailed. The timber of

A. irya takes on a fine polish and would make attractive furniture;

but the tree is never big and the amount of wood obtained from

each tree would be small.

Twigs

Twigs are of greater diagnostic value and they should not be

neglected when making identifications. Their individual details are

given in the systematic part, but here a few general notes may show

a clearer picture. Thickness, colour and the presence of tomentum

often give helpful clues to identifications. The thickness of a twig

is very striking when one compares extremes such as those of

Knema hookeriana and K. stenophylla or K. curtisii. No one would

ever mistake the thin, slender, glabrous twigs of the last two for the

stout, woolly-tomentose ones of the first. Colour is often useful.

Many are reddish or greyish brown, but the following have pale,

straw-coloured twigs:—#Horsfieldia bracteosa, H. sucosa and Kne-

ma curtisti. In Myristica elliptica, the older parts of the twigs are

straw-coloured, but the younger apices are darker.

Twigs are usually striate in the older parts in Horsfieldia, Knema

and Myristica and this alone distinguishes them from the non-

striate twigs of Gymnacranthera. It is important to have a twig of

sufficient length and not just to rip off a short, terminal portion so

that it may fit easily into a collecting press. The upper part is

usually different from the lower, older part. This is especially so in

Knema, many species of which may be recognized with a little prac-

tice from the twigs alone. It is not so easy, and often not possible

to recognize species of Horsfieldia and Myristica from their twigs.

217

Gardens Bulletin, S.

In some species of Myristica, the bark of the twigs tends to crack

and flake off while other parts of the twig are smoother and more

shining than in Horsfieldia. This flaking is very well seen in Myris-

tica lowiana, M. maingayi and in the Bornean M. villosa. It is also

diagnostic for Knema furfuracea, K. hookeriana, K. latericia and

K. elmeri. In other species the bark is finely striate and the twigs

cannot be distinguished from a Horsfieldia or a Knema.

Certain Horsfieldia species have two lines or ridges on each side

of the twig, stretching from one leaf base to the next. These may not

be clear or present on every single interval from leaf base to leaf

base, but some of the intervals ought to show them if a sufficient

length of twig has been collected. Such ridges are present in H. irya,

H. subglobosa var. brachiata, H. ardisiifolia, H. novo-guineensis and

to a slight extent in H. sylvestris. The following New Guinea species

of Myristica also have these ridges:—M. hollrungii, M. subalu-

lata, M. tristis and M. velutina. They are confined to the

younger parts of the twigs which are swollen here and there and

inhabited by ants. The exact nature of this myrmecosymbiosis is

not understood. The destruction of some of the pith by the ants

may stimulate the growth of the twigs. Myristica costata has four

ridges. In Horsfieldia wallichii, the twigs are hollow in parts, but

there are no such lines or ridges. The apical portions of Gymnac-

ranthera twigs may be angled, but here the angles do not extend

from leaf base to leaf base or in any regular pattern.

Lenticels are found in all four genera. They are most numerous

and conspicuous in certain species of Horsfieldia, especially in those

which have the two ridges on the twigs, though they are also found

in some species without the ridges. Among the latter, the best

examples are H. fulva, H. punctatifolia and H. superba. In fulva

and superba they are pinkish. In Gymnacranthera, they are abun-

dant in G. bancana. In Myristica they are less numerous or very

small, while in Knema they are least conspicuous or absent.

Tomentum

Tomentum is found on young twigs, petioles, the leaves (where

it may be present on both sides when young, later disappearing, or

it may remain on the lower surface), on the inflorescence axis, the

outside of the perianth and often on the fruit. The tomentum exhi-

bits a diversity of form in scales, stellate hairs, branched hairs,

which appear like a growth of algal rhizoids, and less often simple

hairs. These are all variations of sympodially branched, uniseriate

trichomes. They are of diagnostic value in quite a few species, such

as Knema hookeriana, K. plumulosa, Horsfieldia flocculosa and

218

Vol. XVI. (1958).

Myristica villosa. The tomentum in American species, according to

A. C. Smith, Brittonia 2, 5 (1937) 404, seems to be of primary

importance for distinguishing genera.

Leaves

(1) General Characters

The leaves are simple, entire, penninerved and without stipules.

They are alternate and appear to be more or less distichous in

arrangement, but are really spiral in origin. Their vernation in

most cases is convolute, but is conduplicate in Dialyanthera. They

are nearly always petiolate, but the petiole is very short in Virola

sessilis and Horsfieldia sylvestris. It is slightly winged in Horsfieldia

sucosa and Dialyanthera species, conspicuously so in D. latialata

and D. lemannii. The texture of the leaf is usually coriaceous, and

especially in Iryanthera macrophylla and Brochoneura. In many

species of Horsfieldia, the leaves are chartaceous and brittle when

dry and tend to break in herbaria, while in some species of Knema

they are rather membranous, but are never extremely delicate.

The shape varies from oblong to lanceolate or elliptic and is less

cften obovate. There is a considerable range in size too. The largest

leaves among the Malayan species are found i in Horsfieldia superba,

25-70 cm. apne and 10-22 cm. broad, K. hookeriana, 30-65 cm.

cm. broad, K. K. glaucescens va: var. r. cordatg. 33 50 cm. _ long and 15— 17

cm. broad, K. m mandaharan, 30-50 cm. long and 8:-5—14 cm. broad,

and K. furjuracea, “10-50 cm. long and 3—14 cm. broad. The spe-

cies with the smallest leaves are K. stenophylla, 8-17 cm. !ong

and 1-5-4 cm. broad and K. curtisii, 4-5—14-5 cm. long and 1-8-6

cm. broad.

The leaf-base may be acute or rounded or occasionally cordate.

Examples of the Cordate. bas€>are seen in Pycnanthus angolensis,

Virola rugulosa, Knema furfuracea (sometimes only slightly so or

not at times), K. glaucescens var. cordata, K. oblongifolia (typical

form) and Myristica subalulata (slightly cordate at times).

The apex is acute or acuminate, and not very often obtuse.

Examples with an obtuse apex are found in Dialyanthera gordonii-

folia, Horsfieldia crassifolia, H. montana, Myristica gigantea (oc-

casionally bluntly acute), M. schleinitzii and in Knema curtisii,

where both acute and rounded leaves occur.

The colour of fresh leaves is usually dark green above and paler

green beneath. In many species of Myristica, most species of Kne-

ma and in some species of Gymnacranthera, the undersurface of

the leaf is glaucous or even white with a waxy bloom. Hairs are

sometimes present beneath but their presence is not confined to any

219

Gardens Bulletin, S.

one genus. The leaves of the following species are brown beneath

due to rusty scales or very short scale-like hairs: —Gymnacran-

thera bancana, Horsfieldia crassifolia, Myristica cinnamomea (the

scales are very minute and so closely adpressed that the surface

appears to be glabrous), M. guatteriifolia and M. fatua. The lower

surface is sometimes punctate, the best example being Horsfieldia

punctatifolia, where it is covered with black dots, visible to the

naked eye.

The midrib is always distinct and often stout. It may be raised or

sunk on the upper surface, but is always raised on the lower. The

nerves and reticulations show some diversity of characters which

is useful in systematic determinations. Two genera Brochoneura

and Staudtia are distinct from the rest in their venation. In these

the primary nerves fork and anastomose in several loops near the

middle of each side of the leaf or far from the margins. Many secon-

dary nerves are present as well, which also fork and are exactly

like the primary nerves in thickness and prominence. The whole

pattern is further obscured by the loose reticulations, so that it is

difficult to tell a primary nerve from a secondary one. In the com-

mon type of venation, found in the other genera, the primary nerves

anastomose only at the margins or near them, while the secondary

nerves are very few or nearly always absent. They are never so

prominent as the primary ones which usually stand out in relief,

especially on the lower surface. Occasionally they are also faint on

the lower surface. The primary nerves may be parallel to each

other or they may curve and bend gradually from the middle to the

margins. The reticulations may be scalariform between the primary

nerves or they may form a dense network, as in dried specimens of

Knema. Fresh Knema specimens do not usually show them.

(2) Anatomy

One of the most characteristic features of the leaf is the presence

of excretory cells in the mesophyll and these produce the aromatic

oil. They are rare, however, in Gymnacranthera. Tannin is also pre-

sent in elongated tanniniferous sacs in the veins. Acicular crystals

occur in the veins, too, and in the sub-epidermis. Stomata seem to

be confined to the lower surface, which is frequently papillose.

Most species have two layers of palisade cells, but the coriaceous

leaves of Iryanthera macrophylla have three. Reticulate scleren-

chymatous fibres occur in the mesophyll of Gymnacranthera, while

stellate, branched, sclerenchymatous idioblasts have been found in

Iryanthera. A few strands of intra-xylary phloem have been found

in the midrib and petiole of Myristica fragrans.

220

Vol. XVI. (1958).

(3) Systematic value

Leaves are of some value in determining certain genera and spe-

cies in as far as the following general characters are concerned,

namely, size, number of veins, presence of tomentum and whether

the base is cordate or not. A few may have some other special fea-

tures which pertain only to individual species. Such features are

mentioned in the systematic part, only if they are deemed to be of

value. It is disappointing, however, that the leaves of many species

of Myristica from New Guinea, have almost exactly the same shape

and appearance and for this reason, it is almost impossible to dis-

tinguish these species by their leaves alone. A group with small

elliptic or lanceolate leaves resembling M. fragrans is especially

troublesome. Also certain Horsfieldia species with a 2-valved

perianth from the same region have leaves which are scarcely dis-

tinguishable from each other. Sterile material of the following

Malayan Myristica species will also give trouble if insufficient or if

only small lengths of the twigs are collected:——M. crassa, iners,

lowiana and maingayi. Knema is slightly easier, but sterile samples

of K. conferta, laurina and scortechinii are often difficult to distin-

guish. One should, therefore, not rely on leaves alone for identifica-

tions unless very familiar with their details.

Inflorescence

(1) General Remarks and Size

The inflorescence is of great use in separating genera and as far

as the Malayan species are concerned, it stands second in import-

ance after the androecium. The inflorescence is always axillary or

in the axil of a fallen leaf. In Virola sessilis and Myristica philip-

pensis it is terminal, but still in the axil of a leaf. It is cauliflorous

in Iryanthera paradoxa. The axis is usually round in cross-section,

but flattened in Virola sessilis, some species of Horsfieldia and in

Myristica maxima. The male inflorescence is generally longer and

more richly branched than the female, especially in Horsfieldia and

Myristica, but not in Knema. However, in Horsfieldia macrocoma

var. canarioides, the female is even longer than the male, 16—23

cm. as against 6-10 cm. The longest inflorescences (male) occur

in Horsfieldia and Virola, with those of H. wallichii up to 33 cm.

long, H. grandis 12-25 cm., H. flocculosa 14—20 cm. and H. brac-

teosa up to 20 cm. Several Virola species have them up to 25 cm.

long, but in V. calophylla they reach 30 cm. The smallest infiores-

cences are found in Knema, where the axis may be only 1 mm. long

221

Gardens Bulletin, S.

when it first starts to produce flowers. The woody tubercle-like in-

florescence of Knema is persistent, producing flowers from time to

time, gradually elongating and eventually reaching a maximum of

about 1:7 cm. after a long time. In one species of Brochoneura, B.

chapelieri, the inflorescence is almost sessile.

(2) Branching

The general type of inflorescence is a panicle. In Horsfieldia and

Virola it is richly branched. Ramifications of the third order are

common in both and in Virola calophylla and rugulosa they are of

the fifth order. The flowers are crowded at the ends of these bran-

ches in umbels, sub-umbels, cymes, capitula, (the capitula rarely

confluent), catkin-like structures, or less often racemes or very

condensed racemes approaching umbels as in Knema and some

species of Myristica. Sometimes there may be more than one inflo-

rescence in the axil of a leaf, as in Compsoneura (1-2), Dialvan-

thera (1-2), Iryanthera (1-3) and Osteophloeum (1-3). The

types of inflorescence are best shown by a table.

GENUS TYPE OF INFLORESCENCE

COMPSONEURA A main axis with the flowers distantly

fascicled or racemose at the ends of its

short branches. 1-2 inflorescences in the

axil of a leaf.

| DIALYANTHERA. A main axis with the flowers fascicled on

very short lateral branches or the lateral

branches rudimentary. 1-2 inflores-

cences in the axil of a leaf.

IRYANTHERA Similar to Compsoneura. 1-3 inflores-

cences in the axil of a leaf.

OSTEOPHLOEUM A main axis with a few short branches

(fewer than in the above three genera),

the flowers sub-umbellate at the ends of

the short branches. 1-3 inflorescences

in the axil of a leaf.

AMERICAN GENERA

VIROLA (1) A much branched, spreading panicle

with the flowers fascicled at the ends of

the ultimate branches. Never double.

(2) A comparatively simple main axis,

little branched.

wecens. <a eee i ny

222

Vol. XVI. (1958).

AFRICAN GENERA

ASIATIC GENERA

GENUS

“BROCHONEURA

‘CEPHALOSPHAERA

“COELOCARYON

PYCNANTHUS

SCYPHOCEPHALIUM

STAUDTIA

GYMNACRANTHERA

HORSFIELDIA

KNEMA

MYRISTICA

TYPE OF INFLORESCENCE

A main axis, the branches catkin-like with

the flowers sessile or nearly so, in dense

capitula, the capitula often confluent.

Flowers monoecious, the male and female

mixed in the same inflorescence or in

different inflorescences.

Paniculate with the flowers distant from

each other in large capitula, the capitula.

stalked or sessile, not confluent.

Paniculate or racemose, the individual

receptacles disciform, stalked or sessile

with the flowers stalked. The inflores-

cence may be reduced to a single iong-

stalked receptacle.

A spreading panicle, the flowers in capitula

at the ends of the ultimate branches, the

capitula on a separate, short peduncle,.

quite separate from each other.

A short thick main axis, unbranched or 2-3

branched, each branch ending in a

spherical umbel with the flowers stalked

and densely packed together (false capi-

tulum).

Inflorescence a spherical capitulum, sessile

or nearly so.

Male a much branched panicle; female

shorter and more contracted. Both with

flowers on the ultimate branches in umbels.

or racemes or a mixture of both.

A much branched panicle, the female

usually shorter and less branched except in

H. canarioides. The flowers stalked,

rarely sessile, in umbels, cymes, racemes.

or combinations of these at the ends of the

ultimate branches. In H. iryaghedhi the:

flowers are in dense, sessile capitula.

Axis a short woody tubercle, 1 mm.—1.7 cm..

long with the scars of previous pedicels:

and bracts, unbranched or occasionally

2-3-fid. Flowers in umbels or contracted

racemes at the ends of these protuberan-

ces.

(1) Asin Knema

(2) A main axis branching di-or tricho-

tomously some distance from its base.

(3) A mixture of (1) and (2) the main

axis smooth, the branches woody and

thickened as in Knema with scars.

Flowers in umbels or racemose umbels.

at the ends of the ultimate branches.

223

Gardens Bulletin, S.

(2) Bracts and Bracteoles

The presence or absence of bracts and bracteoles is also useful

in separating some genera. From the table it will be seen that bracts,

although sometimes caducous, are present in all genera, and that

bracteoles are nearly always absent, except in Jryanthera, Knema,

Myristica and Osteophloeum.

+ = present; O = absent

GENUS BRACTS BRACTEOLES

Compsoneura + caducous 0

Dialyanthera = 0 Sometimes rudimentary

on female pedicels.

Iryanthera + caducous a.

Osteophloeum + caducous, very

small ac

Virola + caducous

Brochoneura —- 0

Cephalosphaera ae 0

Coelocaryon + 0

Pycnanthus + 0

Scyphocephalium + 0

Staudtia -|- 0

Gymnacranthera a 0

Horsfieldia + caducous 0

Knema | + caducous zy

Myristica | + y 4

Not much else may be said about the bracts. They are leaf-like

and usually quite small, the largest being those of Horsfieldia floc-

culosa, which are 1:5—1-7 cm. long.

The bracteoles are smaller than the bracts. Their position on the

pedicel in the genus Knema is often a useful clue to the identity of .

224

Vol. XVI. (1958).

the species. (See table under Knema in the systematic part). The

ciliate bracteole in Myristica malaccensis is one of the best diagnos-

tic guides for determining that species.

(3) Pedicels

The pedicels vary in length and thickness or may be entirely ab-

sent, especially in the female flowers of some species. When pre-

sent they are usually shorter in female flowers than in male. They

are of use in systematic determinations, especially in Knema,

where their comparative lengths are of value. See table under

Knema in the systematic part.

Flowers

(1) Sex

The flowers are nearly always dioecious, except in Brochoneura

and Iryanthera. In Brochoneura they are monoecious, where the

male and female may be mixed in the same inflorescence or on

separate branches of the inflorescence or in separate inflorescences

on the same tree. In Jryanthera they may be dioecious or monoeci-

ous or with male and female mixed in the same inflorescence. King

found some rudimentary stamens in the female flowers of Myristica

suavis, now reduced to M. crassa. I have found male and female

flowers in the same inflorescence of M. crassa collected in Singa-

pore. Other anomalies occur in M. fragrans, which is normally

dioecious. Male trees have been known to produce a few female

flowers at first as well as male and later on all female flowers. In

the Botanic Gardens, Singapore, there is a tree which produces

fruit; the majority of its flowers are female and a few are male.

Horsfieldia macrocoma var. canarioides is monoecious. I suspect

that it may be dioecious as well, but I have never seen a living tree

of this species and it seems to be extinct in Singapore now.

(2) Colour

The perianth of Horsfieldia and Gymnacranthera species is gene-

rally yellow, that of Myristica being of a paler yellow or at times

almost white. In H. fulva it is orange and in G. bancana golden

yellow with a brownish tinge. The outside of the perianth may be

glabrous or pubescent to tomentose, but the inside is usually glabr-

ous and often has some drops of glistening nectar. If the outside is

glabrous then the whole perianth will be uniform in colour as in

Myristica elliptica and M. fragrans, where it is cream, but if hairs

are present as in Knema and certain species of Myristica, then it

will be of a brownish colour outside and some other shade inside,

220

Gardens Bulletin, S.

such as red, pink or cream. The flowers of Virola cuspidata are

orange and those of Iryanthera sagotiana reddish-yellow. The

colours observed in some Knema species are given here.

eee

KNEMA SPECIES COLOUR OF PERIANTH

K. communis Pink

K. conferta Yellow with a greenish tinge and a

brownish pink spot at the base of

each lobe

K. furfuracea Bright red

K. glaucescens Rose red

var. cordata Greenish white

f. rubens Pink

K. globularia Cream

K. hookeriana Bright red :

K. intermedia Cream

K. kunstleri Red

K. latericia Pink

K. laurina Red

K. malayana Cream

K. plumulosa Cream or pale pink to greenish

K. scortechinii Cream

Here it may be noted that the red colour of the perianth of K.

laurina would distinguish it from that of K. conferta, with which

it is often confused.

(3) Scent |

The flowers of many species are scentless, but some of these may

have a scent when crushed, e.g. Gymnacranthera bancana, while

G. forbesii has a faint scent resembling lemons. Most Horsfieldia

species have a powerful penetrating, sweet odour. There are still no

records of scent for several of the rarer Malayan Horsfieldia spe-

cies, but the following have been noted and all have a strong, sweet

scent:—H. bivalvis, H. bracteosa, H. crassifolia, H. irya and H.

superba. H. polyspherula has a faint scent, but H. wallichii is scent-

less. The flowers of the following Myristica species have a sweet

odour:—M. crassa, M. fragrans, M. lowiana (when crushed), M.

maingayi and M. maxima. The flowers of Knema are mostly all

scentless and collectors have said little or nothing about their odour.

I find that these of K. malayana have a sweet odour and those of

K. communis sweet only when crushed. I have personally tested

for scent the flowers of all these examples given here.

226

Vol. XVI. (1958).

(4) Shape

The perianth, representing sepals, is normally divided at the apex

into 3 teeth or lobes, but in one group of Horsfieldia species it is

2-lobed or 2-valved. Species with 3 teeth in the normal flower may

occasionally have a flower or two with 2-4 or 5 teeth and likewise

the bi-valved species a flower with 3 teeth in the same inflorescence.

The lobes of the perianth may, on opening, spread horizontally as

in Knema or become reflexed as in Myristica. However, where the

lobes are very short, as in most species of Horsfieldia, they will

not alter their position very much, the flower remaining globose.

In Horsfieldia macrocoma var. canarioides, however, the lobes

reach down nearly to the base of the flower and expand widely and

may be reflexed. In most Knema and Myristica species the perianth

is split down for 1/3 to 1/2 of its length to form the lobes. Various

shapes of perianth are met with. In many of the Horsfieldia species

with the bi-valved perianth, the flower is laterally compressed and

looks like a cockle shell, but other species of Horsfieldia have a

globose perianth. The perianth of Knema in bud is globose, sub-

globose or slightly 3-angled. The different shapes may be noted

here.

Globose—Brochoneura, Horsfieldia, Knema (often three

angled).

Flask-shaped—Gymnacranthera, Myristica (some species),

Osteophloeum.

Funnel-shaped—Dialyanthera (or cup-shaped or rotate)

Compsoneura, Virola, Iryanthera (perianth cleft nearly

to the base in one section).

Clavate—Pycnanthus, Horsfieldia iryaghedhi.

Conical like a skittle or cylindrical—Myristica (some species).

(5) Size

The flowers are small and for this reason foresters unfortunately

do not bother much about them. Some use is made of their com-

parative size in systematic keys as this is a useful character at times.

Dried flowers in herbaria are slightly smaller than fresh ones and a

shrinkage of 0-5 mm. should be allowed for in measurements. Cau-

tion is necessary in considering measurements of flowers from her-

barium sheets as very often these flowers may be immature. This is

especially true in Horsfieldia, since the mature ones open very

slightly at the top and the immature will look like them if they

have been squashed in the pressing. The flowers of Knema may

remain closed for a long time on the tree and then expand a bit

227

Gardens Bulletin, S.

more before eventually opening (six weeks in the case of K. inter-

media where every day they looked as if they might open). Except

for some Horsfieldia species, the American genera usually have

smaller flowers than the Asiatic. The African genera have still

smaller flowers, the largest being those of Scyphocephalium (3

mm.).

The largest flowers of all are found in Knema hookeriana. The

male is 1 cm. long and the female 1-7-2 cm. long. Other large

Knema flowers are those of K. plumulosa, the male 6—7 mm. long

and the female 1 cm. long and K. mandaharan, female 1 cm. long.

The largest flowers in Myristica, the female of M. neglecta, are 1-1

cm. long, while in M. maxima and philippensis the male are 5—8

mm. long and the female 8-9 mm. long. In Gymnacranthera the

largest are in G. bancana—the male 6 mm. long. In Horsfizidia

those of H. superba attain the greatest length, the male 7-8 mm.

long and the female 5 mm.—1 cm. long. The smallest flowers of all

are found in H. irya, the male being 0-5 mm—1 mm. long. Other

species of Horsfieldia with small flowers are H. crassifolia, male 1—

1-5 mm. long, H. polyspherula, male 1 mm. long, H. subglobosa,

male 1—2 mm. long and H. ridleyana, male 1 mm. long. The female

flowers of these Horsfieldia species are slightly larger.

Androecium

(1) General Remarks

The androecium differs very much in shape and character, more

so than in any other single organ in the whole family. Consequently

it provides some excellent systematic characters by which most

genera and many species may be distinguished. Warburg used it

for separating the American genera, but the knowledge gathered

from new specimens has shown it to be an unsatisfactory character

for these American genera. A. C. Smith points out that genera

which, according to Warburg, had only one type of androecium,

now have others. It is, however, still of prime importance as far

as the Asiatic and African genera are concerned.

Unfortunately the androecium is small since the flowers are

small. It conforms to the dimensions and shape of the flowers.

If we examine its morphology for the Asiatic genera alone we

shall get a very one sided view of the family as a whole and shall

not be able to see its development and diversity. This may be

grasped more conveniently, quickly and clearly from a table than

from many pages of description. However, I shall not go into any

great details about African and American genera in this revision,

which is intended primarily to deal with those in Malaya.

228

Vol. XVI. (1958).

‘soseo AUPUT Ul

sioyjue oy} JO uoluN

Opis dy} 99s 0} IMIG

*xode o[119}s 1104s

e YM osje uwnjod

*USUL

-yoeye jo yurod oy}

UIOIJ SUISIOAIP S1OyjUY | -SUO] OY) UY} 19,10Ys SIOyUY

‘TeoruodqgoO uUUUNTOD | AIOA Y[eIS “IOsUOT sIOyU'Y

‘adeys Ul ayed[vj sIOyIUY | Y]eIS OY} UY) IOBUO] sIOyUY

SYIVWDY

“UUUN[OO

oY} JO yIeIS 94} 0} [enboa Io

UY} JOBUO] 10 19)10Ys SIOYUY

‘ULUN[OS oY} JO YI[eIS

JOYS OY} ULY} 1OdUO] SIOYUY

‘UNOS 9Y} JO Y[eIS 94}

UY] 1OBUO] 10 19}1OYS SIOYIUY

“ULUNTOS Poy]e}s

‘y10Ys

“ULUNTOS OY} JO

I 07 sroyUYy

jo yysuoT [vuonsodoig

pure uWIN{OD 9Y} JO IVS

‘Ayjensn ¢

(9-7)

‘SopIs 119y} Aq 19Y}O Yous 0}

A]VBI[s 10 JOU pue syoeq IY}

Aq UUUIN[OS 9Y} 0} 9}vUpK SIOyUY

“SOpIs JOY}

Aq 19410 YORd 07 puv syxovq I19Y}

vI-cl

Aq UWIN[OS sy} 0} d}vUpPR SIOyJUY

‘poxyiseq pue do1j sioyjuy (Z)

"Oly

A[IV9U SOpIs I19Y} “syoeq I19yY} Aq

p-€ ULUNTOS dy} 0} dJeUpe sJoyUY (1)

*xodv sj je 901 ING ULUNTOS

oy} 0} 9}eUUIOS AT] VSIOP sorIdods

€ uO UI JO poxyiseq ‘soly sisyUY

‘SOpIs I19Y} 1”

IOY}O YORI WIOIJ 9dIJ SIOYJUR 9Y}

JO ouIOs ‘UWINTOS 9Y} 0} SyoRq

6-S Jay} Aq poyse}se “oly JOU SIOyJUY

—!9]2JSOIUOD UOTWDES

‘uUINTOS oY}

0} 9}yeUUI0S AjoIvI IO (poxyiseq)

UWINTOD dy} 0} Soseq SWI9I}xO

OI-v

sIoyUy

Joquinyy

Jaq) Aq poyovyye ‘oly sioyjUVy

—pANaUosduoINnyT UWO0Wd9S

UWUNT[OD 94} 0}

s19yUY JO UOIUG, JO s0130q

VIOUIA

WNdO THdOALSO

VUAHLINVAUI

VUAHLINVATVIG

VUNANOSdNOOD

snuoy

229

Gardens Bulletin, S.

“UUUNOS oY} JO ¥[eIS OY}

uey} Jay10Ys A[JYSI[s s1oyjUy OI-9

‘UWUNTOS oY} JO

YIe}S OY} UvYy} Josuo] sisyUY pre

‘yjuelIod 94}

jO yNOUW 94} puossq

“UUWUNT[OS 94} JO

suipnyjoid wmniss01puy

y[eIs OY} URY} 19}10Ys sIOyUYW p-Z

‘UUINTOS IopUusTsS J o-€

mbieetal (ere)

94} JO W[VIS 94} ULY} 19}10Ys

spy] B IO 0} Jenbs sroyjUYy S-€

“SOPIS OY} 1¥ ddIj JOU ‘syoeq IE}

Aq uuIN]OS sy} 0} s}eUpe sIOyUW WNITTVHddOOHdADS

"syorq I10y}

Aq UUWN][OS 94} 0} d}eUpe sIOyjUYy VILGNV.LS

‘sopre oy} 1¥ 9o1j “SyoRg I19y}

Aq WuIN[OS 9Y} 0} BeUpe sIOyUYy SONHINVNOAd

*syorq I10y}

Aq WUIN[OS 9Y} 0} d}eUpR sIOyjUYy NOAYWVOOTAOD

"syoeq 1104}

Aq UUINTOS 9Y} 0} dJeUpR sIOyUY VudVHdSOTVHdH)o

| he —

‘plyajnojnvy .

snuos s,dinqieAA (Z) Or-0Z

‘yed snoydjepouour

94} St YISUZ] OUIKS OY} JNOGe

IO UUIN[OS oY} JO ARIS 9Y}

uvy} Josuo] ApWYSI[s sIoyjUY 07-9

HI 0} sIOyVUYy sIoyjUy

SYIVUIDY

JO yVSusT [euonsodolg jo

pure uuINnj[oD 94} jo yes

JoqunN,

‘(sorsods 7) s}uoweyTy 310Ys

0JUI poprArIp 1oddn oy} ‘snoydyjep

-OUOU! ULUN[OS 9y} JO 11ed IOMO]

oy} ‘poxyiseq ‘ooly sioyjUYy (Z)

“soprIs I19y} Aq

I9YIO YORd 0} puv syoeq I19q} Aq

ULUNTOS 9Y} 0} aJvUpe sIOyIUY (1) VUNANOHOOWA

—

uWIN,OD sy} 07

sloyjuy JO UOIUA JO 901390q snuepy

230

Vol. XVI. (1958).

. : —

*xodv 9[L19}s “Sopts 1194} Aq 1940

9snjqo IO 93nde Uv UI ‘"}1.0} jenbo Io uWNyOS 9y} Jo Youd 0} poyoryje pure syoeq 1194}

SPUD SOLUTJOWOS ULUN[OD | AVIS oY} ULY} JoZUO] s1OyUY O£-8 Aq UWWNjOS 9Y} 0} 9}vUpPe sIOyJUY VOLLSIIAW

‘JoyjUe

oy} JO opis JOMOT

uo soeg ‘poxoyop “uuUN[OS 1181JANI

Io onbijqo Ay YsI{s 94} JO AJVIS oY} ULY} J9}10Ys ‘y ul

SOUUNJOUIOS “| VOTJAOA Io 0} [enbo ‘1osu0], sioyjUYy Sp-SE ‘OSIP Poyeys Iepnsuei4s}

JoAou ‘Te}UOZIIOY ‘oyIS ‘pUuDIAayYOOY “YY Ul BSUOT IO 93¥)[9d B 0} A[o}eTI9IS soseq

-Sos IO iia sa Veale el sioyjuy | “WW p “SUOT “WU Z—|[ YI[VIS €7-9 Jisyy Aq poyoryje jnq so1j s1oyUY VWANY

‘posure

‘(SpreMuMOp xode WOIJ) 99}

-SIP JOYS & IOJ SOPIS OY} Je pue

sooide I19Y} 12 901] “SYOVQ 1194} Aq

UWUNTOS dy} 0} dJeUpe sIOyUY (Z)

‘pojsue-¢ yOu uUUIN]OO

OY} JO WOT}DIS-ssOIDQ) + “SOpIs 9Y}

ye do1J A[YSI[S 10 Sopis 1194} 18

‘dnd oy} 0VUI

yoeq BuIpusq s19y}

-¢ UUWIN]OS JO wUWOTdIS-ssoID

-ue oY} YIM podeys

-dnd 10 poyesuoje ‘poyyeys Apsoys AIDA I9qy}0 Yovd 0} pue sydeq 1194) Aq

‘ssogoys §=owiniss0i1puy | A[[euOrsedd0 JO ojIssos ULUINTOD O0£-9 UNOS 9Y} 0} dyeUpe sIOyVUYV (1) VIC THIASUOH

*‘1OYIO YORO WIOIJ SOPIs 9Y} Jv puvL

‘soorde oy} *‘SuO] “UW UWUN[OS oY} JO xodv 9Y} 3v 99IJ

ye PosAINSUI sioyjUYy ¢’o Ajuo yuosoid jy ‘“yIeIS “ABM J[eY SVs] 7 IOJ SyOuq IIoy}

*jeortpury Ao uuunyo> | Aue A[Q0189S IO aISSOS ULUNTOD ZI-9 Aq uuwInjoo oy} 0} s}eupe sioyUY VUAHLNVUOVNWAD

}T 0} S1OIUY stoyuy uUUN{OD oY} 07

SYIVUIDY — - Jo yBueT yeuonsodoig jo sIoyjUY JO UOIUL Jo ceIS0q snusy

pue uunjod oy JO ¥IVIS JOquINN

231

Gardens Bulletin, S.

(2) The Column and the Method of Attachment of the Anthers

The extrorse, 2-celled anthers are generally attached to a longer

or shorter column by their backs and this arrangement is found in

most genera. They may also be attached to each other by their sides

either fully or partially. If their sides are free it is easy to count the

number of anthers. If not, counting may be difficult, especially

where the anthers are numerous as in some species of Horsfieldia;

here during the anthesis one may mistake one split anther for two.

Sometimes the column may have a short sterile apex as in Osteo-

phloeum and in some species of Myristica. Its apex may be acute,

obtuse or truncate. The column may be sessile or may continue

below the anthers as a stalk. The stalk may be much longer than

the anthers as in Dialyanthera, Pycnanthus and some species of

Iryanthera, or equal in length to the anthers as in some species of

Virola and Myristica, or shorter as in Compsoneura, Osteoph-

Joeum, and in some species of Brochoneura and of Myristica. The

sessile androecium is found in most species of Horsfieldia and Gym-

nacranthera. Once again, the monodelphous column to which the

anthers are completely attached by their backs is the general pat-

tern in the family. Exceptions to this mode of attachment will now

be briefly mentioned.

Free, basifixed anthers are oe in Compsoneura, section Eu-

compsoneura, Dialyanthera (except in one species), Iryanthera (a

few species) and Knema. In Knema the column ends in a peltate

or triangular disc with stalked or sessile anthers arranged stellately

round it; they are usually horizontal or slightly deflexed at maturity

or sub-erect, but never erect, and the pollen sacs are on their lower

surfaces. In Brochoneura two species deviate from all the other

genera in having a short monodelphous column ending in free fila-

ments with basifixed anthers. One of these species was formerly

placed by Warburg in the genus Mauloutchia (M. chapelieri). In

the other species of Brochoneura, we have the normal Myristica-

type of column.

In Gymnacranthera the anthers are adnate to the column by

their backs for a part of their length, but are free at the apices and

at their sides. Their apices are slightly incurved. In some Hors-

fieldia species the anthers show various degrees of freedom from

the column, but they are never completely free.

(3) Number of Anthers

The number of the anthers is often quite useful in systematic

diagnosis. The smallest number is two in some species of Virola

and Pycnanthus, while the maximum number is found in Knema

curtisii, 35-45. The numbers in other species of Knema are given

232

Vol. XVI. (1958).

in the table in the systematic part. It will be obvious that the num-

ber of anthers is useful for separating closely related species only

when there is a considerable difference between the numbers in

their anther counts. Thus if one species has 6—10 anthers and the

other 7—9, this numerical classification would not serve to distin-

guish between them; if on the other hand, the numbers were 7—9

and 15-18, the classification would be of value since there is no

overlapping of numbers. The table further shows that species do

not tend to have any constant number of anthers, but that there is

a variance of 2~3 above or below the average.

(4) Pollen Grains

The pollen grains are more or less similar to those in Anno-

naceae, Canellaceae and Magnoliaceae. R. P. Wodehouse has done

some work on the pollen of the American species—Smith and

Wodehouse “The American Species of Myristicaceae” Brittonia 2,

5 (1937) 397. Little is known about those of the African or the

Asiatic species. Often pollen morphology among the angiosperms

is of taxonomic value, both as to classification of the smaller groups

within the families and to the relationships of the families to each

other.

The grain is one-furrowed, elongated, globular or boat-shaped.

The exine is moderately pitted, reticulate or granular throughout,

or with these markings at least on part of the dorsal side of the

grain. The grain with the single furrow is considered to be primi-

tive and it is found in primitive families such as Magnoliaceae,

Annonaceae, Nymphaeaceae, Saururaceae and Piperaceae. In less.

primitive families, the groove tends to become modified or elimi-

nated. The Myristicaceous grain is less related to that of Laura-

ceae where the groove is continuous round the entire grain. Of the

American genera, Dialyanthera seems to be the most primitive

in the development of this groove and the other genera follow in

ascending order, Virola, Osteophloeum, Compsoneura and Iryan-

thera.

Ovary, Style and Stigma

Unlike the androecium, the gynoecium is so uniform and varies

so little as to be of no use in classification. The ovary itself is glo-

bose, sub-globose or ovoid and often covered with hairs or tomen-

tum of some kind. It is generally 1-2 mm. in diameter, but 4—6

mm. in Knema hookeriana and K. mandaharan. Female flowers of

Myristicaceae can usually be recognized without opening them, due

to the globose ovary which takes up more room laterally than the

androecium and so the flowers are larger and more swollen. As was.

233

Gardens Bulletin, S.

stated earlier, female flowers usually have shorter pedicels than the

male or they may be sessile.

The superior ovary consists of a single, erect, anatropous ovule

with two integuments and is placed near the base of the ovary or

a little distance above it.

The style is usually absent or very short, while the stigmas are

very small, usually bi-lobed or connate with a groove indicating

two lobes. Only in Knema is there any significant variation, where

it may be a peltate, sessile, slightly bi-lobed disc with numerous or

few laciniations at the edge (c.f. the peltate androecium of the

male flower), or simply bi- or tri-lobed, in which case there may be

a short style. If bi-lobed, each lobe may aga be shortly bifurcate

so that there are four lobes in all.

Fruit

The fruit is of great use in the classification of different species,

but of limited use for genera. There is much variation in shape and

size and in the nature of the tomentum. The fruits themselves are

attractive by their orange, yellow, reddish-yellow or rusty-brown

pericarp, while their arils are usually some shade of red or orange-

red or if unripe, white. They dehisce, splitting into two valves by a

longitudinal, circumferential ridge or furrow of ‘dehiscence. This is

usually prominent on the pericarp. Often the suture may ‘be a ridge

on one side of the fruit but continuing down the opposite side as a

furrow. In Horsfieldia wallichii, the dehiscence of which I have

personally observed, there is a weak spot at the base of the fruit on

one side. The splitting starts here and proceeds slowly up to the

middle. This may take up to five hours. Then the splitting continues

from here up to the apex and down to the middle on the other side

in a very short time, five minutes or less. The splitting at the actual

apical part may take only seconds. There is then a slowing down

“once more from the middle to the base. Scyphocephalium has a

very thick succulent pericarp, which does not dehisce.

The shape of the pericarp is very diverse. It may be oblong, ellip-

tic, globose, sub-globose, ovoid, obovate or lanceolate. The

apex may be acute, obtuse, or as in Knema retusa shortly beaked,

or tapering into a long beak and at the same time narrowed simi-

larly at the base as in Myristica tubiflora from New Guinea. In all

genera except two, Osteophloeum and Iryanthera, the length of the

fruit is greater or equal to the breadth. In these two genera the

fruit is transversely elongated but some species of Iryanthera may

have a globose fruit.

Among the Malayan Myristicaceae, some of the largest fruits

.are found in the genus Myristica, but some New Guinea species of

234

Vol. XVI. (1958).

this genus have quite small fruits. M. maingayi has fruits up to 10-5

cm. long and 6—6:5 cm. in diameter. Other large ones are M.

maxima, 7—9 cm. long, M. lowiana, 6—8 cm. long, M. cinnamomea,

6—9 cm. long, and M. fragrans, 6—9 cm. long. Of the Malayan Hors-

fieldia species, the largest fruits are those of H. superba, 7-9 cm.

long and 5-5—6-8 cm. in diameter and H. punctatifolia 6-9 cm.

long and 6 cm. in diameter. The largest fruit in Knema is that of

K. hookeriana, 4-5-8 cm. long and 3-4-5 cm. broad. Most Knema

fruits are much smaller, the smallest being about 1 cm. in diameter.

The pericarp is usually thin in Knema and in some species of Hors-

fieldia. Those with the thickest pericarp are also the ones men-

tioned above because of their maximum length and diameter. Their

pericarp is 1—2:5 cm. thick but that of K. hookeriana is a little

thinner, 8 mm.—1 cm. thick. The pericarp is firm or leathery and

often fleshy but never soft and pulpy like that of a tomato.

The majority of the American species have glabrous fruits. The

African and Asiatic ones may be glabrous or tomentose. Tomen-

tose fruits are very characteristic of Knema and especially striking

is the pale-buff, lanose fruit of K. hookeriana which can never be

mistaken for any other. Entirely glabrous fruits are very rare in

Knema. Those of K. curtisii and some forms of K. glaucescens are

tomentulose when young but eventually become glabrous when

ripe or dried. Practically all Horsfieldia species on the other hand

have glabrous fruits but these may have powdery or rusty scales

when young. The majority of Myristica fruits are glabrous, but

quite a few are densely tomentose. In Malaya that of M. lowiana is

a good example of a tomentose fruit.

The Aril

The aril is a tough, but fleshy covering surrounding the seed and

attached to the basal part of it. Its function has already been dis-

cussed under the heading “Habitat and Distribution”. Whether the

aril is entire or laciniate, is an important and useful character in

the classification of genera. It covers the seed in nearly all cases

except in Brochoneura chapelieri (Mauloutchia) where it is pre-

sent only round the basal portion of the seed; in other species of

Brochoneura it is complete but laciniate. In Knema retusa it is also

present only at the basal portion of the seed, but the condition

there is probably pathological as the plant is known from one

gathering only. More material might show it to be like that of other

Knema species. In some genera it is laciniate to the base or nearly

to the base; in others it is laciniate at the apex only or completely

entire. The epidermal surface may be smooth or minutely striate

as seen under a lens. There are three views on the origin of the aril.

235

Gardens Bulletin, S.

(1) It may be an aril proper, originating from the funiculus at or

just below the hilum, (2) an arillode, arising from around the

micropyle or (3) from both hilum and micropyle. The latter is

probably the correct view.

A list of genera with Jaciniate or non-laciniate arils is now given.

ARIL LACINIATE ARIL NON-LACINIATE

Dialyanthera Compsoneura (or slightly laciniate

Virola (laciniate to the base or in at the apex)

some species at the apex) Iryanthera (or incompletely laci-

Brochoneura niate at the apex

Cephalosphaera Osteophloeum

Coelocaryon Scyphocephalium

Pycnanthus Staudtia (complete or sometimes

Gymnacranthera not reaching to the apex)

Knema (laciniate at the apex only) Horsfieldia

Mpyristica

In the following genera the aril is striolate under a lens:—

Virola, Coelocaryon, Pycnanthus, Gymnacranthera, Knema and

Myristica. It is not striolate in Compsoneura, Iryanthera, Scypho-

cephalium, Staudtia and Horsfieldia.

The Seed

(1) Testa

The seed is of various shapes; oblong, ovate, globose, ovate-

lanceolate, elliptic-lanceolate, or in Osteophloeum and some spe-

cies of Iryanthera transversely elliptic. It varies in length from 1—5

-cm. and in breadth from 5 mm.—2:5 cm. The testa consists of three

layers, the outer one is thin, shining, fleshy or membranous. It is

rugose in Iryanthera. It may have shrivelled up and disappeared in

seeds from herbarium specimens. The middle layer is thick, hard

and lignified, while the inner is membranous.

(2) Endosperm

The endosperm or albumen is hard and nearly always contains

fat and a fixed oil. This oil is absent in Compsoneura or only pre-

‘sent in very minute quantities. Starch is present in quite a number

of genera and is very abundant in Myristica. The genera in which

it is absent are:—Dialyanthera, Virola, Coelocaryon, Pycnanthus,

Scyphocephalium Gymnacranthera and Horsfieldia.

The nutmeg, Myristica fragrans, contains a large percentage of

fat, about a fourth of its weight. This is the “nutmeg butter” of

commerce or myristin. Its volatile or essential oil, 3—8 per cent,

contains an aromatic narcotic known as myristicin which is poison-

ous and has caused some deaths. There is also a fixed oil present.

236

Vol. XVI. (1958).

The endosperm also contains protein crystals and a few tannin

cells.

The endosperm is ruminate in the majority of genera, but not in

the following and some use has been made of its ruminate character

in classification:—Compsoneura, Iryanthera (or at times very

slightly), Brochoneura, Cephalosphaera and Staudtia. There are

no details for Osteophloeum. The endosperm is solid, except in

Horsfieldia irya and Coelocaryon, where it is hollow in the centre.

(3) Embryo

The very small, straight embryo is nearly basal or slightly above

the base and lies above the micropyle. Its radicle is minute, inferior

and cylindric. There is some considerable diversity in the shape and

position of the cotyledons in the different genera. Their two lobes

may be sub-erect, as in Virola, Brochoneura, Cephalosphaera,

Coelocaryon, Pycnanthus, Staudtia and in some species of Knema.

In other species of Knema and in Compsoneura, Dialyanthera, Iry-

anthera, Scyphocephalium, Gymnacranthera, Horsfieldia and My-

ristica they are divaricate or horizontal. There are no details for

Osteophloeum. In some genera they are connate at the base and in

others not or very scarcely connate. The genera in which they are

not or scarcely connate are:—Virola, Coelocaryon, Pycnanthus

and Knema. The connate cotyledons may be united into a peltate

disc, as in some species of Horsfieldia and Myristica and Dialyan-

thera. This disc may be flat or in the form of a disc or cup with the

edges smooth or undulate.

HISTORY OF NUTMEGS

To-day the world’s supply of nutmegs and mace comes from

the West Indies, principally from Grenada, but there were once

flourishing plantations in Singapore, Malacca and Penang. The

original home of the nutmeg, Myristica fragrans, is Banda and

Amboina in the Moluccas, and it was cultivated there by the

Portuguese in 1512. In the early seventeenth century, the Dutch

drove them out and held a monopoly of the nutmeg trade. They

endeavoured to limit the trees to Banda and Amboina, but fruit

pigeons carried the seeds to neighbouring islands. In 1769 the

French succeeded in introducing the tree into Mauritius. In 1796

Christopher Smith, who sailed with Captain Bligh on the “Pro-

vidence”’, collected seeds from the Moluccas. These produced fruit

for the first time in Penang in 1802. By 1836 the trees were in

full production and the nuts and mace were superior to those

grown by the Dutch. Unfortunately in 1866 many trees were killed

237

Gardens Bulletin, S.

off, probably due to the attack of a ‘powder post’ beetle and at

the present day home-grown nutmegs are no longer exported from

Malaya. A few trees are still to be seen here and there and local

people use them for their own domestic purposes. They may still

be found in Penang on hillsides and places remote from houses.

A newcomer, unacquainted with the local flora would probably

think that the tree was native. No doubt many of these trees are

self-seeded. Actual figures for the acreages of nutmegs under culti-

vation taken from the Annual Reports of the Department of

Agriculture for the Federation of Malaya in the years 1946-1955

are as follows:—

1946..M a 1951) ae

1947: yar ood 19522 1s eae

1948" FBT 1953 tions

1949 .. 70 1954:

19507 oars 19552" eee

USES OF NUTMEGS

Two oils are obtained from the seeds of Myristica fragrans, the

essential or volatile oil obtained by distillation and the fixed or

expressed oil obtained by subjecting the nutmegs to pressure

accompanied by heat. The essential oil is a pale, yellow liquid

containing various substances such as eugenol, d-camphene,

geraniol, free myristic acid and most important, the narcotic

myristicin. Some cases of poisoning have been recorded by persons

taking an overdose of this oil, which is used medicinally as a

carminative and as a flavouring agent. As a culinary spice, the

nutmeg is more commonly grated in small quantities from the

actual nut and used to flavour cakes, puddings, custards, punches

and possets.

The fixed oil is known as “nutmeg butter” and is a very aromatic,

orange-coloured mass when extracted. It is not of culinary interest,

but is used medicinally as a mild stimulant for external application

and in the manufacture of perfumes. According to Redgrove,

Spices and Condiments (1933) 293, it consists of 73 per cent of

trimyristin, 12-5 per cent of essential oil, 3:5 per cent of fat, some

myristicin and 8-5 per cent of unsaponifiable materials.

The mace or aril is very aromatic and is also used as a spice.

It contains a volatile oil similar to that present in nutmegs and

also moisture, fat, amylodextrin and starch.

238

Vol. XVI. (1958).

In Malaya a table jelly (preserve) is made from the fresh husks

or pericarp of the ripe fruit. In Penang the pickled pericarp is

sometimes eaten as an appetizer with alcoholic drinks.

Besides M. fragrans, other kinds of nutmegs have been sold or

used commercially. These have hardly any aroma in either the

seed or the aril. Ridley reports that the only one at all aromatic in

Malaya is M. cinnamomea and that it is by no means strongly frag-

rant. The seeds of Gymnacranthera farquhariana (canarica) have

been used in India for candle making. Those of M. malabarica have

been used for oil and for adulterating nutmegs. Its aril is known as

Bombay Mace. M. argentea, the Papua or Macassar nutmeg, is also

used for making oil or soap, but is useless as a spice. The seeds of

Virola surinamensis are used commercially for making wax.

CLASSIFICATION

The characters which are of importance in classification have

already been mentioned and may at this point be conveniently

summarized.

(1) Androecium. It seems to be the most important of all as far

as the Malayan species are concerned. The way in which the an-

thers are attached to the column will distinguish the four Malayan

genera. The number of anthers will distinguish many of the Knema

species from each other, as their anthers are free and may be easily

counted. The shape of the column in Horsfieldia will distinguish

certain species.

(2) Fruit. The fruit is second in importance. The shape, size and

character of its tomentum will distinguish many of the Malayan

species from each other. The aril, if entire, will distinguish Hors-

fieldia from the other three Malayan genera and if laciniate at the

apex only will single out Knema.

(3) Inflorescence. The type of branching of the inflorescence is

nearly as important as (2). It will separate many genera from each

other, especially the African genera with capitate and umbellate

inflorescences.

(4) Bracts, bracteoles and pedicels. The presence or absence of

bracts and bracteoles, especially the latter, but not so much the

former, is of value. The position of the bracteole in Knema will be

helpful for certain species. The length of the pedicel for many

species of Knema and to a less extent in Horsfieldia and Myristica

is a useful character (see table in systematic part).

239

Gardens Bulletin, S.

(5) Flowers. The size of flower is useful; colour in some species

of Knema; whether the perianth is bivalvate or trivalvate for Hors-

fieldia.

(6) Leaves. Leaf characters are less helpful except in the Ame-

rican species.

(7) Twigs. Useful for many species of Knema.

(8) Bark. Useful for many of the Malayan species in the field if

one can remember what it looks like.

(9) Seed. The non-ruminate endosperm of the following genera

will distinguish them from the rest, which have a ruminate endos-

perm in the seed:—Compsoneura, Iryanthera, Brochoneura, Ce-

Phalosphaera and Staudtia. The embryo is very small and not easy

to find or dissect, but it may show whether the cotyledons are con-

nate or free at the base and whether the lobes are sub-erect, divari-

cate or horizontal. These are useful characters to a certain extent.

(10) Starch. Presence or absence of starch in the endosperm will

distinguish certain genera, e.g. Gymnacranthera and Horsfieldia

(absent in both) from Knema and Myristica (present in both).

PHYLOGENY AND EVOLUTION

Comparison with other Families

The family is a relict one and certain distingvishing characters

set it aside from its closest neighbours. It had equal opportunities

along with its relatives under the same climatic conditions and

started off in competitive evolution with them, perhaps in the

Jurassic or the early Cretaceous.“ Apparently it originated in the

tropical rain forest amid large clumsy creatures like mastodons,

sabre-toothed tigers and primitive birds, many of them with teeth

that could crush its hard seeds with their masses of endosperm and

fat. Although it traversed the belt of tropical rain forest from

Malaysia to Africa and America, it stayed in that belt of generous

and equitable climate amidst a pachycaul vegetation, brightened

by scarlet fruits, orange or red arils and conspicuous black seeds,

dangling on long funicles—the true, primeval picture of that early

dawn when spiny durian-like fruits were abundant. Its primitive

features and intolerance of adverse climatic conditions did not give

it a chance to migrate to the temperate regions. It was, more or

less, doomed to remain in the shelter of the primary tropical rain

forest.)

It is a uniform family and its floristic and carpological characters

show few marked evolutionary trends. In fact the female flowers are

so uniform as to be of little use in the systematic separation of

240

Vol. XVI. (1958).

genera—a reason why botanists failed for so long to recognize more

than one genus in the family. It is more closely related presumably

to the Annonaceae than to any other family. Both have the same

geographical distribution and the same habitat, the same mono-

podial growth form with whorled branches and a distichous ar-

rangement of the leaves, which are themselves similar, simple and

without stipules and with the veins anastomosing at the margins.

The trimery of the flower is also suggestive of Annonaceae, but the

Myristicaceae have lost the inner whorl (petals) and their sepals

have become gamosepalous. However, in many Annonaceae the

sepals are often united at the base. With the loss of the inner petals,

the union of the sepals will make up for their lack of extra protec-

tion of the sexual organs. The Myristicaceae are nearly all dioeci-

ous, except for Brochoneura and some species of Iryanthera. In

this respect they differ from most of the Annonaceae, which are

generally hermaphrodite, but separation of the sexes in that family

does occur. The dioecious condition is probably more advanced

than the hermaphrodite flower. The complex branching of the in-

florescence in Myristicaceae is regarded by some as primitive.©?

Single flowers do not occur as in the Annonaceae, but that family

has inflorescences with just as great a diversity of branching. The

Annonaceae possess many more genera and thus have a better

chance to develop diversity in their evolutionary trends. Unique

and difficult to explain is the diversity of the androecium in the

Myristicaceae. In this respect the family stands apart from others.

The monodelphous condition is generally regarded as more primi-

tive than anthers with separate filaments. The degree of the fusion

of the anthers to the monodelphous column is an important syste-

matic character and may throw some light on the development of

the family and its genera. The anthers themselves have two loculi

whereas those of Annonaceae have four, but both have extrorse

dehiscence.

The single, apocarpous, one-seeded carpel of Myristicaceae is

more advanced than the numerous, stalked, many-seeded carpels of

the Annonaceae with their massive torus. It is not so advanced as

the syncarpous examples which the latter does possess. There is

no syncarpy in Myristicaceae and no thick torus. There is no ad-

vantage in having a thickened torus when the carpel is sessile and

single.

The presence of an aril is often regarded as a primitive feature.)

An aril is also met with in some of the Annonaceae, but not very

frequently and it is neither so well developed nor so brightly co-

loured. The aril is originally entire and splits later, but in some

genera it remains entire or is fimbriate at the apex only. There is

241

Gardens Bulletin, S.

no real argument, however, for claiming that an entire aril is more

primitive than a laciniate one. The only advantage I can think of

is that seeds with an entire aril which traps air may float better

than those partially covered by an aril. In Brochoneura chapelieri,

as has been stated, the aril surrounds the base of the seed only, but

this condition may be pathological rather than constant since that

species was collected once only and more material should be ex-

amined when available. Thus the genus ought not to be classed as

primitive or advanced from this character as the other species of

Brochoneura have a complete and laciniate aril.

The great quantity of endosperm in the Myristicaceae is prob-

ably a primitive feature and consequently the cotyledons are very

small. The endosperm in the majority of the genera is ruminate

and in this respect it is similar to that of the Annonaceae. I sup-

pose it was originally entire and then the cracks of rumination start,

giving it the crazy pavement-like pattern. Some genera with certain

advanced features have a non-ruminate endosperm and others with

a ruminate endosperm are less advanced in other ways. The two

most advanced genera in America, Compsoneura and Iryanthera,

advanced from the pollen morphology point of view, have a non-

ruminate endosperm, but I am unwilling to say that a non-ruminate

endosperm should be regarded as something advanced.

On pollen morphology and embryology, the Myristicaceae are

closer to the Annonaceae than to the Lauraceae, but on wood ana-

tomy the alliance is with the Lauraceae.®: 9 1% The single, superior,

sessile ovary of the Myristicaceae with one ovule is also found in

the Lauraceae, but there is no endosperm in the seed of the latter.

Lastly the presence of tanniniferous tubes in the rays which secrete

the red sap kino is a special feature of the Myristicaceae and is

found in no other family.“! 12) Summing up, the points of resem-

blance, therefore, between the Myristicaceae and the Annonaceae

are more numerous than between the Myristicaceae and any other

family.

The Significance of Geographical Distribution

As has been pointed out, there are three geographical regions

where the Myristicaceae occur, Indo-Malaysia, Africa and Ame-

rica. The chief centre of distribution is the Indo-Malaysian with the

majority of species in New Guinea and with the largest genus Myris-

tica, also well represented in New Guinea. The total number of spe-

cies in the four Indo-Malaysian genera is 170-190, if we take into

account those still to be described from unnamed collections and at

the same time allow for reductions. In Africa there are six genera

242

Vol. XVI. (1958).

with some twenty-one species. In America there are five genera

with seventy-three species.

It is generally supposed that botanically the Malaysian region is

very old. Several Malaysian genera have been found in the London

Clay deposits, but no Myristicaceae. Reid and Chandler state that

the most marked relationship of the flora of the London Clay is

with that of Indo-Malaya, particularly with the Malay Islands.“

If we suppose that the Palaeo-Myristicaceae arose in the Malaysian

region, then how did they migrate to the other two centres? One

can suggest west by Ceylon to Madagascar and on to Africa by

some former land connection or also from Western India; west

again from Africa across the Atlantic to the West Indies, Central

America and Northern South America. There was again another

migration east from New Guinea to Fiji, Tonga, Samoa and the

nearer Pacific Islands but I do not think that there was any further

crossing of the wide, deep Pacific Ocean, although that would not

be impossible according to Croizat (his ideas are discussed in a

later paragraph). The flora and fauna of the islands off the South

American coast contain a high percentage of endemics, a few South

American genera, but no Myristicaceae and very few other angios-

perms came to South America from these islands, the migration

being west to these islands. Many botanists are against the idea of

there being former “land bridges” connecting continents and ac-

cording to others, the migration may be better explained by Wege-

ner’s “Theory of Continental Drift” Good states that this theory of

continental drift would explain plant distribution to-day to a re-

markably complete degree if it could be substantiated in the follow-

ing terms and made to incorporate the following points: —“*)

1. That at least some time between the Cretaceous and the

middle or later Tertiary, the continents were, more or

less joined into one and were fixed in position.

2. That as a consequence of a thermal reaction of the kind

pictured by Joiy,“>) the sima became molten in the latter

part of the Tertiary.

3. That continental drift resulted first, perhaps, at a rapid rate

and then as the sima cooled, more slowly.

4. That an immediate result of this drifting was the uplifting

of mountain ranges on the forward sides of the moving

masses.

5. That a later consequence of the rapid loss of heat was the

onset of the glaciations of the Pleistocene.

243

Gardens Bulletin, S.

Croizat,“) a phytogeographist, has little time for Wegener’s

theory of shifting continents or wandering poles. He says that phy-

togeographists need no theories. He, himself, began by plotting the

factual records of plant distribution on maps of the world and after

studying the distribution of many genera and species on many maps

saw that the migrations were very orderly and followed definite

tracks. From the direction of these migrations he could see reason

for the presence of many former land masses and oceans and on

page 521 of his book gives a short account of these former geogra-

phic configurations at the start of angiosperm migrations in the

latter part of the Jurassic. He says that without former land masses

dispersal would make no sense, and that the angiosperms streamed

generally northward from the southern hemisphere, across certain

geologic lands now covered by the modern oceans. The most

important of these former geologic lands was “Gondwana”, which

in Pre-Cretaceous times, welded as one, Africa, Asia, Malaysia,

Australia and additional lands in the Western Pacific. In the south-

ern hemisphere there were three main “gates of angiospermy from

which the tracks of all angiosperms took their start.

If all this theory is not nonsense, then we ought to be able to say

that genera furthest away from the Malaysian region, namely the

American, should be more distinct from the Malaysian genera with

more variations than the African.“® This, I think is true generally,

but there is just as much variation in the Malaysian genera as

in the exotic. The African genera have specialized in the

development of a more compact inflorescence, the capitulum or a

panicle which ends in capitula. This already was evident in Ceylon

with Horsfieldia iryaghedhi, which has a capitulum. The American

genera specialized in a tendency to have the anthers free from the

column. However this tendency was already indicated in the Malay-

sian region. It is incipient in Horsfieldia and Gymnacranthera, but

well advanced in Knema with the free anthers attached to a peltate

disc. °

The trouble is that there are too few genera in the family to say

which are primitive and which are more advanced. The Annona-

ceae have more genera and it is much easier there to say which

genera are advanced. A. C. Smith, after studying seventy-three

species in the five American genera, states that he is unable to say

which genus is the most primitive or the most advanced. R. P.

Wodehouse, after examining the pollen grains of some of these,

comes to the conclusion that Compsoneura and Iryanthera are the

most advanced and that Dialyanthera is the most primitive.®

If we regard species with the anther completely fused to the an-

droecium column as the most primitive, (because that condition is

244

-* Go»

Vol. XVI. (1958).

the basic one, found in the majority of genera and also in Myristica

which has the greatest number of species), and the genera with free

anthers the most advanced, then we can add this evidence to the

case of Compsoneura and Iryanthera to strengthen R. P. Wode-

house’s statement that these two genera are the most advanced.

Also from what the logic of the distribution theory postulates, we

can deduce that they came later. Dialyanthera also has free anthers

and we are left to explain why it should not be so advanced since

its pollen grains were found to be primitive. Evolution does not

proceed evenly at the same rate in all organs.“!”) A. C. Smith says

he does not know why there should be more species in Virola than

in the other American genera. This genus has the basic type of

androecium which is the most primitive and the most universal and

hence the reason for the large number of species. This also is the

reason why Pycnanthus has the largest number of species in Africa

and Myristica more species than any single genus in the family.

Genera which have departed from this basic type of androecium

may be considered to be more recent. Those with free anthers, es-

pecially the American, do not have so many species. The exception

is Knema which has a fair number of species, with one centre of

distribution in Malaya and another in Borneo.

The reason why Myristica has flourished in New Guinea is that

the majority of its species are small trees and have small

fruits. Myristica is not so common in Malaya, where its species are

taller trees with large fruits and seeds. By not so common I mean

that not only are the different kinds of species fewer, but that there

is a scarcity of individuals of any one species in a given area of

forest. The tall trees will take longer to produce seeds than the

small. Also Knema in Malaya has been successful because the trees

and the fruits are small. Here, too, many of them are small trees.

Horsfieldia species with small fruits and small stature outnumber

the taller, large fruited species both in Malaya and in New Guinea.

Gymunacranthera has only some six species, but is very widely dis-

tributed from India throughout Malaya to the Philippines and New

Guinea. Its fruits and seeds are small. Small seeds are more easily

swallowed and distributed by birds than the large ones which prob-

ably fall close to the tree and rot. Mention has already been made

of the distribution of the small, hollow seeds of Horsfieldia irya by

water, a species which is very common in Malaya by streams and

has a wide distribution in Indo-Malaysia extending from Ceylon,

the Andamans, Burma, Siam and the Malay Islands to New Guinea,

Polynesia and Micronesia. Myristica fragrans, the commercial nut-

meg, owes its wide distribution to man and as a cultigen under

artifical care, it produces both globose and pear-shaped fruits.

245

Gardens Bulletin, S.

Once again it is very difficult to say which genera in this relict

family are most primitive and most advanced, since, owing to un-

equal evolutionary developments, some which are advanced in one

character are behind in another. Also the available fossil evidence

is of little value. All that we have is the leaf fragment of one named

Myristicophyllum from the Island of Labuan, found in Tertiary

deposits.4®) If I were to make a decision for the Malayan genera

regarding their order of rank, I should place them thus: —Myristica

the most primitive, but the most successful and the basic genus in

the family, Horsfieldia next, with Gymnacranthera close to it and

Knema the most advanced. They may also have evolved in this

order in time sequence. It must be emphasized that they all stand

close to each other and that the degree of primitiviness between

the first and the last is not great. The evolutionary trends in the

family may now be summarized thus: —

Knema is the most advanced because of the free anthers, the

simple inflorescence, the small fruit, the more numerous stigma

lobes offering more exposed surface to the pollen, and the free

cotyledons. The aril is more primitive, being laciniate at the apex

only. The other three genera have a branched type of inflorescence

which is considered to be primitive, although the Knema type is

sometimes found in Myristica. Horsfieldia has an entire aril and is

less advanced than the others in this respect, but it is ahead of

Myristica chiefly because of the greater diversity seen in the sta-

minal column and in the tendency for the anthers to become free.

Also the degree of union of the perianth segments is greater in

Horsfieldia, resulting in the globose perianth with a very small

2pical opening. Gymnacranthera is only slightly ahead of Hors-

fieldia because of the greater freedom of the anthers and of the

divisions of the aril extending to the base.

References to section on Phylogeny and Evolution

1. Croizat, L.: Manual of Phyto-Geography. The Hague, 1952.

2. Corner, E. J. H.: “The Durian Theory”. Annals of Botany, 13, 52

(1949) 367-414.

3. Lawrence, G. H. M.: Taxonomy of Vascular Plants. The MacMillan

Co., New York, 1951.

4. Wilson, C. L.: “The Evolution of the Stamen”. Chron. Bot. 6 (1941)

245.

5. Corner, E. J. H.: “The Durian Theory Extended—II. The Arillate Fruit

and the Compound Leaf”. Phytomorphology 4 (1954) 152.

6. Martin, A. C.: “Morph. Seeds.” 4m. Midl. Nat. 36 (1946) 513-660.

7. Gundersen, A.: Families of Dicotyledons. Chronica Botanico Co., Mass.,

U.S.A. 1950.

8. Smith, A. C. & Wodehouse, R.P.: “American Species of Myristicaceae”.

Brittonia 2, 5 (1937) 401.

246

Vol. XVI. (1958).

9. Joshi, A. C.: “A Note on the Development of the Pollen in Myristica

fragrans and the Affinities of the Family Myristicaceae”’. Journ. Ind.

Bot. Soc. 25 (1946) i39-143.

10. Erdtman, G.: Pollen Morphology and Plant Taxonomy. The Chronica

Botanica Co., Waltham, U.S.A. & Almgqvist & Wiksell, Stockholm,

1952.

11. Garratt, G. A.: “Systematic Anatomy of the Woods of the Myristi-

caceae”. Trop. Woods 35 (1933) 6-48.

12. Metcalfe, C. R. & Chalk, L.: Anatomy of the Dicotyledons 2 (1950)

1136. Oxford, 1950.

13. Reid, E. M. & Chandler, M. E. J.: The London Clay Flora. British

Museum Publication, 1933.

14. Good, R.: The Geography of the Flowering Plants. Longmans, Green

& Co., London, New York, Toronto, 2nd edit. 1953.

15. Joly, J.: The Surface of the Earth. Oxford, 1925.

16. Willis, J. C.: Age and Area. Cambridge, 1922.

17. Bessey, C. E.: “Phylogenetic Taxonomy”. Ann. Missouri Bot. Garden 2

(1915) 112.

18. Geyler, H. T.: “Uber fossile Pflanzen von Labuan aus Vega”. Exped.

vetensk, Jakttagelser, Stockholm 4 (1887) 499 Tab. 33, Figs. 3-6.

ACKNOWLEDGEMENTS

I am indebted to the following institutions for the loan or ins-

pection of herbarium material:—Forest Research Institute, Ke-

pong; Forest Department, Lae, New Guinea; Indian Botanic

Garden, Sibpur, Calcutta; Forest Research Institute, Dehra Dun;

Sarawak Museum; Muséum National d’Histoire Naturelle, Paris;

Botanical Museum and Herbarium, Utrecht; the Rijksherbarium,

Leiden; Herbarium of the Botanical Institute of the University,

Wroclaw (Breslau); Herbarium Universitatis Florentinae, Firenze;

Royal Botanic Gardens Kew, Royal Botanic Garden, Edinburgh;

the British Museum; the Botanische Staatssammlung, Munich; the

Conservatoire botanique and the Boissier Herbarium, Geneva; the

Arnold Arboretum (Knema and Gymnacranthera only); the

Naturhistoriska Riksmuseum, Stockholm; the Botanical Museum

of the University of Uppsala; the Philippine National Herbarium,

Manila; the Botanisches Museum, Berlin-Dahlem; the Herbarium,

Section of Forest Botany Bangkok and the Herbarium of the Agri-

cultural Department, Bangkok; the New York Botanical Garden;

Department of Agriculture and Stock, Brisbane; Botanical Museum

and Herbarium, Copenhagen; National Herbarium of Victoria;

Melbourne; Jardin Botanique de I’Etat, Bruxelles and the Her-

barium of the University of California, Berkeley (Knema and

Gymnacranthera only). Material on loan from a few other herbaria

has not arrived yet and cannot now be included in this paper. I am

grateful to Mr. G. H. Addison, Botanic Gardens, Singapore, for

the photographs of bark.

247

Gardens Bulletin, S.

SYSTEMATIC PART

The following abbreviations are used here: —

Mat. F.M.P. .. Material for a Flora of the Malayan

Peninsula, by J. S. Gamble.

F.M.P. he .. Flora of the Malay Peninsula, by H. N.

Ridely.

S.F.N. keh .. Singapore Field Number.

CF: nr ..- Conseryator of Forests (a series of field

numbers in Kepong).

F.D. te ..- Forest Department (a series of field num-

bers in Kepong).

K.F.N. he .. Kepong Field Number (the present series

of field numbers in Kepong).

F.M.S. Mus. Herb. .. Federated Malay States Museums Her-

barium (a series of field numbers used

by that museum but all the material is

now incorporated in Singapore Her-

barium).

F.R. oi -» Forest Reserve.

So ss vt .. Sungei (river).

G. Shs .. Gunong (mountain).

D.F.O. Se .. District Forest Officer.

The herbaria in which the cited specimens are deposited are

indicated by the usual capital letters or abbreviations adopted in

the Index Herbariorum. No herbarium is quoted if I have not per-

sonally seen the particular specimen I cite even although I know

the specimen to be there.

MYRISTICACEAE

Dioecious, rarely monoecious monopodial trees from 30 cm.—40

m. high (majority 5-8 m. high); stilt roots sometimes present;

branching whorled, the crown often pyramidal. Bark brown, black

or greyish, smooth to fissured or less often flaking, exuding a red

sap on cutting. Wood white or soon brownish on cutting, soft; rays

narrow, tanniniferous tubes present; growth rings usually distinct

and visible to the naked eye. Twigs stout or slender, striate or

smooth, their apical portions often tomentose with various types

of hairs, branched and rhizoid like, stellate, or less often simple.

Leaves simple, entire, alternate, often appearing distichous, exsti-

pulate, often tomentose on both surfaces when young and later

glabrous or sometimes remaining tomentose on the lower surface;

primary nerves distinct, anastomosing at the margins; secondary

nerves sometimes present but rarely as numerous and distinct as

the primary; reticulations present or sometimes very faint or not

248

Vol. XVI. (1958).

visible. Male inflorescences axillary or in the axils of fallen leaves,

on the twigs or on the older branches, very rarely cauline, the axis

often much branched or unbranched, one, less often two and

three main axes arising in one leaf axil, with the flowers at the ends

of the branches in racemes, cymes, umbels, racemose-umbels,

capitula (the capitula occasionally aggregated and confluent,

forming catkins), if unbranched, then the inflorescence short and

woody with the scars of previous pedicels and bracts and the flow-

ers in racemose-umbels at its ends; bracts present or more often

caducous; bracteoles absent or occasionally present; pedicels

usually present, less often absent. Female inflorescence similar to

the male, but less branched and shorter, rarely longer. Perianth

gamophyllous, rarely split to the base, 3-toothed or lobed, less

often 2-toothed, valvate, funnel-shaped, campanulate, urceolate,

patelliform, globose, sub-globose or rarely rotate, the female

usually larger, more swollen and urceolate than the male.

Androecium, elongate, cylindrical, globose, cup-shaped or pyra-

midal, stalked or sessile, sometimes with a short sterile apex; an-

thers adnate to the column of the androecium by their backs and

to each other by their sides or less often free at their sides and api-

ces or entirely free from each other, their bases only attached to

the column, the latter sometimes expanded into a peltate or tri-

angular disc with stalked or sessile anthers arranged round it stel-

lately, 2-45 in number, elongate, extrorse, 2-loculed. Ovary single,

sessile, superior, globose or sub-globose; style absent or rarely pre-

sent; stigmas 2, connate or rarely a peltate disc with few to many

laciniations; ovule 1, anatropous, situated near the base of the

ovary or slightly above it. Fruit globose, sub-globose, pear-shaped,

oblong, or elliptic, rarely transversely elongated, dehiscing by a

longitudinal circumferential suture into 2 valves, rarely indehiscent;

pericarp coriaceous and fleshy, glabrous, pubescent, tomentose or

lanose. Aril red or orange-red, covering the seed, laciniate to the

base or only at the apex or sometimes entire, attached to the base

of the seed between the hilum and the micropyle. Seed conform to

the fruit, the testa of three layers; albumen hard, ruminate or less

often non-ruminate, containing a fixed oil and sometimes starch.

Embryo very small near the base of the seed or a little above it;

cotyledons connate at the base or less often not, sub-erect or divari-

cate to horizontal, flat, if connate then peltate, cup-shaped or

patelliform, their edges smooth or undulate; radicle very small,

inferior and cylindric.

249

Gardens Bulletin, S.

KEY No. 1 (Male flowers)

a. Inflorescence a short woody axis or tubercle-like protuberance

with the scars of previous pedicels and bracts; bracteoles

present

b. Androecium a stalked, peltate or triangular disc with the

anthers free, attached stellately to the disc only by their

bases. Bracteoles at the base of the flower or median on

the pedicel 1. KNEMA

b. Androecium a stalked column with the anthers completely

fused to it by their backs, the apex of the column often

sterile. Bracteoles embracing the base of the flower on one

side 2. MYRISTICA

(section with M. crassa)

a. Inflorescence not a short woody axis but a smooth, often

branched panicle or the basal part always smooth and the

upper part occasionally woody with a few scars; bracteoles

present or absent

c. Androecium globose, cup-shaped, trigonous or cylindric,

nearly always sessile, rarely very shortly stalked. Anthers

adnate to the column by their backs, their apices some-

times free; bracteoles absent 3. HORSFIELDIA

c. Androecium an elongated column, the anthers free at the

apex or not, column stalked or sessile. Bracteoles present

or absent

d. Anthers not free, column stalked, often with a sterile

apical portion. Bracteoles present at the base of the

perianth on one side. Twigs striate at least in the older

portions 2. MYRISTICA

d. Anthers free at the apex, column sessile or very shortly

stalked. Bracteoles absent. Twigs not striate

4. GYMNACRANTHERA

KEY No. 2 (Fruit)

a. Aril laciniate to the base or nearly to the base

b. Twigs striate, at least in the older parts. Reticulations of

the leaves often visible above but never a dense network

2. MYRISTICA

b. Twigs not striate. Reticulations of the leaves not visible above

3. GYMNACRANTHERA

250

Vol. XVI. (1958).

a. Aril entire or laciniate at the apex only

c. Aril entire, covering the seed, slightly lobed or convoluted

at the apex. Leaves not whitish beneath; reticulations

few and faint above 3. HORSFIELDIA

c. Aril laciniate at the apex only. Leaves whitish beneath;

reticulations forming a dense network above in dried

leaves 1. KNEMA

KEY No. 3 (Sterile material)

a. Twigs not striate, often angled. Leaves whitish beneath; reticu-

lations not visible above 4. GYMNACRANTHERA

a. Twigs striate or at least in the older parts. Leaves whitish or

not beneath; reticulations distinct above or not

b. Leaves often whitish beneath, texture not granular

c. Reticulations forming a dense, close network above in

dried material, raised 1. KNEMA

c. Reticulations few above, not forming a dense network,

mostly sunk 2. MYRISTICA

b. Leaves not whitish beneath, texture often granular

3. HORSFIELDIA

Gardens Bulletin, S.

"11

Suloviquis jyey “IoMOop

jo oseq ye uoTIsod

*OUNI} Idj1OYS IO IOSUO]

eB IOJ Bunsisiod ‘juoso1g

*snoo

-nped pur [jews “juUssolg

‘(Z) pur

(1) jO woneurquiod (¢)

“‘DuaUuy Ul Se

souviognjoid Apoor (7)

poysuvig (1)

‘yuns ‘viuauy Ul se osuop

Iaaou =ynq = MIpjaifsioyy

UI UvY} SNOJOUINU s1OW

Ajyysis ‘aaoqe d{QISIA

SOUNJOWIOS SUOTL[NIO1

‘yyeoueq )=YsIyYM Udo

“SUIUTYS Ud}JO

‘AJQSO]D OS 9}1INb jou ynq

s}jied JoplO sy} UT 9}eINS

BOTSLIAW]

. APWLOYS

“IOMOP JO WSvq 1v 10 [99

“Ipod jo o[ppru ye vol

-isod ‘Sunsisiod ‘juosolg

‘snoonped ‘jussotd

‘xode oy} 1% py-¢-Z

A][WUuOIsed50

‘sIvoS YIM s0uRIOqQN}

-O1d 9¥I[-qouy APpOOM

‘OAOQLR YIOMJOU Posiv4

QSO[D OsSUSP B BSUTWIIO]

Ud}JO JSOLU SUOTILINIT}IOI

‘yyeoueq =YsIyYM Udo

"YJOOWS SOUTJOWUOS

Sunod ‘s}ied sJOplo

oy} Ul oes ApIsO[D

eWloUy

‘yuOSqYy

*snoonprs ‘juUNsoIg

‘poyourig

“OTHHIQ

Ajyensn $= ‘“repnuess

ud}IJO Jeo] JO 91N}Xx9}

:yuns pure Moy yng

QAOGe IQISIA SOUT}

-OWOS SUOlP[Nd}O1

‘yyeoueg YsHIyM JON

‘sjived Iopyo oy} UT AT

-[B1ID9dS9 9}¥11]S A[OSO[D

eIp}ays1OH

‘yuosqy

‘snoonpes “judso1g

‘(po}oe1] U0

a] eUW9]) poyourlg

“oAoge

Q[GISIAUL suOoT}PR[NS

“Hor SyyeoUeq YSHIYAA

‘syied

SuUNOA OY} UI Sururys

pue yJoows ‘pojsue

ud}jO *9}V11]S-UON

vIDYURIOVUWUAD

SHTOALOVad

SLOVUd

HONAOSAUOTANI

SHAVHT

SOIML

slajovieyy

VYHNAD NVAVIVW AHL NI SHONHAHAHIG GNV SHILIAVTIINIS YOLVW AO ATAVL

252

Vol. XVI. (1958).

‘O}VOLIVAIP

soqo7T ‘podeys-dno uso

pure oseq oy} 3% o}eUU0D

"yoieys pue

[IO poxy v YIM o}vUIWNY

‘OSE 94} 0} JSOW]Y

IO seq OY} 0} 9}RIUIDRT

‘paqoy-1q Aja NUT,

‘jUOSqV

‘poyoryie Alo}o[/du109

ole sioyjue 9)esU0T9

oy} YoIyM 07 ‘xode

QUIDI}XO OY} 3 OILIO}S

udyjO “UUUN,[OD poyXeIs YW

BOTISLIAJI

‘O}VILIVAIP IO

yoroqns soqoyT ‘aseq

oy} We 9JBUUOD AT}YSITS

A19A A[UO IO ApIDIBIS

“‘Yyoivys pue

[10 poxy B YIM o}RUIUUIN YY

A[uo xodv oy} 18 9}VUINRT

‘POpIAIp UTese UD]JO

S9qO] SY} “paqoj-!q 10

o}eIUIoe] pur WIOFIOSIG

‘“ou0 JOYS B IO JUaSqY

‘posuviie A[9}v][9}S

OB SIOYJUL dOIJ DTISSOS

IO poyyeis oy} Yorum

punol osip sjve[nsue

-I1} 10 9}e)[0d poyleIs V

euloUy

‘O]BVOLIVAIP SOqo’T

‘oseq oy} ye o}yeUUOD

"yoie}s OU pure [IO

poxy B YIM o}eUTUUNY

‘A{UO xodv oy) 1%

SOqO] Moj B YIM JO

Pods OY} BULIDAODOINUA

‘paqo]-iq Ajo NUIT

"yuosq Vy

"SOpIs

pue sooide s10y) 3e

1OU 10 9013 puv Syoeq

I1oy} Aq HI 0} poysr}

-je sioyjue 9}e3U0]O

oy} SuIAvy pruresdAd

JO ostp ‘dno ‘uuinyos

“poyleys Ajorel “oyIssos YW

PIP[OYS10

panuyuod—-WAANAD NVAWIVW AHL NI SHONAYAAKIG GNV SAILIMVTIINIS UOLVW AO ATAV.L

‘OIVOLIVAIP Soqo’]

‘oseq oy) We d}eUUOD

"Yo1e}s OU pur [IO

poxy & YIM oyeuTUNYy

‘OSEq 94} 0} }SOLUTe

IO OSeq 94} 0} O}VIUIORT

‘paqo|-iq Ajaynuryy]

‘yuosqy

*‘poxoyul

pur so1j soorde 1104}

*poyoej}e div s1oyjue

ayesuo0]9 OY) YIM

0} UWIN{[OD poyYris VY

| |

BINYJURIOVUWAD

SNOCHTALOD

NAWNE TV

Wav

VWOILLS

ATALS

WNIdOUadNvV

sI9}ov1eyD

253

Gardens Bulletin, S.

1. KNEMA Loureiro, Fl. Cochinch. (1790) 604 et Ed. Willd.

2 (1793) 742; Warb., Monog. Myrist. (1897) 131 and 543;

Gamble, Mat. F.M.P. 5, 23 (1912) 236; Ridley, F.M.P. 3

(1924) 66.

[Myristica sect. Knema Bl., Rumphia 1 (1836) 187; Hk. f. et

Th., Fl. Ind. (1855) 155; A. DC., Prodr. 14, 1 (1856) 204; King

in Ann. Roy. Bot. Gard. Calc. (1891) 284].

Dioecious trees 3—30 m. high, average height 10 m.; stilt roots

present in a few species. Twigs striate in the older parts, the younger

parts usually smooth and non-striate. Leaves glaucous beneath,

often covered, when young, with furfuraceous tomentum, soon

glabrous, or less often pubescent on the lower surface; nerves often

distinct, interarching at the margin; reticulations forming a close

network, often scalariform between the nerves, distinct on both

surfaces or occasionally faint or invisible above but always pre-

sent beneath. Inflorescence axillary or from the axils of fallen

leaves, consisting of short, woody, peduncular tubercles, 1 mm.—

1:7 cm. long (of slow, unlimited growth, the youngest at

the tips of the twigs), covered with the scars of previous

pedicels and bracts, usually unbranched but occasionally shortly

bi- or trifurcate; bracts caducous. Pedicels: male up to 2 cm. long,

average 7 mm.—1 cm., occasionally only 2 mm. long; female on the

average shorter than the male or the flowers sessile; a persistent

bracteole present at the middle of the pedicel, less often at the base

of the flower. Flowers in racemose or sub-umbellate clusters on

the woody peduncles, rather numerous, globose or sub-globose in

the male, usually urceolate and larger in the female, brown-tomen-

tose or pubescent outside, less often lanose or puberulous; inside

glabrous, cream-coloured, pink or red, 3-lobed (an occasional

one 4-lobed in the same species); lobes reaching 1/3 to 1/2 way

down or even to the base of the perianth, spreading horizontally

when mature but remaining a long time closed in bud before open-

ing. Filaments united, forming a peltate or triangular, usually sti-

pitate, less often nearly sessile disc; anthers 6-23 but 35—45 in K.

curtisti, attached stellately round the disc by their bases, sessile or

on a short filament, horizontal or less often sub-erect, extrorse,

free and quite separate from each other, or less often, when nu-

merous, touching each other. Style short or absent. Stigma disc-

shaped with many lobes or 2 to 4-lobed; if 4-lobed, bifurcate at the

junction of the style and each lobe divided again into 2 smaller ones.

Ovary pubescent or tomentose, globose or sub-globose. Fruit

stalked or less often sessile; pericarp thick, fleshy, usually with

some kind of indumentum, lanose, tomentose or pubescent, less

254

Vol. XVI. (1958).

often later glabrescent but never initially entirely glabrous as in

Horsfieldia and in some species of Myristica; aril laciniate at the

apex only; albumen ruminate with a fixed oil and starch; cotyledons

divaricate or sub-erect, scarcely or only very slightly connate at the

base.

TYPE OF GENUS: K. corticosa Lour., Fl. Cochinch. (1790) 605 =

K. globularia (Lamk.) Warb., Monog. Myrist. (1897) 601.

DISTRIBUTION: India, Burma, Siam, Indo-China, China, Malay Pe-

ninsula, Malay Islands, including the Philippines but rarer in the east-

ern group. One species K. tomentella extending to Morotai and New

Guinea.

Knema is distinguished from the other Malayan genera by its

androecium which consists of a stalked, peltate or triangular disc,

and of shortly stalked or sessile anthers arranged round the margin

of the disc. It also differs in nearly always having a dense network

of raised reticulations on both surfaces of the leaf or at least on

the lower surface. In Gymnacranthera these are absent on the

upper surface of the leaf and are few, far between or absent on the

lower. In Myristica and Horsfieldia, reticulations are sometimes

present but they never form such a dense network and if present

on the upper surface, they are sunk, never raised. If present on the

lower surface, they are level with it or slightly raised. From Gym-

_ nacranthera it also differs in the striate twigs; in the aril which is

entire or laciniate at the apex only and not cleft to the base; in the

sub-umbellate inflorescence which is unbranched and consists of

short, woody tubercles with numerous pedicel-scars of former flow-

ers (these tubercles may occasionally be very shortly bi- or trifur-

cate); the presence of a bracteole; the presence of a short style in

‘Many cases; starch in the albumen and the cotyledons scarcely or

‘only slightly connate at the base. From Horsfieldia it differs in the

leaves being glaucous or whitish beneath; the inflorescence never

branched; the pedicels having a bracteole; the perianth never 2-

lipped; a short style sometimes present; the presence of starch in

the albumen, and the cotyledons not connate at the base. From

Myristica it differs from some species in the unbranched inflores-

cence but agrees with others which have the same kind of woody

tubercles; also in having the aril not cleft to the base and in the

cotyledons not connate at the base. It agrees with some species in

the absence of a style and with others in having a bi-lobed stigma,

although many Knema species have a many-lobed stigma. Again,

the bracteole in Knema may be at the base of the perianth as is

valways the case in Myristica, or it may be median on the pedicel.

The older twigs of Knema are usually more closely striate than

those of Myristica and in this respect, they are not different from

those of Horsfieldia.

255

Gardens Bulletin, S.

The genus is a difficult one especially for beginners as many of

the species are close to each other. However the Malayan species

can often be identified even if sterile, provided that a sufficient

length of twig is present, showing, both the young apices and the

older stages where the tomentum is shed and where the striations

start. Twigs are, in fact, a very good aid to identification and even

their thickness is important since there are species with very slender

twigs and others in which the twigs may be quite stout even at the

apices. In three species hookeriana, furfuracea and latericia, the

bark on the older twigs tends te crack and this will not be apparent

unless a considerable portion of the twig has been selected. Sterile

specimens may present difficulties at first which should also dis-

appear after constant practice and observation; a greater care will

be required in the case of laurina and conferta, especially in the

former, where many forms which vary in degree of indumentum

and in size and shape of the leaves occur. One should not expect

to arrive at the correct identification of a species from the key by

taking measurements from immature flowers. These should be open

or well advanced in bud. If too young, the anthers may appear

sessile, when in reality they are stalked when mature. Again the

bracteole may appear to be at the base of the perianth, where in

mature flowers it will be median. In dried specimens, fruits may

appear to be mature, when in fact, they are immature and would

have increased a good deal more, thus upsetting ones measure-

ments. Leaves should be dried and examined in order to see the

reticulations which do not show up well in the fresh specimen.

Two keys are given here. One based on sterile material and

fruits, does not give any indication either of the section into which

a species ought to be placed or of the real systematic relation oi

species. The other key, though based mostly on male flowers, in-

cludes other useful systematic characters, such as female flowers,

leaves twigs and fruit, and attempts to place each species in close

systematic grouping. There is a small skeleton key to these groups

and all three should be used except when the specimens is sterile.

The chief systematic characters are also tabulated and a quick

glance at this table, prior to using a key, may decide the correct

identity of a species or at least its relevant group. These groups are

not defined as sections or any other botanical taxa, as I feel it is -

premature to give them an exact taxonomical category at this stage,

though this might be done when the position of the other Malay-

sian Knema species is known. No one has yet divided Knema into

sections but I found that certain systematic characters especially

those of the male flowers form a good basis for classifying the spe-

cies into closely related groups; certain closely related species

256

Vol. XVI. (1958).

naturally came together and could be placed in these groups. The

most useful systematic characters were:—whether the anthers are

sessile or stalked; if the bracteole is at the base of the perianth or

medial (this can be observed in female as well as in male flowers

but is more distinct in the male on account of the longer pedicel) ;

the stigma many-lobed or few-lobed, up to 6 in number and less

important whether a style is present or not. These, all taken to-

gether, do give a good indication of the systematic position of a

species but since the flowers are so small and are not always pre-

sent, foresters will probably not bother to examine them at all,

while, even in the herbarium, flowers have to be boiled to examine

and count the anthers.

From these characters there appear to be five groups of Knema

in Malaya as is shown in the table. Some of these groups can be

sub-divided into smaller groups, some with only one species, in

which case the species is more isolated from the others. At this

stage, I cannot say whether these single species from the Malay

Peninsula will continue to stand alone since I have not yet grouped

the other Malaysian species into the pattern with them, but it is

likely that some of these Malaysian species may go with them.

When the species from the rest of Malaysia have been properly

placed in the scheme, one should then get a clearer idea of the

relationship as a whole and might be able to decide whether the

smaller groups can be called sections or whether the five larger

groups are to be the sections and finally whether more sections

ought to be made.

The following will give some idea of which species are related: —

Group 1 Group 2 Group 3

hookeriana curtisii (rigidifolia

furfuracea a. ~ conferta

icsdcin | scortechinii

kunstleri

Group 4 Group 5 b. malayana

plumulosa oblongifolia L ae

’ L intermedia oblongifolia var. “ SenOpHyuS

b JS retusa monticola d. glaucescens and its

* | mandaharan globularia vars.

laurina

Species closest to each other are bracketed in this scheme. The

group of curtisii is a very small one and this species is different

from all the others in having such a large number of anthers.

Retusa has been tentatively placed with mandaharan as it seems

to be more closely related to it than to the rest but until male flow-

ers are available, its exact position cannot be clarified. Glaucescens

257

Gardens Bulletin, S.

q€

G€

ha 3

0t-F1

vI-L

87-81

87-1

Sc-LI

91-6

17-61

esojny

-U9WI0}-YsIppol ‘xode

38 osn}qgo ‘podeys-ivod

19}e] WUSOSIIG

-R]3 ‘snoynisqnd

~YsIppol “prosdiyyo

1938] JUSNS0IG

“P38 ‘QsO[NJUSWIO}

-AjsnI ‘9sOqo]3-ploAo

Joe] WsdsoI1QeIs

*yusosoqnd-9}e] [93s

-Aysni A[Zurieds ‘xode

ye yey soqye1l ‘proao

asojn}

-UdUI0}-AjsnI ‘d9sOqo]s

snoiqe]3

19}®] ‘snoynioqnd

-AjsnI ‘puso yore

ye poMmolieu ATYSITS

9SO}USUIO}-A}SNI

‘mOlIVU oSseq ‘osn}qo

xode ‘podeys-ived

wWin}USUIO}

qroys ‘ur §= ‘yng

ajed yWM pus yoeo

ye popunol ‘Zu0;]qGo

jOom osoury]

‘gnq oyed yim prosdiyye

“WS ZT I-T X

“Wd (-S°T

“md ¢*] x

“WD $°%-Z

"Wid €°7-Z X

"WD €-S'Z

"Wd $°Z-Z X

"Wd $-S"E

"Wd ¢

X "WD ¢

“wd € X

"wd S-¢

mo ct x

"WD LT

"wo ¢ Xx

"WD $"-S"E

"WD S*p-£ X

“Wd §-S"p

o]Issas

*P2q°]-b-Z

9]ISSOS

“‘P9Q°1-b-€

oyISsos

‘paqol-¢

oyIssos

‘poqol-¢

umouyun

Q]ISSOS

‘paqo]

—9 ‘Moy

9TISSes

*poqo]

oir

oyISses

*poqoy

-AUBU

O]ISSOS

*peqo]

-Aueul

uvlpour

uvIpoul

ueIpow =

uvIpour

utipow

uvIpow

uvipow

uvipew

ueripour

"Wd S°I-T

"uld |

—"wur /

“Wd U I~]

“WD |

—urW £

oinyewUut

rum ¢

“uu ¢-Z7

“WU ¢

‘wu ¢

“WU S—p

oinjyewwt

‘WUT ¢

"WU

“WU ¢

“ws /-9

pox[eis

AyV0ys €I-6

poxlers ZI-6

poyess ¢I-01

Poayeys 8I-€1

poxTeys

Ajqeqo.id SI-ZI

ayIssos S¥-SE

pox]eys

AjOYS AIOA ZI-6

a] ISsos €I-Ol

ayisses | (0Z)-gI-SI

dnoin

Jeo] Ul

SOAIOU JO

Joquinn

yNIy JO UONdIIOSOGg

g]Oo}9eIq

jo uolIsog

s]sorpod

oyeul jo

yiaueT

SIOMOY

g]eul jo

yiWwuseT

a] Issos sioyjue

IO poy]eys jo

sisyjUuy | Joquinn

SYHALOVAVHO OCILVAWALSAS TOHHSN AO AIEAVLI—-VWANY

euedryeul

epsuny

HUTYy993100S

B119;U09

BL[OJIPISII

tstyino

BIDIIO}IE]

eooRInyIny

eUuRLIOyOOY

so1seds Bulsuy

258

Vol. XVI. (1958).

CT-TI

07-SI

9IOU!

JO GZ

SZ-SI

O7-LI

O7-€1

yeoy ut

S9AIOU JO

Joquinn’

YO sqni yorym janos

9yeT[9IS-Ajsni ‘prosdiyjo

gsoury-Aysni ‘prosdiy[s

9sO[NUSUTO}

-UMOIG Yep ‘spuo

y10q 38 9snjqQo ‘suo[qo

asojny

-usU0}-AjsniI ‘*xode

ye A[WYSI[S puv oseq

ye pomolieu ‘prosdiyyo

aso[ny

-UsWIO}-AjsnI ‘spud

y10q 18 98Nj}qQO ‘BZuo[qo

Arp

usyYM yoviq ‘Ysory

usYM MOTIPA 10 Yuld

‘yusoso1qe|s spud

qyI0q 18 9sNIqQO ‘Buo;qo

snoiq

-B]3 SuIW0d9eq ‘asojny

-UudUIO}-Ajsni ‘xodv

popunol yIIM proaogo

esoiny

-U9UI0}-AjsNI ‘prloaogo

yn JO UONdIIs9q

“WD 77-8 T X

"Wd p-¢

"Wd $°7-Z Xx

"Wd p

“Wd L*T-$°T X

“WD 'C-7

"Wid ¢°Z-] x

"WD S*p-S°Z

“uid | x

“Wd €°]

“WD ¢*I[-T xX

"WD Q°]

9]ISsos

“poqoy]

-AUBUI

o]ISsos

*poqo|

-AUBUl

usos JOU

a[ISsos

“poqo|-p-Z

umouyun

o]Issos ==

*P9qol-S—p

O]ISSOS

*‘poqo]-9-¢

oyAis

IOYS AIOA

“poqo|-p

qiuetiod

jo oseq ——

qjueiiod

jo aseq

yjuerod

jo oseq

MOTAq 10

oyppru

2a0qe

ueripow =

ueIpoul

eTP

~pru oy}

2A0qe IO

uvipow =

uvIpour

uvipow

a]09}981q

jo uONIsog

"Wd |

—wu g

Wd |

—WwuI g

oinqyeuUt

‘Ww g-9

"wd |

—"WIU ¢

"WU p-¢

s[eorpod

ayeur jo

yiaueT

“WIL §—p

‘Wu 1-9

oun eur

"UI ¢-Z

“ww S-¢

"WU ¢-Z

‘WU p

“WU §°%-7

"wu ¢

SIOMOY

oyvu jo

yisusT

o]ISsos

JSOUW]Y 10

o]ISSOs

(9]Issos

ON} BUrUIT)

poy[eys

Ayoys

pores

poxrers

poy[eys

Apioys

pox]eis

Ayjsoys

pox [ess

Apoys

O]ISSOS

IO poy]eis

s1oyUy

SI-TI

SI-€l

vI-€l

vI-Ol

CI~L

LI-cl

8-9

6-L

sioyjue

Joquinny

vIpouliajut

vsornuinyd

B1ep109

‘IVA =, SUBOSOONL]S

** BrlAIoUTjUOTed

"IVA susdsoonvys

suoqni "J SusdsooneV]s

susosoone]s

i vy [Aqdouajs

us stunuwiwo9s

so1oods BMoUuy

penuyjuod—SAALOVAVHO OILVWALSAS TOXASN AO ATAVLI—VNANY

259

AY) a LC CC tC -

aS ’ —

8 yussseqnd yuesoid

'G) 0} 9S0}USUIO}-A}sNI "WD 7-S*I x ay Ays qijuerisd

S ee-TI ‘plosdijjs 10 ploao "WS €-S°C “Peqo[-Z JO oseq “WUE C—-€ “wu ¢-€ 9TISsos 01-9) * vupMNey]

snoiqe]s yuesoid

Ayjeuy = ‘jus soqnd “WoC "J—-o 1x ay Ais qjuersiod

S SI-€1 -Aysni ‘gsoqoysqns "10: 7=C"T ‘poqo]-p-Z jo oseq “mu ¢ “wu ¢ a]Issos €I-O1 eriepnqgoys

yuosoid

"wD 7-S'*] x ay Ais yyueried a]Issos

¢ S7-81 ‘od "WD ¢°7-Z ‘paqol-¢-Z yo oseq JO “WU | “UIU ¢ o]ISsOs 01-9 PRyOonUuOW

‘IBA = BITOJIBuUOTGo

wn}

-UdWI0}-AjsnI ‘y10Ys yuosoid

‘gsuop‘xede }e poyurod "Wd 7-SG°T x ay Ais yjueried

¢ O€ ‘PIOAO IO ploAogo-qns ‘wd Z ‘poqol-Z jo oseq “WU g-9 “WU p oTISSoS 01-9 BITOJISUO]GO

9soyuoUl

-O}-Ajsni ‘pouszey

pue poepunol xode a] ISsosS

‘jeJUOZIIOY pue prolq "wd g*€-¢ X ‘peqoy qjueried Bunod AOA

ar 97-77 aseq ‘3U0] Sev pvolq se “UID S*p-€ -Auelu jo oseq “mIU ¢ “WU p-¢ o]ISsas LI-91T | ° uereyepuew

9SO[N}UIUIO}-9}eTO

-d0Y9 ‘aseq je snoqg "wd ¢ x

ar $7-91 -qid ‘oyenoide ‘proao "ud 9-¢ uMouyuUN umouxun uMouyuUN umouyun uMouyun | UMOUYUN | * esnjol

jes] ul sjeorpod SIMO O]ISSOS sioqjue

dnoipy | soaiou jo yindy JO UOTdIIOSEd pa pst BUIZIIS ipa aa a]eu jo o[eUl jo IO poy]eis jo so1seds eulsuy

Joquinyy | J IS J HISOd yisusT yigsueT sioyjuy Joquinny

panuyuods —SadLOVAVHO OILVWALSAS TOAXSN AO ATAVL—VWANY

260

Vol. XVI. (1958).

is near to kunstleri, malayana and communis, but has been put

into a separate sub-group. K. stenophylla seems to stand half way

between glaucescens and the kunstleri sub-group but it could easily

go with either.

KNEMA—TABLE OF GROUPS

Group Systematic Characters Species

1 Bracteole median, anthers ses- hookeriana, furfuracea, latericia

sile, stigma many-lobed (anthers very shortly stalked in

latericia)

Zz Bracteole median, anthers ses-_ curtisii

sile, stigma few-lobed

(about 6)

3 Bracteole median, anthers rigidifolia, conferta, scortechinii,

stalked, stigma few-lobed kunstleri, malayana, communis,

stenophylla, glaucescens (brac-

teole may be above middle or

near base of flower in glauces-

cens and some of its vars.)

4 Bracteole at the base of the plumulosa, intermedia, retusa, man-

perianth, anthers sessile, daharan

stigmas many-lobed

5 Bracteole at the base of the oblongifolia, oblongifolia var.

perianth, anthers sessile, monticola, globularia, laurina.

stigmas few-lobed

KEY TO THE GROUPS OF KNEMA

a. Bracteole median. Anthers sessile or stalked

b. Anthers sessile

c. Stigmas many-lobed. Anthers not over 30, but 20 or less 1

c. Stigmas few-lobed. Anthers over 30 and up to 45 .... 2

Pee eM Gai et Gals laug FS dis whee ek ne wk ees 3

a. Bracteole at the base of perianth. Anthers sessile

Se SE SS re ae oa 4

ERAN EWN dE ls as en odecduw «) ecaveud sue sos 5

Note:—If the anthers are not mature, they may appear to be

sessile. They are shortly stalked in K. glaucescens but appear to be

sessile in immature anthers of that species. The bracteole in K.

glaucescens is usually above the middle, less often at the middle or

a little below the base of the flower.

261

Gardens Bulletin, S.

SCIENTIFIC KEY TO KNEMA

(Based with emphasis on male flowers, but including other useful

diagnostic characters)

Group 1

a. Bracteole median (often above middle in glaucescens). Anthers

sessile or stalked

5, Anthers sessile (very shortly stalked in latericia)

c. Stigma with many laciniate lobes (more than 6), sessile.

Anthers 9-20. Bark of twigs tending to crack or to scale

off. Apical portions of twigs more than 2 mm. thick and

tomentose or lanose. Leaves never elliptic; nerves more

than 16 pairs. Fruit tomentose or lanose

d. Flowers the largest in the genus, densely pale brown-

lanose outside; male 1 cm. long, female 1-7-2 cm.

long; fruit the largest in the genus, densely pale brown-

lanose, the hairs up to 5 mm. long, 4-5—8 cm. long and

3-4-5 cm. broad, apical portion of twigs and petioles

of younger leaves also covered with the same pale-

brown wool (1) K. hookeriana

d. Flowers smaller, tomentose but not lanose, the tomentum

much shorter; fruit smaller, tomentose but not lanose.

Apical portion of the twigs and the petioles of young

leaves scurfy-tomentose but not lanose

e. Leaves large (sometimes small ones at the tips of the

twigs), 10-50 cm. long and 3—14 cm. broad; nerves

24-35 pairs. Male flowers pale yellowish-brown out-

side, 6-7 mm. long; female also pale yellowish-

brown outside, 1 cm. long; male pedicels 7-10 mm.

long, female nearly sessile, 2 mm. long. Fruit pale

yellowish-brown, oblong, 3:5-4:5 cm. long and 3

cm. broad; stalk 4-5 mm. long (2) K. furfuracea

e. Leaves smaller, 12—20 cm. long and 3—5 cm. broad;

nerves 9—21 pairs. Male flowers reddish-brown out-

side, 5 mm. long; female same colour, 6 mm. long;

male pedicels 3-4 mm. long, female sessile. Fruit

reddish brown, pear-shaped, 1:7 cm. long and 1:5

cm. broad, sessile. (3) K. latericia

262

Vol. XVI. (1958).

GrRouP 2

c. Stigma few-lobed (6), sessile. Anthers very many, over 30

and up to 45. Bark of twigs not tending to crack. Apical

portions of twigs 2 mm. thick and glabrous. Leaves

elliptic, elliptic-oblong or elliptic-lanceolate; nerves 9-16

pairs. Fruit becoming glabrous and black when dry

(4) K. curtisii

GRouP 3a

b. Anthers stalked (stigmas few-lobed, 2—6-lobed)

f. Fruit ovoid or globose. Leaves rigidly coriaceous or

coriaceous, reticulations very distinct on both surfaces,

oblong or oblong-lanceolate with more or less parallel

sides, never elliptic. Twigs tomentose or pubescent, the

_ tomentum extending down a short distance. Male flowers

in bud 4—6 mm. in diam.

g. Mountain trees of high altitude, 1,000 m. or over. Fruit

globose, 3 cm. in diam., nearly sessile or stalk not

over 3 mm. long (5) K. rigidifolia

g. Not mountain trees of high altitude

h. Twigs striate at the tips and there shortly but densely

furfuraceous-tomentose. Leaves pubescent beneath,

the pubescence tending to persist; reticulations very

distinct. Male flowers numerous, 30—40 or more in

a cluster, densely rusty-tomentose; pedicels 7 mm.—1

cm. long; anthers 13-18. Fruit 3-5-4:5 cm. long;

stalk 1 cm. long (6) K. conferta

h. Twigs not striate at the tips and there slightly

pubescent. Leaves scarcely pubescent beneath except

when young, mostly glabrous; reticulations less

prominent. Male flowers 7—10 in a cluster; pedicels

1—1-2 cm. long; anthers 10-14. Fruit 2-5-3 cm. long;

stalk over 1 cm. long (7) K. scortechinii

Group 3b

f. Fruit obovoid, ellipsoid or oblong. Leaves coriaceous or

not, reticulations distinct or not, the sides parallel or not,

sometimes elliptic. Twigs tomentulose or glabrous. Male

flowers in bud generally smaller, 2-5 mm. in diam.

i. Twigs reddish brown. Leaves coriaceous or not. Fruit

obovoid or ellipsoid, drying reddish-brown, tomentu-

lose, 1-5-2:5 cm. long and 1-1-5 cm. broad

263

Gardens Bulletin, S.

j. Reticulations of leaf visible on both surfaces

k. Male pedicels 7 mm.—1-5 cm. long; anthers 9-13.

Leaves up to 19 cm. long, often elliptic

1, Leaves very coriaceous, elliptic or oblong-elliptic;

reticulations very close and raised on both sur-

faces; nerves 7-14 pairs. Male pedicels 7 mm.

—1 cm. long. Fruit ellipsoid, 2—2:'5 cm. long

and 1-5 cm. broad (8) K. kunstleri

1, Leaves membranous, oblong lanceolate to narrow-

ly elliptic; reticulations very close but not so

prominent or raised on both surfaces, finer on

the lower; nerves 14—20 pairs. Fruit obovoid,

smaller, 1:5—2 cm. long and 1-1-2 cm. broad

(9) K. malayana

k. Male pedicels 4-5 mm. long; Anthers 7—9. Leaves

up to 25 cm. long, narrower, not elliptic

(10) K. communis

GROUP 3c

j. Reticulations very fine or invisible on both surfaces

m. Twigs very slender 1 mm., thick at apex and 2 mm.

thick 10 cm. lower down, shining, dark brown

to blackish. Leaves 1:5—4 cm. broad, nerves faint

beneath. Male flowers 2—2:5 mm. in diam.; anthers

6-8; male pedicels 2-5 mm. long. Fruit obovoid

(11) K. stenophylla

Group 3d

m. Twigs stout, 3 mm. thick at apex and 4 mm.

thick 10 cm. lower down, dull, dark brown.

Leaves 3—7:5 cm. broad; nerves distinct beneath.

Male flowers 4—5 mm. in diam.; anthers 10—14;

male pedicels 5 mm.—1 cm. long. Fruit ellipsoid

(12) K. glaucescens var. patentinervia

i. Twigs less often reddish, but straw-coloured or greyish

with blackish or reddish brown portions here and there.

Leaves coriaceous. Fruit oblong, tomentulose, glabre-

scent or glabrous and often blackish brown when

mature, 25-45 cm. long and 1-2:5 cm. broad

(1:3-1:5 cm. x 1—1:3 cm. in glaucescens var. paten-

tinervia )

n. Leaves not cordate at the base, twigs often slender

and 1-3 mm. thick at the apex

264

Vol. XVI. (1958).

o. Fruit 2‘5—4-5 cm. long and 1—2-5 cm. broad. Nerves

of leaf generally oblique; reticulations on upper

surface visible or not

p. Leaves 4-9 cm. broad; midrib dull beneath;

reticulations of upper surface often distinct or

visible. Anthers 12—17; male pedicels 8 mm.—1

cm.long (12) K. glaucescens var. glaucescens

p. Leaves 3—5 cm. broad, midrib beneath shining

and reddish brown; reticulations above invisible

or very faint. Anthers 7-12; male pedicels 3-4

mm. long (12) K. glaucescens f. rubens

o. Fruit 1:3—1-5 cm. long and 1—1:3 cm. broad. Nerves

of leaf patent, leaving midrib at an angle of 70°—

90°; reticulations above invisible

(12) K. glaucescens var. patentinervia

n. Leaves cordate at the base, twigs 5 mm. thick at the

apex, reddish or greyish brown

(12) K. glaucescens var. cordata

GrouP 4a

a. Bracteole at base of the perianth. Anthers sessile

q. Stigma many-lobed

r. Staminal disc mammillate. Young twigs underneath the

tomentum smooth and glossy, dark reddish brown. Leaves

11-30 cm. long and 3-8-5 cm. broad

s. Petioles of the apical leaves and the apical portions of

the twigs densely covered with dark, rusty wool, the

individual hairs 2 mm. long. Male flowers 6-7 mm.

iong, female 1 cm. long, both densely dark rusty-lanose

outside; male pedicels 8 mm.—1 cm. long and 2-3

mm. thick, female 3 mm. long and 4-5 mm. thick.

Fruit rusty-lanose, perianth persisting in fruit

(13) K. plumulosa

s. Petioles of the apical leaves and the apical portions of

the twigs covered with very short rusty-stellate scurf,

not lanose, soon glabrous. Male flowers 4—5 mm. long,

female 5-7 mm. long, both rusty-tomentose outside;

male pedicels 8 mm.—1 cm. long and 1 mm. or less

thick; female 7-8 mm. long and 1:5—2 mm. thick.

Fruit not lanose but covered with rusty-stellate scurf,

becoming glabrescent, perianth not persisting

(14) K. intermedia

265

Gardens Bulletin, S.

GrouP 4b

r. Staminal disc not mammillate. Young twigs not glossy or

smooth but greyish brown except the shortly tomentose

tips. Leaves very large 30-55 cm. long and 8:5—21 cm.

broad

t. Leaves obtuse and often retuse at the apex; lower sur-

face covered with minute brown scales. Fruit 5-6 cm.

long and 5 cm. broad, beaked at the apex and gibbous

at the base on one side. Flowers unknown

. (15) K. retusa

t. Leaves acute and not retuse at the apex; lower surface

not covered with minute brown scales. Fruit 3—4:5

cm. long and 2—3-5 cm. broad, not beaked at the apex

and not gibbous (16) K. mandaharan

GrRouP 5

q. Stigma few-lobed, 2—6 lobed

u. Leaves glabrous at least when adult

v. Leaves large, rounded or cordate at the base, 20—43

cm. long and 9-16 cm. broad; nerves about 30 pairs.

Male flowers stalked (17) K. oblongifolia

vy. Leaves smaller, acute at the base, 15—25 cm. long and

1-5-7 cm. broad; nerves much fewer than 30 pairs.

Male flowers stalked or sessile

w. Mountain trees of altitude 1,000 m. or over. Twigs

often reddish-brown and glossy. Leaves thinly

coriaceous or coriaceous, 15—25 cm. long and 4~7

cm. broad; midrib not striate beneath when dry;

nerves 18—25 pairs. Male flowers sessile or on a 1

mm. long pedicel. Fruit 2-3 cm. long; stalk 3-5

mm. long, stout

(17) K. oblongifolia var. monticola

w. Not mountain trees. Twigs usually blackish brown,

smooth or slightly rough. Leaves membranous, 8—17

cm. long and 1:5—5 cm. broad; midrib striate be-

neath when dry; nerves 13-18 pairs. Fruit 1-8-2

cm. long; stalk 5 mm.—1 cm. long, slender

(18) K. globularia

u. Leaves stellate-scurfy pubescent or tomentose especially

beneath, the indumentum varying a good deal (some-

times almost glabrous forms are found) (19) K. laurina

266

Vol. XVI. (1958).

KEY TO KNEMA

(Sterile material and fruits)

a. Apex of twigs, the petioles of young leaves and the fruits

densely lanose with hairs 2-5 mm. long

b. Tomentum of these parts pale yellowish-brown. Bark of

twigs tending to crack or to scale off, rough, blackish or

greyish-brown when not covered with tomentum. No stilt

roots. Leaves 30—65 cm. long and 9-5—13 cm. broad; nerves

20-30 pairs. Fruit 45-8 cm. long and 3-45 cm.

broad; perianth not persisting in fruit (1) K. hookeriana

b. Tomentum of these parts reddish-brown. Bark of twigs not

tending to crack or to scale off, smooth, shining, reddish-

brown. Stilt roots often present. Leaves 14-30 cm. long and

S—8:5 cm. broad; nerves 15—20 pairs. Fruit 4 cm. long and

2—2:5 cm. broad; perianth persisting in fruit

(15) K. plumulosa

a. Apex of twigs, the petioles of young leaves and the fruits not

densely lanose but tomentose, pubescent or glabrous; hairs

less than 2 mm. long

c. Bark of young twigs tending to crack or to scale off. Bark

on trunk also scaly and tending to flake off

d. Twigs very stout at the apex, 5 mm. thick. Leaves large

(sometimes small, narrow ones at the tips of twigs) 10-50

cm. long and 3—14 cm. broad (average 10-14 cm. broad);

nerves 24-35 pairs. All tomentum pale yellowish-brown.

Fruit oblong, 3-545 cm. long and 3 cm. broad; stalk

4-5 mm. long (2) K. furfuracea

d. Twigs more slender at the apex, 2-3 mm. thick. Leaves

smaller, 12-20 cm. long and 3—5 cm. broad; nerves

9-12 pairs. All tomentum reddish-brown. Fruit pear-

shaped, 1-7 cm. long and 1-5 cm. or less broad, sessile

(3) K. latericia

c. Bark of young twigs not tending to crack or to scale off.

Bark on trunk tending to scale or not (usually not)

e. Leaves large, 35-55 cm. long and 14—21 cm. broad,

often retuse at the apex; lower surface pale brown

with minute scales. Fruit 5-6 cm. long, beaked at the

apex and gibbous at the base on one side

(17) K. retusa

e. Leaves smaller, 6-25 cm. long and 2:5—7 cm. broad,

not retuse at the apex; lower surfaces covered with

267

Gardens Bulletin, S.

minute, rusty-stellate scales when young only, later

glabrous. Fruit obovoid 1:8 cm. long, not beaked at

the apex and not gibbous on one side at the base

(14) K. communis

e. Leaves large or not, not retuse at the apex and not

covered with white or brownish scales (exception

young leaves of K. communis see above, this species

again included in key later in case leaves have become

glabrous or do not show the scales)

f. Leaves large, the average usually over 30 cm. and

up to 50 cm. long; breadth 8-5-17 cm.

g. Leaves membranous, twigs reddish brown and shin-

ing. Sides of leaf nearly parallel; nerves depressed

above, often with a slight bullate appearance, 30

pairs. Fruit sub-obovoid or ovoid, pointed at the

apex (20) K. oblongifolia

g. Leaves stoutly coriaceous, twigs not shining nor

smooth, greyish brown except the rusty, shortly

tomentose tips. Sides of leaf not parallel; nerves

raised above and can be felt by rubbing with the

finger, 22—25 pairs. Fruit not sub-obovoid or

ovoid and not pointed at the apex

h. Leaves rounded at the base but not cordate,

medium brown and shining above when dry;

midrib at base on upper surface raised;

scalariform reticulations below distinct; distance

between two nerves at the middle of the leaf

1:5—2:3 cm. Fruit tomentose, medium-brown

when dry, 3—4:5 cm. long and 3-3-5 cm. broad,

its stalk 5—7 mm. long and 5 mm. thick

(18) K. mandaharan

h. Leaves rounded at the base and slightly cordate,

rather dark and dull above when dry; midrib

at base above not raised but flush with the upper

surface; scalariform reticulations visible below

but very fine; distance between two nerves at

the middle of the leaf 1-1-3 cm. Fruit tomentu-

lose, dark chocolate-brown when dry, 3-3-5

cm. long and 2-2-5 cm. broad, more oblong;

stalk 7 mm.—1-—3 cm. long and thinner, 2 mm.

thick (19) K. glaucescens var. cordata

268

Vol. XVI. (1958).

f. Leaves smaller, occasionally up to 30 cm. long

i. Mountain trees of altitude 1,000 m. or over

j. Leaves stoutly coriaceous, rounded or slightly

cordate at the base. Young twigs not angled

and not smooth or shining. Fruit globose, 3 cm.

in diam., sub-sessile; stalk 3 mm. long

(5) K. rigidifolia

j. Leaves thinly coriaceous to membranous, usually

acute at the base, occasionally rounded. Fruit

sub-obovoid or ovoid, pointed at the apex, 2

cm. long; stalk 5 mm. long

(20) K. oblongifolia var. monticola

i. Not mountain trees of high altitude

k. Leaves stellate-pubescent or stellate-tomentose

beneath, the degree of tomentum varying very

much, nearly glabrescent in some forms of

laurina. The leaves of scortechinii are pubescent

when young and later become glabrous (Species

Nos. 6, 7, 22 are difficult to distinguish when

sterile)

l. Reticulations of leaves above very prominent

and areolate, veins slightly raised above; sides

of leaves not usually parallel. Tips of twigs

striate and closely tomentose. Fruit ovoid,

3-5-4 cm. long and 2-2-5 cm. in diam.; stalk

1 cm. long (6) K. conferta

1. Reticulations of leaves above less prominent;

sides of leaves nearly parallel, the leaf soon

becoming glabrous beneath. Tips of twigs not

Striate, pubescent for a short distance down-

wards, 5 cm. or so. Fruit ovoid-globose,

2-5-3 cm. long and 2-2-3 cm. diam.; stalk

over 1 cm. long (7) K. scortechinii

1. Reticulations of leaves above finer still, veins

sunk above; sides of the leaf parallel or not.

Twigs usually not or only faintly striate at

the tips but the tomentum, if dense, often

erect. Fruit very variable in shape and tomen-

tum, ovoid, ellipsoid or oblong, 2-5-3 cm.

long and 1-5—2 in diam.; stalk 2—3 mm. long

(22) K. laurina

269

Gardens Bulletin, S.

k. Leaves glabrous beneath (K. communis has

minute, brown, stellate scales when young)

m. Twigs often pale straw-coloured

n. Twigs stout, 3-4 mm. thick at 10 cm. down

from apex. Leaves, average breadth 5-9

cm.; whitish beneath with dark brown or

blackish veins when dry. Fruit glabrous and

blackish when dry

(8) K. glaucescens var. glaucescens

n. Twigs more slender, 1—2 mm. thick at 10

cm. down from apex. Leaves, average

breadth 1-8-3-5 cm., not whitish beneath

but reddish-brown or pale straw-coloured.

Fruit blackish or not when dry

o. Twigs shining and smooth, not striate.

Leaves elliptic-oblong or elliptic-lanceo-

late, the sides not parallel; length 4-5—

14-5 cm.; nerves. 9-16 pairs. Fruit finally

glabrous, drying blackish, 3-5 cm. long

and 3 cm. in diam., narrowed at each

end; stalk 1-1-3 cm. long

(4) K. curtisii

o. Twigs not shining nor smooth, striate.

Leaves lanceolate, the sides nearly

parallel; length 9-20 cm.; nerves 17—20

pairs, reddish beneath when dry. Fruit

minutely rusty-pubescent, oblong, obtuse

at both ends, 2—2:2 cm. long, 1:5—-1-7

cm. broad; stalk 3-5 mm. long

(8) K. glaucescens f. rubens

m. Twigs not pale straw-coloured

p. Young twigs smooth, shining, reddish-brown

q. Nerves and reticulations of leaf very dis-

tinct above and below; length 11—22 cm.,

occasionally up to 30 cm.; breadth 3-

4:5 cm. Fruit ellipsoid, 3-4 cm. long and

1-8—2:2 cm. broad; stalk 8 mm. long and

3—4 mm. thick. Stilt roots present

(16) K. intermedia

270

Vol. XVI. (1958).

q. Nerves and reticulations of leaf not visible

above, faint below; length 7-17 cm.;

breadth 1:5—4 cm. Fruit obovoid, 1:3

cm. long and 1 cm. broad; stalk 1-2

mm. thick. No stilt roots.

(12) K. stenophylla

p. Young twigs dull, rough or striate, black,

reddish or greyish-brown

r. Leaves elliptic or elliptic-oblong to ovate;

sides not parallel; reticulations very pro-

minent, raised and close on both surfaces

when dry; nerves beneath drying reddish-

brown. Fruit ellipsoid 2—2-5 cm. long and

1-5 cm. broad (9) K. kunstleri

r. Leaves more elongate or more lanceolate,

to oblong-lanceolate; sides nearly parallel;

reticulations never so prominent but often

very close and fine. Fruit ellipsoid or not,

mostly obovoid

s. Leaves coriaceous

t. Leaves drying greenish or greenish-

brown above, brown stellate scales

present beneath when young; reticu-

lations above very fine and close but

visible; nerves 17—20 pairs, oblique.

Fruit obovoid, narrowed towards the

base, 1-8 cm. long and 1-1:3 cm.

broad; stalk 1 cm.—1:5 cm. long

(14) K. communis

: Leaves drying a rich brown above, no

stellate scales beneath; reticulations

above invisible or scarcely visible;

nerves 21-25 pairs, very patent.

Fruit ellipsoid, narrowed at the base

and slightly at the apex, 1:3—1-5 cm.

long and 1-1-3 cm. broad; stalk 7

mm. long

(11) K. glaucescens var. patentinervia

s. Leaves thinly coriaceous

u. Leaves usually drying olive-green or

‘blackish-green above with a metallic

lustre, brownish beneath (this

271

Gardens Bulletin, S.

character not always reliable); 1-5-5

cm. broad; midrib striate longitudi-

nally beneath when dry, often minu-

tely pubescent; reticulations beneath

extremely fine and close, not raised,

sometimes scarcely visible. A tree of

rocky seashores in Malaya. Fruit

sub-globose (21) K. globularia

u. Leaves usually drying brownish-green

above, whitish or brownish-white be-

neath (this character not always re-

liable); 3-7 cm. broad; midrib not

striate beneath, glabrous; reticula-

tions beneath fine but visible. A tree

of lowland forest. Fruit obovoid

(10) K. malayana

(1) K. hookeriana (Hk. f. et Th.) Warb., Monog. Myrist. (1897)

551 T. 24 Figs 1-4; Gamble, Mat. F.M.P. 5, 23 (1912) 237;

Ridley, F.M.P. 3 (1924) 67; Corner, Wayside Trees of Malaya

1 (1940) 476.

Basionym: Myristica hookeriana Hk. f. et Th., Fl. Ind. 1

(1855) 156; A. DC., Prodr. 14, 1 (1856) 204; Miq., Fl. Ind.

Bat. 1, 2 (1858) 69 et Suppl. (1861) 384; Hk. f., Fl. Br. Ind.

5 (1886) 109; King in Ann. Roy. Bot. Gard. Calc. 3 (1891)

325 Pl. 163.—Fig. 1. Plate IA.

Tree 6-30 m. high, crown conical with spreading, deflexed

branches. Bark blackish brown or greyish brown, not furrowed but

flaky, the flakes in a longitudinal direction, sap copious, watery,

red. Twigs stout with rough,)flaky bark, glabrous except at the tips

which are covered with a light-buff wool. Leaves glabrous except

when young, when they are densely covered with a stellate, woolly

tomentum, coriaceous, dark green and glossy above, with the raised

nerves and broad midrib pale green; glaucous beneath, lanceolate

or oblanceolate, acute at the apex, widest above the middle, grad-

ually narrowed to the rounded base; nerves 20—30 pairs, commonly

28, very prominent on both surfaces, interarching at the extreme

edge; reticulations visible on both surfaces, scalariform between

the primary nerves, not straight; length 30-65 cm.; breadth 9:-5—13

cm.; petiole 2:5 cm. long, stout, at first invested with tawny wool,

slowly becoming glabrescent. Flowers in clusters on woody tuber-

cles on the branches, pyriform in bud, densely tawny-lanose out-

side, glabrous and bright red inside, very coriaceous. Male: on

272

Vol. XVI. (1958).

Fig. 1. Knema hookeriana (Hk. f. et Th.) Warb.

A, Leaf. B, Apex of twig. C, Female flowers. D, Female flower from

above. E, Female flower dissected to show ovary. F, Fruit. G, Fruit

cut open to show aril. H, Aril. I, Seed. J, Male flower. K, Staminal

column. A—H from living material of tree at Potting Yard, Botanic

Gardens, Singapore. J and K from Sinclair 8907.

273

Gardens Bulletin, S.

1-2 cm. long pedicels with an obtuse, ovate, decurrent, 5 mm.

long bracteole at the middle of the pedicel; perianth 1 cm. long

with longitudinal furrows inside; staminal column 4 mm. long,

bearing 15-18-(20) sessile, curved anthers round the disc.

Female: pedicels 1:5—2 cm. long, perianth as in the male but

larger, 1-7-2 cm. long, furrowed, 3, sometimes 4-lobed, the lobes

reaching down half way and having two concavities, one touching

the ovary, a ridge present above this concavity and the second just

under the thickened apex of the lobes; ovary 5 mm. long, sub-

globose, woolly; stigma 2 mm. long, greenish white, disc-shaped,

sub-sessile with many minute teeth but split on one side, giving the

appearance of two main lobes. Fruit ellipsoid, 4-5-8 cm. long, 3—4:5

cm. broad and pericarp 8 mm.—1 cm. thick, densely covered with

buff wool. Aril thin, scarlet, enveloping the seed, slightly fimbriate

at the apex. Seed smooth, testa greyish black.

KepaH: F.D. 17908 (SING); Mohamad F.D. 17570 (SING); Gu-

nong Baling, Best S.F.N. 21260 (SING); Teloi F.R., Baling, Shariff b.

Omar K.F.N. 66387 (KEP).

PENANG: Wall. Cat. 6802 (BM, K, NY); Wall. Cat. 6802A (A, BM,

CAL, DD, E, G & Prodr., K holotype, M); Telok Bahang, Curtis

2479 (BM, CAL, K, SING); Government Hill, Curtis 202 (CAL, K,

SING) and Nauen S.F.N. 35848 (K, SING); Maingay sine loc. 1279

(CAL, -K;.L).

PERAK: Gopeng, King Nos. 5754 (BM, CAL, FI, G, K, L, MEL,

SING) and 6007 (CAL, DD, E, FI, G, K, P, SING, UPS); Larut,

King 6656 (CAL, FI, G, K, L, MEL); Segari Melintang F.R., Mat

Hassan K.F.N. 69416 (KEP); Saiong F.R., Kinta, Mohd Rani K.F.N.

65949 (KEP).

TRENGGANU: Bukit Kajang, Kemaman, Corner, 12th April, 1935

(SING).

PaHANG: Ayer Surin, Pulau Tioman, Henderson S.F.N. 21684 (A,

DD, SING). |

SELANGOR: 18th mile, Ginting Simpah, Strugnell F.D. 12895 (KEP,

SING).

NEGRI SEMBILAN: Sungei Menyala F.R., Wyatt-Smith K.F.N. 64076

(KEP).

Matacca: Griffith 4342 (K); Maingay 1279 (A, K).

JouorE: Gunong Banong F.R., Batu Pahat, Suliman b. Manja K.F.N.

70213 (KEP).

SINGAPORE: Raffles 1382 (G); Anderson 10 (BM, CAL, MEL);

Wood Neuk Garden, Hullett 314 (CAL, SING); Bukit Timah, Ridley

Nos. 3701 (CAL, SING) and 3362 (SING); Burkill S.F.N. 2032 (K,

SING); Furtado, 22nd Sept., 1929 (SING); Seletar, Ridley 4813

(BM, K, SING); Botanic Gardens, Cantley 13 (SING); Botanic Gard-

ens’ Jungle, Burkill, date 1913 (SING) and Cantley 2904 (K); Band-

stand, Botanic Gardens, Ridley 2109 (CAL, SING); Lawn S, Nur

S.F.N. 26166 (DD, K, SING); Lower Ring Road, Nur Gardens No.

402 (SING).

DISTRIBUTION: Borneo and Sumatra.

There should be no difficulty in distinguishing this species from

the others. It will stand out in the forest from many trees by its

large leaves. The buff wool on the young twigs, petiole and young

274

Vol. XVI. (1958).

leaves, on the flowers and on the fruit at once serve to identify it.

The flowers and fruit are larger than those of any of the other

species. The large leaves recall those of K. furfuracea but those of

hookeriana are larger still and not cordate at the base. The bark

tends to flake and is very similar in this respect to that of K. fur-

furacea and latericia. These three species fall in one distinct group.

(2) K. furfuracea (Hk. f. et Th.) Warb., Monog. Myrist. (1897)

581 T. 24 Figs 1-2; Gamble, Mat. F.M.P. 5, 23 (1912) 245;

Ridley, F.M.P. 3 (1924) 70; Corner, Wayside Trees of Malaya

1 (1940) 476.

Basionym: Myristica furfuracea Hk. f. et Th., Fl. Ind. 1

(1855) 159 [non A. DC., Prodr. 14, 1 (1856) 206 — K. plu-

mulosa]; Mig., Fl. Ind. Bat. 1, 2 (1858) 70; Hk. f., Fl. Br. Ind.

5 (1886) 112; King in Ann. Roy. Bot. Gard. Calc. 3 (1891)

J18-P1. 155.

Synonym: Myristica longifolia Wall. ex Bl. sensu Hk. f. et Th.,

Fl. Ind. 1 (1855) 156 quoad sp. Penins. Malayanae tantum.—

Fig. 2. Plate IB.

Tree 10-20 m. high, pyramidal in outline with many slender,

horizontal branches which often hang down; lowest branches about

4 m. from the ground. Bark greyish-brown, of no great depth, not

furrowed, outer layers thin and brittle, flaking longitudinally into

fairly long strips, wood white; sap watery, pink, copious. Twigs

stout, even at the densely furfuraceous, light-brown apices, the older

portions with the bark tending to crack. Leaves coriaceous, dark-

green above and rather dull, glaucous beneath, the whitish-green

midrib and veins standing out on both surfaces, glabrous except

the very young leaves, oblong-lanceolate, widest above the middle,

apex acute, base gradually narrowed and cordate; nerves 24-35

pairs, curving gradually and anastomosing faintly at the margins,

an occasional secondary nerve present between the main ones;

reticulations mostly scalariform, at right angles to the main nerves

but not straight, forming a network with other finer irregular ones,

faint above, more distinct below; length 10—50 cm.; breadth 3-14

cm. but variable in size on the same tree; petiole stout, 3-3-5 cm.

long. Inflorescence from the axils of fallen leaves, forming warted

accrescences (6 or 7 often fused in one group, 7 mm. long, their

apices free). Male flowers: pedicels 7 mm.—1 cm. long, bearing a

minute, obtuse bracteole near the middle; perianth 6—7 mm. long

and 4-5 mm. in diam., pale brown, stellate-tomentose outside,

glabrous inside, divided into 3 blunt, triangular teeth; androecium

275

Gardens Bulletin, S.

Fig. 2. Knema furfuracea (Hk. f. et Th.) Warb.

A, Leafy twig. B, Leaf. C, Twig with female flowers. D, Female flower

from above. E, Female flower dissected to show ovary. F, Fruit. G,

Dehiscent fruit showing aril and seed. H, Aril. I, Seed. J, Seed in

longitudinal section. K, Male flower. L, Staminal column. A—J from

living material of tree in Arboretum, Botanic Gardens, Singapore,

Sinclair S.F.N. 39488.

276

Vol. XVI. (1958).

a stout, furrowed stalk with its disc circular or slightly 3-angled;

anthers 10-13, sessile. Female flowers in clusters, nearly sessile,

their pedicels about 2 mm. long; perianth 1 cm. long, densely

light-brown stellate-tomentose outside, glabrous, bright-red and

minutely papillose inside, coriaceous, the lobes 3, occasionally 4,

oblong, split down to more than half way, obtuse at the apex

with a thickened, often incurved projection (breadth of lobe

6 mm.); ovary globose with chocolate-coloured, stellate tomentum,

3-4 mm. in diam.; stigma sessile, 1 mm. long and 1-5-2 mm.

across, disc-shaped with many acute lobes or teeth, depressed in

the centre. Fruit oblong, rounded at each end, pale yellowish-brown

and thinly tomentose, 3-5—4-5 cm. long and 3 cm. broad, ridged

longitudinally with a circumferential dehiscence suture, pericarp

1 cm. thick; stalk 4-5 mm. long. Aril fimbriate at the apex only,

crimson. Seed 2 cm. long, oblong, outer coat succulent, pale pin-

kish-white; inner hard, dark brown or nearly black.

KepAH: Bukit Tanjong, Haniff S.F.N. 13147 (BRI, SING, UC); Gu-

nong Raya, Pulau Langkawi, Haniff S.F.N. 15510 (SING) and Corner,

15th Nov., 1941 (SING); Penarah F.R., Mat. F.D. 12372 (KEP); Ulu

Sungei Tawar, Boswell F.D. 12595 (KEP).

PENANG: Herb. Hook., no data (A, K) type material; Phillips, date

1824 (K); Waterfall, Curtis 2827 (CAL, K, SING); Government Hill,

Curtis 1459 (CAL, K, SING); Penang Hill, Kiah S.F.N. 35330 (A,

BM, BRI, DD, KEP, K, SING); Moniot Road, Curtis 2456 (CAL,

SING); Pulau Boetong, Curtis 2769 (BM, CAL, K, SING).

PERAK: Larut, King Nos. 5600 (BM, CAL, DD, FI, K, MEL,

SING); and 7551 (CAL, K, SING, UPS); Gopeng, King Nos. 5819

(CAL, DD, G, K, L, MEL); 6025 (CAL, FI, G, K, L, SING) and

6059 (CAL, FI, G, L, UC); Chanderiang, King 5720 (CAL, DD,

K, MEL); Ulu Kal, King 10349 (CAL, DD, E, K); Tapah, Haniff

14260 (K, SING); Pulau Lalang, Wyatt-Smith K.F.N. 76532 (KEP);

Sungei Siput, Jinal F.D. 8831 (KEP); Parit F.R., Kinta, K.F. Nos.

54665 (KEP); 54683 (KEP); 54685 (KEP); 54696 (KEP); 54806

(KEP); 54808 (KEP) and 54826 (KEP).

TRENGGANU: Bukit Kajang, Corner S.F.N. 25916 (A, CAL, K,

KEP, SING).

PAHANG: Tahan River, Ridley 2261 (BM, SING); Raka Hill Forest,

Mat Zin F.D. 27513 (KEP).

SELANGOR: Weld Hill, Hamid C.F. 934 (KEP, SING); Sinclair

S.F.N. 40080 (E, SING); Abdul Rahman C.F. 2802 (KEP); Klang

Gates, Burn-Murdoch 101 (BM, KEP, SING); Bukit Cheraka F.R.,

Walton F.D. 28396 (DD, KEP, SING); Bukit Lagong, Motan K.F.N.

70460 (KEP).

NEGRI SEMBILAN: Seremban Road, Tampai, Nur S.F.N. 1424 (K,

SING).

Matacca: Griffith 4346 (A, CAL, K, P) type material; Maingay

1288 (CAL, G); Maingay 1287 (CAL, K, L) type material of var.

major; Ayer Panas, Derry 152 (SING); Tubong, Goodenough 2001

(CAL, P, SING); Batu Tugali, Holmberg 773 (SING).

277

Gardens Bulletin, S.

JoHORE: Sungei Sedili, Corner, 28th Aug., 1932 (SING).

SINGAPORE: Arboretum, Botanic Gardens, Sinclair S.F.N. 39488

(BK, BO, DD, E, K, L, P, PNH; SAN, SING); Nur Gardens’ Herb.

No. 1466 (SING); Furtado, 11th Jan., 1928 (SING).

DISTRIBUTION: Siam, Anamba Islands, Sumatra, N. Borneo, Labuan.

The pyramidal shape of the tree and the flaking nature of the

bark are similar to that of K. hookeriana. The slightly smaller

leaves with cordate base and non-lanose tomentum will however

distinguish it from that species. They may recall specimens of K.

latericia having larger ones than normal but the cordate leaf base

will at once distinguish it from that species as well as from most of

the other large-leaved Malayan ones.

Wall. Cat. 6810 listed as K. glaucescens (non Jack) and quoted

as a synonym of K. furfuracea in Fl. Ind. P. 157, was named K.

intermedia var. dubia Warb. It is actually K. plumulosa Sinclair,

a species which Warburg and others have wrongly interpretated

as K. cantleyi.

King’s var. major of K. furfuracea is not really a variety but

simply typical K. furfuracea. The leaves, supposed to be larger,

are not different. The leaves on the same tree of K. furfuracea vary

much in size, those at the apex of the twigs tend to be smaller.

Collectors often take the smallest leaves in order to fit them con-

veniently to the size of their collecting presses or vascula. King

6515 quoted by Gamble as K. furfuracea is K. mandaharan.

(3) K. latericia Elm., Leafl. Philip. Bot. 5 (1913) 1815.

Synonyms: K. meridionalis J. Sinclair in Gardens’ Bull. Singa-

pore 13, 2 (1951) 297 Fig. 1. Myristica ridleyi Gandoger in

Bull. Soc. Bot. France 66 (1919) 226 in clavi—Fig. 3. Plate

IIA.

Shrub or small tree 5-10 m. high, pyramidal in outline. Bran-

ches numerous, slender, sometimes hanging down almost to the

ground. Bark reddish brown, of no great thickness, rough, crack-

ing in fairly large elongated portions but not furrowed. Twigs

when young and also the young foliage densely furfuraceous-stel-

late-tomentose, the young leaves appearing at flowering; bark of

the older twigs tending to crack. Leaves, adult coriaceous, lanceo-

late, dark green above, glaucous beneath, apex acute, base acute

or sometimes rounded; midrib and veins conspicuously raised on

both surfaces, whitish-green when fresh, reddish in dried material;

nerves 9-21 pairs, curving and anastomosing at the margin; retic-

ulations scarcely visible above, distinct beneath, especially in

278

a ——

AWS

Z\\ G

JuRAin Dee

Fig. 3. Knema laterica Elmer.

A, Portion with leaves and male flower. B, Twig with female flowers.

C-—D, Female flowers top and side views. E-F, Male flowers top and

side views. G, Fruit. A, E and F from Sinclair S.F.N. 38914 (living

material). B—D from Sinclair S.F.N. 38915 (living material) G

from Sinclair S.F.N. 38561 (holotype).

Gardens Bulletin, S.

dried material; length 12—20 cm.; breadth 3—5 cm.; petiole 7 mm.—

1 cm..Jong, pubescent. Male flowers 3—5, arising from woody,

axillary tubercles; pedicels 3—4 mm. long, rusty-tomentose with a

minute median bracteole; perianth segments 3-4, about 5 mm.

long, coriaceous, ovate-rotund, reaching down to near the base of

the perianth, rusty stellate-tomentose outside, glabrous and pink

inside, apex sub-acute or obtuse, erect or slightly incurved; sta-

minal disc triangular or peltate, the stalk red, about 1 mm. long;

anthers 9-12, very shortly stalked, sub-erect. Female flowers ses-

sile in groups 7-10 on woody tubercles; perianth as in the male

but more coriaceous and rigid, 6 mm. long; stigma sessile, green,

funnel-shaped, the apex multi-lobed; ovary hemispherical, dark

brown, stellate-pubescent. Fruit about 1:7 cm. long and 1:5 cm.

or less broad, reddish-stellate-tomentose, pear-shaped with obtuse

apex, gradually narrowed to the base. Aril pale orange, laciniate

at the apex. Seed basal, black, filling the carpel.

KEDAH: Inas Forest Reserve, Kulim, Syed Harun K.F.N. 55759

(KEP).

PERAK: Scortechini (DD, MEL).

JoHorE: Sungei Sedili, Corner S.F.N. 29277 (K, KEP, SING); S.

Berassau, Corner S.F.N. 28975 (CAL, K, SING); 54 mile, Kota

Tinggi-Mawai Road, Corner S.F.N. 28711 (A, K, SING); 134 mile

Mawai-Jemaluang Road, Corner S.F.N. 29420 (K, SING); Sungei

Tementang, Mawai-Jemaluang Road, Corner S.F.N. 29290 (A, CAL,

K, KEP, SING); Kulai Young Estate, Corner S.F.N. 33598 (KEP,

SING); North of Gunong Blumut, Holttum S.F.N. 10606 (BRI, CAL,

K, SING); Bukit Tanah Alang, Lake & Kelsall 4012 (K, SING);

Bukit Abu Bakar, Nur & Kiah S.F.N. 7757 (K, SING); Mount Aus-

tin, Ridley, date 1906 (SING).

SINGAPORE: Murton 150 (K); South side of Mac Ritchie Reservoir,

Sinclair S.F.N. 38561 (K, SING holotype of K. meridionalis) and

Sinclair S.F.N. 39481 (B, BK, Delhi Univ., K, M, NY, SING); Bukit

Timah, Ngadiman S.F.N. 34957 (SING); Cantley, date 1885 (K) and

21 (K, SING); Cantley 69 sine loc. (K, SING); Chan Chu Kang,

Goodenough, 27th Nov., 1890 (SING); Ridley 6737 (SING); Kranji,

“Ridley 2043 (CAL, SING); Bukit Panjang, Ridley 12541 (SING);

Sungei Morai, Ridley 6450 (CAL, SING); Maranta Avenue, Botanic

Gardens, Sinclair S.F.N. 38914 (E, K, SING); Lawn 0, Botanic Gar-

dens, Sinclair S.F.N. 38915 (E, K, L, SING); Ridley 2037 (CAL,

SING); Furtado S.F.N. 21177 (K, NY, SING); Arboretum, Botanic

Gardens, Furtado S.F.N. 21153 (K, NY, SING) and 21154 (K, SING)

and Nur Gardens No. 1459 (SING); s.l. Ridley, date 1898 (LY) type

of Myristica ridleyi Gandoger.

DISTRIBUTION: Palawan and North Borneo.

Tyee: Elmer 12757, Mt. Pulgar, Palawan (BM, BRSL, CAL, E, G,

This Malayan species is common in Singapore in the Mac Ritchie

Reservoir Forest and in the Botanic Gardens’ Jungle. It has been

confused with K. furfuracea or left unnamed. It has smaller flowers

and fruit. Its leaves too, are smaller with fewer veins and are never

280

Vol. XVI. (1958).

cordate at the base as in the former. The midrib and veins are

conspicuously raised on the upper as well as the underside of the

leaf and are usually red in dried material. These characters will

distinguish it from sterile material of K. malayana. The flaking

bark of the twigs and trunk is a good diagnostic character and one

which groups it with hookeriana and furfuracea. There is a speci-

men in Kew, Agama 21612 from Palawan, named K. badia Merr.

and indicated as a type but it is not different from K. latericia. I

cannot find any record of its publication.

(4) K. curtisii (King) Warb., Monog. Myrist. (1897) 567 T. 25

Figs 1-2; Gamble, Mat. F.M.P. 5, 23 (1912) 240; Ridley,

F.M.P. 3 (1924) 70.

Basionym: Myristica curtisii King in Ann. Roy. Bot. Gard.

Calc. 3 (1891) 326 Pl. 167. M. sp. Hk. f., Fl. Br. Ind. 5 (1886)

113, Maingay 1301, foot note -—Fig. 4.

Tree 3-10 m. high. Bark characters not known. Twigs glabrous,

smooth, angled at the apices, slender, often shining, straw-coloured

to brown, never reddish. Leaves thinly coriaceous, narrowly ellip-

tic-oblong or elliptic-lanceolate to elliptic-obovate, the apex acute,

sub-acute or obtuse, the base acute, glabrous on both surfaces, the

upper surface olivaceous when dry, the lower paler, sub-glaucous,

nerves 9-16 pairs, rather faint, interarching near the margins;

reticulations very faint, close and fine above, less distinct still be-

neath; length 4:5-14-5 cm.; breadth 1-8-6 cm.; petiole 1-1-2 cm.

long, slender, glabrous. Male flowers in umbels of 3—10, on short

axillary tubercles; pedicels 8 mm—1 cm. long, puberulous with a

minute bracteole about the middle; perianth lobes about 4 mm.

long, broadly triangular, minutely puberulous; staminal disc tri-

angular, very concave (at first flat or slightly convex), bearing

35—45 sessile anthers, stalk 1 mm. long. Female flowers slightly

larger, 5 mm. long, the stalks stouter and 4-5 mm. long; ovary

_ sessile, ovoid, rusty-tomentose, length including the sessile, six

lobed stigma, about 2 mm. Fruit 3—5 cm. long and about 3 cm.

broad, at first rusty-puberulous, finally glabrous, slightly narrowed

at each end, single or in pairs with a circumferential, longitudinal

ridge of dehiscence; pericarp 3—4 mm. thick; stalk 1-1-3 cm. long.

Aril blood-red, entire or obtusely lobed at the apex, the lobes

fringed with minute teeth. Seed pale buff.

KEDAH: 48th mile Jeniang Road, Kiah S.F.N. 35975 (SING).

PENANG: Waterfall, Curtis 1024 (CAL, K, SING) type material;

Pulau Boetong Curtis 1024 (BM, SING); Penara Bukit, Curtis 1024

(K, SING) type material; near bungalow, Ridley 9374 (BM, SING).

281

Gardens Bulletin, S.

Fig. 4. Knema curtisii (King) Warb.

A, Leafy twig with male flowers. B, Leaf from Corner S.F.N. 37128.

C, Male flower. D, Staminal column. E, Staminal column, top view.

F, Fruit. A, C, D, E from Curtis 1024. F, from Henderson S.F.N.

21675.

282

Vol. XVI. (1958).

PERAK: Scortechini 292b (CAL, FI, G, K, L, SING) type material;

Haram Perah, Scortechini 763 (CAL); Hermitage, Curtis 13220 (K,

SING) type material; Taiping, Wray 2112 (CAL, K, SING) type

material; Ridley 14681 (BM, SING).

gaia Bukit Bauk F.R., Dungun, Wood K.F.N. 76076 (KEP,

PAHANG: Tanjong Duatah, Pulau Tioman, Burkill S.F.N. 1111 (SING):

ae ‘<a Pulau Tioman, Henderson S.F.N. 21675 (CAL, K,

NEGRI SEMBILAN: Gunong Angsi F.R., Kuala Pilah, Syed Ali F_D.

23674 (SING); Tampin Hill, Ridley 1854 (CAL, SING).

Maracca: Maingay 1301 (CAL, K, L) type material; Alvins Nos.

702 (SING) and 2240 (SING); Batu Tiga, Derry Nos. 1204 (SING)

and 1172 (CAL, K, P, SING).

JoHorRE: Pulau Aor, Henderson S.F.N. 18220 (K, SING).

SINGAPORE: Bukit Mandai, Goodenough 2376 (CAL, K, SING):

Mandai Road, Kiah S.F. Nos. 37148 (A, BM, BRI, K, KEP, KING)

and 37123 (A, KEP, SING); Corner, 15th April, 1934 (SING);

Jurong, Corner S.F.N. 26155 (CAL, K, SING).

DISTRIBUTION: Borneo, Simaloer Island near Sumatra and Sumatra.

This species is distinct from all the other Malayan ones by the

concave staminal disc which has a very large number of sessile

anthers, at least over 30 and usually 35-45. It seems to fall in a

group of its own with perhaps K. membranifolia Winkler as the

nearest relative. Sterile material of K. curtisii may possibly be con-

fused with K. globularia, but the leaves of curtisii are not so long in

proportion and the twigs are much lighter in colour. The flowers

are not tomentose but pubescent or puberulous and the pedicels

are more slender. The leaves are usually acute to sub-acute but

even on the same tree we find some which are obtuse. The Singa-

pore specimens have obtuse leaves and in Corner 26155 from

Jurong, Singapore, the leaves are smaller, 4 cm. long and 1-5-2

cm. broad, oblong-obovate and obtuse at the apex. Those from

Mandai Road, Singapore, have leaves which are larger and about

the same size as the type but less obovate than the Jurong ones.

They are obtuse but a few sub-acute ones are found on the same

twigs. As their flower structure is the same and since obtuse and

acute leaves can be found on the same specimens, these characters

cannot be called varietal.

The nearest probable ally. as already stated, is K. membranifolia

Winkler in Engl., Bot. Jahrbucher 49 (1913) 368. The texture of

the leaves, the venation and the colour of the twigs are similar in

both but in membranifolia the leaves are much larger, and nar-

rower in proportion, the twigs stouter, the flowers larger and more

densely tomentose, the staminal disc similar with fewer anthers

(17-23 in number and with distinct spaces between each other)

and the stalk of the fruit shorter and thicker.

283

Gardens Bulletin, S.

(5) K. rigidifolia J. Sinclair, sp. nov.—Fig. 5.

Species K. scortechinii et K. confertae valde affinis; a priore foltis

magis coriaceis, reticulationibus prominentioribus; ab altera flori-

bus masculis in fasciculo paucioribus, foliis adultis glabris; a duabus

illis, fructibus sessilibus vel fere differt.

Arbor ad 13 m. alta. Cortex brunneus laevis vel leviter fissus;

latex ruber. Ramuli crassi, apice ferrugineo-pubescentes, infra

glabri, conspicue striati. Folia crasse coriacea, rigida glabra, ob-

longo-lanceolata vel lanceolata, supera.nitida (in sicco modice

brunnea), subtus glauca, apice acuta, basi rotundata vel leviter

cordata, 16-28 cm. longa, 4-8-5 cm. lata; nervi 17—25 pares

obliqui, supra tenues, subtus prominentes ad marginem indistincte

anastomosantes; reticulationes densae, utrinque distinctae; petioli

1:5 cm. longi, crassi. Flores masculi immaturi tantum visi, 3 mm.

longi, pedicelli crassissimi 3 mm. longi, bracteola medio praediti;

perianthium coriaceum triquetrum minute ferrugineo-tomentosum;

discus staminalis concavus (deinde probabiliter planus), subtus

striatus, 15 antheris immaturis sessilibus (maturis probabiliter

stipitatis) coronatus. Flores feminei ignoti. Fructus ferrugineo-

tomentellus, 3 cm. in diam., subsessilis cum stipite 3 mm. longo;

pericarpium 3—4 mm. crassum.

Tree up to 13 m. high. Bark brown or greyish brown, smooth or

slightly fissured, sap red, not copious. Twigs stout, minutely rusty-

pubescent at the tips, later glabrous and coarsely striate. Leaves

stoutly coriaceous, glabrous, oblong-lanceolate or lanceolate, shin-

ing and medium-brown above when dry, glaucous beneath, apex

acute, base rounded or slightly cordate; nerves 17—25 pairs, obli-

que, fine above, prominent beneath, interarching rather indistinctly

at the margins; reticulations forming a close network, distinct on

both surfaces; length 16-28 cm.; breadth 4—8-5 cm.; petiole 1:5

cm. long, stout. Male flowers rather immature, as yet only 3 mm.

long on a very stout, 3 mm. long pedicel with a median bracteole;

perianth coriaceous, triquetrous in bud, rusty-tomentulose; staminal

disc concave (probably flat later), striate on the undersurface;

anthers 12-15, erect, sessile (most probably become stalked later).

Fruit globose, minutely rusty-tomentulose, 3 cm. in diam.; sub-

sessile on a 3 mm. long stalk; pericarp 3—4 mm. thick.

PAHANG: Forest plantation, Cameron Highlands, D.F.O. of B.

Padang 27196 (KEP); Rhododendron Hill, Cameron Highlands, Hen-

derson S.F.N. 23328 (BM, K, KEP, SING) and 10th April, 1930 (A,

K, SING, UC); Fraser’s Hill, Nur S.F. Nos. 10532 (K, SING) and

11313 (KEP, K); Kalong F.D. 22434 (KEP, SING); Strugnell &

284

Vol. XVI. (1958).

SAS A

Ss SS SSS

ane > a

A, Leafy twig. B, Male flowers. C, Male flower. D, Male flower in

Fig. 5. Knema rigidifolia J. Sinclair.

E from

longitudinal section. E, Fruit. A-D from Purseglove 4218.

Corner S.F.N. 33226.

285

Gardens Bulletin, S.

Mahmood F.D. 22519 (KEP, SING); Carrier 27317 (KEP); Hender-

son F.M.S. Mus. Herb. 11261 (KEP); Pine Tree Hill path, Fraser’s

Hill, Corner 33226 (E, K, L, SING holotype); Purseglove san (BM,

O, E, K, L, SING).

DISTRIBUTION: No other localities known.

This is a mountain species very close to K. conferta and scorte-

chinii. It differs from both in the sessile or almost sessile, globose

fruit. The leaves are more coriaceous with more distinct reticula-

tions than in scortechinii. They are often a degree more coriaceous

than those of conferta. They are glabrous in the adult stage while in

conferta some tomentum usually persists. In scortechinii the tomen-

tum does not usually persist but if it does it is much less than in

conferta.

(6) K. conferta (King) Warb., Monog. Myrist. (1897) 578 T. 24

Figs 1-2; Gamble, Mat. F.M.P. 5, 23 (1912) 243; Ridley,

F.M.P. 3 (1924) 70.

Basionym: Myristica conferta King in Ann. Roy. Bot. Gard.

Calc. 3 (1891) 315 Pl. 150.—Fig. 6.

Tree 12-24 m. high. Bark reddish-brown, dark red inside, shal-

low, slightly rough but not fissured; sap dark red, not very copious;

wood white. Twigs when young deeply striate with harsh, rusty-

stellate tomentum, later glabrous. Leaves coriaceous, oblong or

oblong-lanceolate, shining and glabrous above except the midrib

which becomes glabrous later, undersurface whitish-green, covered

with scurfy, stellate tomentum in varying degrees, apex acute, base

rounded, sometimes slightly cordate, less often acute, margins

slightly recurved; nerves 14—28 pairs, oblique, nearly parallel, in-

terarching near the margins; reticulations scalariform between the

primary nerves, also interwoven into a second set, forming a very

close network, very distinct on both surfaces when dry but in the

fresh condition those on the upper surface very faint; length 13-27

cm.; breadth 4-5—6:5 cm.; petiole 1-1-5 cm. long. Male flowers in

densely crowded, axillary clusters on woody tubercles, light-brown,

stellate-pubescent outside, glabrous and yellow inside, with a slight

greenish tinge and a brownish-pink spot at the base of each lobe,

coriaceous, 4-5 mm. long and 44-5 mm. broad, sub-globose, dep-

ressed and bluntly 3-angled in bud, the lobes acute at the apex

when the flower opens; pedicels 7 mm.—1 cm. long with a minute,

obtuse, median bracteole; staminal column 2 mm. high, pink,

shortly stalked, the disc rather large, filling the flower, at first

slightly convex, later flat; anthers 13-18, white in bud, orange-

yellow when ripe, on short filaments which are slightly recurved at

286

Vol. XVI. (1958).

WAALS

> \/X) 7s 3 |

ROS ose

= KSA LAB

OFT

Fig. 6. Knema conferta (King) Warb.

A, Twig with leaves and fruit. B, Fruit in longitudinal section. C, Arif

and seed. D, Group of male flowers. E, Male flower. F, Male flower

from above. G, Staminal column. H, Ovary. A, B, C, H from Sin-

clair S.F.N. 39505. D, E, F, G from Sinclair S.F.N. 40046.

287

Gardens Bulletin, S.

first and later flat. Female flowers larger than the male but fewer

in the clusters, obovoid or urceolate; pedicels about 5 mm. long;

ovary rufous-tomentose, style sessile with a 3-lobed stigma. Fruit

in clusters, ovoid, rather flat at the apex, at first shortly and spar-

ingly covered with a rusty-stellate tomentum, later almost glabres-

cent, line of suture distinct, length 3:5—4 cm., breadth 2—2-5 cm.;

pericarp 6—7 mm. thick; stalk 1 cm. long. Seed covered by the aril

which is laciniate at the apex.

PERAK: Ulu Bubong, King 10295 (CAL, K, SING) type material;

Larut, King 6211 (CAL, FI, G, K, L) type material; Taiping, Wray

2377 (CAL, K, SING) type material.

MaLacca: Griffith 4345 (A, CAL, K) type material; Maingay 1288

(L).

JOHORE: Sungei Rhu Reba, Jason Bay, Corner S.F.N. 28505 (A,

CAL, K, KEP, SING); Mawai, Ngadiman S.F.N. 34714 (SING);

Sungei Kayu, Kiah S.F.N. 32034 (A, BM, K, KEP, SING).

SINGAPORE: Maingay 1297 (CAL, K) type material; Bukit Timah,

Ridley 442 (SING) type material; Bukit Timah, catchment area, Liew

S.F.N. 37256 (SING); Chan Chu Kang, Ridley, Feb. 1894 (SING);

Mandai Road, Corner S.F. Nos. 33145 (A, DD, KEP, L, SING) and

37129 (A, BM, BRI, K, KEP, SING); Liew S.F.N. 37254 (SING);

Sinclair S.F. Nos. 39505 (B, BK, BO, DD, Delhi Univ., E, K, M, P,

PNH, SAN, SING); 40026 (BO,E, K, L, SING) and 40046 (BK, BM,

BO, DD, E, K, L, P, PNH, SAN, SING); Bukit Mandai, Ridley 6735

(SING); Corporal 442 (CAL); Changi, Ridley, date 1892 (SING).

DISTRIBUTION: Indo-China (as var. tonkinensis), Siam, Banka, Bil-

liton, Riouw (Pulau Karimum), Sarawak and Borneo.

This species has been confused with laurina, see notes under

that species. Maingay 1294 (CAL, K) quoted by King as type

material of conferta is laurina. K. conferta is closest to rigidifolia

and scortechinii (see notes under these species). It also is like K.

pectinata from Sarawak which probably should be placed in the

conferta group and has many more veins, larger flowers and a

larger staminal disc than conferta. K. lenta Warb. [Pierre 26, holo-

type, Cochinchina (P)] is extremely like conferta but the leaves

are glabrous beneath. I think it should rank as a variety of glau-

cescens.

There is one sheet in Herb. Kew, Maingay 1289, also numbered

1004. The digits 1289 are scored out with pencil and 1288 is

added. This is clearly K. intermedia but the Leiden duplicate is

conferta.

(7) K. scortechinii (King) J. Sinclair, comb. nov.

Basionym: Myristica scortechinii King in Ann. Roy. Bot.

Gard. Cale: 3.01892), 317-71. 43).

Synonyms: Knema conferta var. scortechinii Warb., Monog.

Myrist. (1897) 580; Gamble, Mat. F.M.P. 5, 23 (1912) 244;

288

Vol. XVI. (1958).

Ridley, F.M.P. 3 (1924) 70. Myristica laurina BI. var. borneen-

sis Mig. in Ann. Mus. Bot. Lugd.-Bat. 2 (1865) 51.—Fig. 7.

Tree 6—20 m. high. Bark dark brown with a dark reddish tinge,

nearly smooth, probably flaking later; sap dark red, fairly copious.

Twigs somewhat stout, rusty-pubescent at the apex but not striate,

glabrous and striate in the older parts. Leaves coriaceous, narrowly

oblong-lanceolate or lanceolate, base rounded and sub-cuneate or

slightly sub-cordate, apex acute, dark green, glabrous and shining

above, glaucous beneath, at first covered with minute, rusty, scaly,

deciduous tomentum, later glabrous; nerves 18—28 pairs, fine

above, prominent below, interarching at the margins; reticulations

forming a dense network, distinct above and slightly less promi-

nent below; length 16—32 cm., average 20 cm.; breadth 4-5—6 cm.;

petiole about 1 cm. long. Male flowers few, 4-10 in umbels from

woody tubercles; pedicels slender, 1—1:2 cm. long with a minute

bracteole at the middle or slightly above; perianth 5 mm. in diam.,

tawny-tomentulose outside, glabrous and cream-coloured inside,

cleft nearly to the base; androecium flat with 10—14 well-spaced,

stalked anthers. Female flowers unknown except for the remains

of the ovaries where the stigma is sessile and 3-lobed. Fruit ovoid-

globose, blunt, minutely rusty-tomentose when young, later sub-

glabrescent, 2-5—3 cm. long and 2—2-3 cm. broad; stalk over 1 cm.

long. Aril thin, red, lobed at the apex. Seeds ovoid-globose.

KepAH: Koh Mai F.R., Kiah S.F.N. 35176 (SING); Jerai F.R.,

Kuala Muda, Rejab Gurun K.F.N. 73503 (KEP).

PERAK: Scortechini 178a (CAL, K, L, SING) type material; Tapa,

Wray 1422 (CAL, SING) type material; Gopeng, King Nos. 5939

(CAL, DD, E, K, MEL, SING, UPS) type material and 6043 (BM,

CAL, FI, G) type material; Chanderiang, King 5617 (BM, CAL, E,

FI, G, K, L) type material; Larut, King 6694 (CAL, DD, K, L,

UPS) type material; Ulu Bubong, King 10635 (CAL, FI, K) type

material; Gunong Batu Pueth, Wray 285 (CAL) type material;

B.P.D., King 7926 (CAL, FI, G) type material.

TRENGGANU: 193 mile Kuala Trengganu-Besut Road (east side),

Sinclair & Kiah S.F.N. 40399 (SING).

PAHANG: Kuala Lipis, Somerville F.D. 10456 (K).

SELANGOR: Bangi F.R., Kuala Lumpur, Wyatt-Smith K.F.N. 63706

(KEP) sterile.

JouorE: Pulau Tinggi, Sinclair S.F.N. 40287 (K, SING).

DISTRIBUTION: Confined to Malaya.

King does not specify any type numbers but gives Scortechini,

Wray and King’s collector, Perak as the type material. He also

adds Derry, Malacca. I have not seen any sheets of Derry at Kew

or elsewhere.

Warburg reduced King’s species M. scortechinii to a variety of

conferta but Gamble has made a remark, page 245 to the effect that

289

Gardens Bulletin, S.

Fig. 7. Knema scortechinii (King) J. Sinclair.

A, Leafy twig with male flowers. B, Male flower. C, Staminal column.

D, Staminal column top view. E, Female flower. F, Ovary. G, Fruit.

A-D from Sinclair S.F.N. 40287. E-F from King’s Plate 153. G

from Wray 1422.

290

Vol. XVI. (1958).

he feels some doubt and that it is possible that further investigation

on the spot will restore it to specific rank. I too, feel that Gamble

and King are right and I have restored it. The differences are rather

slight but they are numerous. In scortechinii the leaves are longer

and narrower with nearly parallel margins (although conferta

sometimes has the margins parallel), less pubescent beneath and

the pubescence not persisting and the reticulations a shade finer.

The young, apical portions of the twigs are less tomentose and

never striate. The flowers are much fewer in their clusters, less

densely tomentose, slightly larger and the pedicels a trifle longer.

The anthers are 10-14 in number as against 13-18. The fruit is

ovoid-globose, shorter, and its stalk longer.

(8) K. kunstleri (King) Warb., Monog. Myrist. (1897) 568 T. 25;

Gamble, Mat. F.M.P. 5, 23 (1912) 241; Ridley, F.M.P. 3

(1924) 69.

Basionym: Myristica kunstleri King in Ann. Roy. Bot. Gard.

Calc. 3 (1891) 314 Pl. 149.

Synonyms: K. coriacea Warb., Monog. Myrist. (1897) 614.

M. coriacea (Warb.) Boerl., Handl. Fl. Ned. Ind. 3, 1 (1900)

92 nom. alt. K. parvifolia Merr., in Phil. Journ. Sc. 13 (1918)

287.—Fig. 8.

Tree 9-15 m. high, occasionally up to 30 m. with spreading

branches. Bark rough, but not fissured. Twigs rusty-pubescent at

the apex, lower down glabrous, striate and dark brown. Leaves

coriaceous, glabrous, medium-green above, glaucescent beneath,

elliptic or elliptic-oblong, apex bluntly acute or obtuse, base acute,

sometimes unequal-sided, slightly decurrent on to the petiole; mid-

rib and the 7-14 pairs of nerves prominent on both surfaces, the

nerves curving horizontally and gradually ascending, interarching

at the margins; reticulations very close, numerous and prominent

on both surfaces; length 9-19 cm.; breadth 4—8 cm.; petiole 1-5

cm. long, glabrous. Flowers rusty-pubescent outside, red and glabr-

ous inside, arising on woody, 3-5 mm. long, axillary tubercles

which are sometimes bi- or trifurcate. Male, pedicels 7-10 mm.

long with a minute, median bracteole; perianth 3 mm. long, the

3—(4) lobes coriaceous, broadly triangular-ovate, split nearly to

the base and thickened at the apex; staminal column 2 mm. long

with a flat disc and 9-12 stalked, horizontal anthers. Female

flowers 4 mm. long with stouter pedicels than in the male; ovary

2 mm. long, rusty-tomentose, stigma sessile with 3—4 lobes. Fruit

1-3 in a group, 2—2:5 cm. long and 1-5 cm. broad, ellipsoid, the

291

Gardens Bulletin, S.

Fig. 8. Knema kunstleri (King) Warb.

A, Leafy twig with fruit. B, Male flowers. C, Male flower. D, Staminal

column from above. A from Kiah & Strugnell S.F.N. 24034. B, C,

D from Wray 3985. )

292

Vol XVI. (1958).

remains of the stigma at its apex, puberulous when young, after-

wards glabrescent. Aril red, very slightly laciniate at the apex,

covering the seed.

2 ei Gua Ninik, Henderson S.F.N; 19684 (A, BRI, SING,

PERAK: Scortechini s.n. (CAL, E, K, SING); Scortechini 175a

(CAL, DD, FI, G, K, L); Larut, King Nos. 2614 (CAL); 2743 (CAL,

meee MEL). 3974 (CAL. DD. FL 'G, K;L, MEL): 9576 -(K):

338938 (BM, CAL, DD, FI, G); 3510 (CAL, L); 4150 (CAL, DD, E,

K, SING, UPS); 2216 (BM, CAL, FI, G, K, L); 4949 (CAL, FT);

5867 (CAL, DD, FI, K); 6440 (CAL, DD, FI, G, K, L, MEL, SING);

Gopeng, King Nos. 4414 (CAL, DD, E, UPS) and 4605 (CAL, DD,

K, MEL); Kinta, King 7180 (CAL, DD, FI, P); D.F.O. of Kinta

K.F.N. 54801 (KEP); Ulu Bubong, King Nos. 10022 (CAL, DD, FI,

K); 10549 (CAL, E, UC) and 10826 (CAL, FI, G, K, L); Blanda

Mabok, Wray 3985 (CAL, FI, G, L, P, SING); Waterfall Hill Wray

2056 (CAL, SING); Reservoir, Taiping, Haniff S.F.N. 13127 (K,

SING); Maxwell’s Hill, Curtis 2051 (CAL, SING); Burn-Murdoch

161 (SING, K) and Wray 2056 (SING); Gunong Keledang, Ridley

9587 (CAL, K, SING).

PAHANG: Ulu Sungei Kuantan, Symington and Kiah S.F.N. 28775

(K, SING); Bukit Chersa, Telom Valley, Kiah & Strugnell S.F.N.

24034 (SING); Raub, Jinal F.D. 20456 (KEP, SING).

SELANGOR: 20th mile Ginting Simpah, Strugnell F.D. 12104 (KEP,

SING); Weld Hill, Hamid C.F. 981 (SING).

MALAccA: Brisu, Derry C.F. 981 (SING).

DISTRIBUTION: Sarawak and Philippines.

TYPE MATERIAL: All Scortechini, King and Wray’s numbers from

Perak.

This species is closest to K. malayana. The leaves are more coria-

ceous and the reticulations are more distinct and raised on both

surfaces than in malayana. The male flowers are very similar in size

and number of anthers but the pedicels are slightly shorter, 7 mm.—

1 cm. long as against 1-1-5 cm. long in malayana. The fruit is

slightly larger, puberulous to glabrescent and ellipsoid, not pear-

shaped and tomentulose as in malayana.

I cannot see that it is different from K. coriacea Warb., Beccari

670 type (FI, K, P), and have reduced the latter to a synonym of

kunstleri. It agrees with Merrill’s K. parvifolia from the Philippines:

which I also reduce as a synonym. Closely allied to K. kunstleri is.

K. alvarezii Merr. from the Philippines but the leaves of the latter

are more acuminate with more veins and the male flowers densely

tomentose with shorter stalks.

(9) K. malayana Warb., Monog. Myrist. (1897) 570 T. 25 Figs:

1-2; Gamble, Mat. F.M.P. 5, 23 (1912) 242; Ridley, F.M.P. 3

(1924) 69.

Synonyms: Myristica glaucescens Hk. f. et Th., Fl. Ind. 1

(1855) 157 pro parte, altera pars prob. M. globularia Lamk. et

293

Gardens Bulletin, S.

alia taxa, (non K. glaucescens Jack); Hk. f., Fl. Br. Ind. 5

(1886) 111 pro parte; King in Ann. Roy. Bot. Gard. Calc. 3

(1891) 323 Pl. 161. M. corticosa (Lour.) Hk. f. et. Th., Fl. Ind.

1 (1855) 158 et A. DC. Prodr. 14, 1 (1856) 205 pro parte non

typica; Miq., Fl. Ind. Bat. 1, 2 (1858) 69 pro parte. M. mala-

yana (Warb.) Boerl., Handl. Fl. Ned. Ind. 3, 1 (1900) 91 nom.

alt.—Fig. 9.

Tree 5-20 m. high. Bark dark greyish-brown, not furrowed,

nearly smooth; inner pinkish to brownish; sap red, watery, not very

copious. Twigs slender, rusty-pubescent at the tips, lower down

glabrous, reddish-brown, finely striate, not angled. Leaves slightly

coriaceous when fresh but membranous when dry, glabrous, ob-

long-lanceolate to narrowly elliptic, apex sub-acute, base cuneate

or slightly rounded, upper surface dark green, dull to slightly shin-

ing, lower sub-glaucous; nerves and midrib raised on both surfaces,

drying reddish-brown; nerves 14—20 pairs, slender, oblique; reti-

culations forming a very fine, close network, distinct on both sur-

faces when dry but scarcely visible when fresh; length 7-17 cm.;

breadth 3—7 cm.; petiole slender, 7 mm.—1 cm. long. Male flowers

pale brown-tomentulose or puberulous outside, cream-coloured in-

side, arising on short, axillary, wood tubercles which are occasion-

ally bi-furcate; pedicels 1-1-5 cm. long, slender, with a minute brac-

teole at the middle or slightly above; perianth in bud depressed-

globose, 2-3 mm. in diam. (open flower up to 5 mm.), the lobes

broadly ovate, acute, glabrous with 6—7 striations inside; staminal

disc flat, stalk very short, less than 1 mm.; anthers 9-13, horizontal,

well-spaced and not touching, shortly stalked. Female flowers the

same colour as the male, sweet-scented, obovoid in bud, 5 mm.

long, coriaceous, on thicker, 5 mm. long pedicels with a median

bracteole; ovary ovoid-globose, dark chocolate-tomentose; style

very short, stigma green, 2-lobed or 4-lobed. Fruit usually in pairs,

pear-shaped, obtuse at the apex, often with the remains of the

stigma, narrowed to the base where the perianth scars form a

small collar, rusty-tomentulose, 1-5-2 cm. long and 1-1-2 cm.

broad; stalk slender, 5 mm. long; pericarp thin. Seed smooth, with

an outer, light brown succulent coat, dark blackish brown beneath.

Aril red, laciniate at the apex, covering the seed.

KepAH: Koh Mai F.R., Kiah S.F.N. 35156 (A, DD, K, KEP,

SING); near Sungei Lugong, Kiah S.F.N. 35097 (A, K, KEP, SING);

Rimbateloi F.R., Abdullah & Ariffin F.D. 12209 (KEP); Lanai, Meh

F.D. 42268 (KEP); Sungei Terenas, Sow F.D. 46158 (KEP).

PerAK: Gopeng, Kinta, King Nos. 4352 (CAL, DD, K, SING,

UPS) and 6128 (CAL, DD, E, FI, K); Chanderiang, King Nos.

5706 (CAL, E, FI, K, SING) and 5726 (CAL) DDE ea

UPS); Larut, King 7599 (CAL, FI, MEL); Ulu Bubong, King 10594

294

Vol. XVI. (1958).

Fig. 9. Knema malayana Warb.

A, Twig with female flowers. B, Female flower. C, Female flower from

above. D, Ovary. E & F, Stigma. G, Young fruit. H, Mature fruit.

I, Male flowers. J, Male flower. K, Staminal column. A-G from

Sinclair S.F.N. 40714. H from King 4352. I-K from Sow & Tachun

F.D. 16852.

295

Gardens Bulletin, S.

(CAL, G, L, MEL); Tapah, Batang Padang, Wray 176 (BM, CAL,

K, SING); Pulau Jarak, Seimund Nos. 1186 (SING) and 67 (SING);

Pondok Tanjong, Burn-Murdoch, date 1909 (K) and No. 179 (KEP,

SING).

TRENGGANU: 36th mile Jerangau Road, Dungun, Kiah & Sinclair

S.F.N. 40493 (BM, BO, E, K, L, SING).

PAHANG: Sungei Nering, Temerloh, Henderson F.M.S. Mus. Herb.

10714 (SING).

SELANGOR: Near Ulu Selangor King 8541 (BM, CAL, FI, G, K,

L, UC); Kuala Lumpur, Curtis, Feb. 1890 (SING); Sungei Buloh

Res., Ja’amat F.D. 15330 (KEP, SING) and F.D. 14938 (KEP); Sow

& Tachun F.D. 16852 (KEP, SING); Weld Hill, C.F. 825. (KEP,

SING).

Matacca: Maingay 1299 (K, L) and 1280 (CAL); Griffith 4343

(A, CAL, P); Griffith, date 1845 (K); Bukit Klana, Alvins 896 (SING).

JoHORE: S. Kayu, Mawai-Jemaluang Road, Corner, 14th April, 1935

(SING); Mawai, Corner S.F.N. 28440 (KEP, SING); S. Rhu Reba,

Jason Bay, Corner S.F.N. 28504 (A, B, CAL, K, KEP, L, SING);

Kampong Chin-Chin, Ulu Batu Pahat, Lake & Kelsall, 6th Nov., 1892

(SING).

SINGAPORE: Cantley 2919 (SING); Cantley 2851 (SING) may be

Malacca; Bukit Timah, Cantley 20 (K); Gardens’ Jungle, Ridley 4136

(CAL, K, SING); Corner S.F.N. 32209 (A, BM, K, KEP, SING);

Taban Path, Bukit Timah, Sinclair, 11th June, 1953 (E); Toas, Ridley

6447 (CAL, SING); Chan Chu Kang, Ridley 1833 (K, SING); road

leading to No. 1 rifle range, Nee Soon, Seletar Forest, Sinclair S.F.

Nos. 40280 (A, B, BM, BO, E, K, L, P, PNH, SAN, SING) and

40717 (BM, BKF, BO, E, K, KEP, L, SAN, SING); South side of

Mac Ritchie Reservoir off Lornie Road, Sinclair S.F.N. 40714 (BM,

BO, E, K; KEP, Lb, PP SING).

DISTRIBUTION: Siam, Borneo, Hubert Winkler 2466 and 2467

(BRSL, SING); Sumatra, Palembang, Boschproefstation J. 683 (L);

Rahmat Si Boea 7023 (A); Sarawak, Purseglove 5041 (A, K, L, NY,

SAR, SING).

TYPE MATERIAL: Wray 176; King Nos. 5706; 6128; 8541; 10594;

Cantley 20; Griffith 4343 (not Tenasserim as stated by Warburg but

Malacca); Griffith 4349 (CAL, K, P) from Tenasserim is K. globularia

and not malayana as quoted by Warburg.

Warburg had to give this species a new name, K. malayana, on

account of K. glaucescens Jack being already in use for a different

plant. Warburg did not indicate a holotype (see his list of type

material above). M. glaucescens Hk. f. et Th. sensu King is cer-

tainly malayana and the specimens he quotes are correctly identi-

fied with malayana except that Maingay 1280 is partly malayana

and partly glaucescens var. patentinervia. However the glaucescens

of Hk. f. et Th. is a mixtum compositum, consisting of more than

two species. The epithet glaucescens Hk. f. et Th. is antedated by

Jack’s K. glaucescens and cannot be used. I do not think that K.

malayana occurs in Burma, Griffith 4349 (CAL, K, P) quoted

by Warburg as malayana is globularia. Griffith 4343, K. malayana,

is from Malacca and not from Tenasserim as stated by Warburg.

296

Vol. XVI. (1958).

The long pedicels and the appearance of the leaves recall K.

kunstleri. The leaves, however, are not coriaceous but membran-

ous when dry and the reticulations are not quite so prominent.

Sterile and even fertile specimens have been confused with globu-

laria but the twigs of the latter are quite different. The twigs of

malayana are finely striate, more slender and never angled. The

lower midrib is not striate when dry as in globularia and the curv-

ing of the nerves of the leaf is different. The pear-shaped fruit will

also distinguish it from the nearly spherical fruit of globularia. The

aril of malayana when eaten is first hot in the mouth, then cold,

tasting like a mixture of cloves and nutmeg.

(10) K. communis J. Sinclair, sp. nov.—Fig. 10. Plate TXB.

K. malayanae et K. kunstleri proxima a quibus pedicellis masculis:

brevioribus, foliis longioribus angustioribus oblongis vel oblongo-

lanceolatis (non ellipticis) differt. Alterae species huic gregi perti-

nentes: K. kunstleri, K. malayana, K. glaucescens.

Arbor 4-16 m. alta. Cortex extus rubro-brunneus tenuis fere

laevis vel scabriusculus, lacunis elongatis vel rotundatis 6-10 cm.

longis irregulariter foveatus, intus ruber; latex ruber copiosus.

Ramuli apice lepido-tomentelli, teretes, infra glabri, rubro-brun-

nei, tenuiter striati. Folia coriacea oblonga vel oblongo-lanceolata

vel lanceolata; apice acuta, basi acuta vel leviter rotundata inter-

dum inaequilateralia, 6—25 cm. longa, 2:5—7 cm. lata, supra glabra

nitida atro-viridia (in sicco viridiuscula), subtus glauca, primo fer-

rugineo-stellato-lepidota deinde glabrescentia; nervi 17—20 pares.

obliqui interdum irregulariter curvati, supra graciles leviter elevati,

subtus prominentes; reticulationes in vivo vix visibililes, in sicco

utrinque distinctae densae tenues; petioli 7 mm.—2 cm. longi, primo

lepidoto-puberuli deinde glabri. Flores masculi ex tuberculis lig-

nosis orti, pedicelli 4-5 mm. longi ferrugineo-tomentelli, medic

bracteolati; perianthium extus ferrugineo-tomentelli, intus glabrum

roseum 3—4 mm. longum, fere ad basi usque in lobos tres ovatos

acutos fissum; discus staminalis albus planus 1-5 mm. latus, stipite:

1 mm. alto et antheris 9 horizontalibus distantibus praeditus.

Flores feminei immaturi 3 mm. longi; pedicelli 1:5 mm. longi;

Ovarium 2 mm. longum pubescens; stigma 4-lobatum. Fructus

obovoideus, basin versus angustatus, ferrugineo-tomentellus, 1:8

cm. longus, 1—1-:3 cm. latus cum linea suturali visibili; stipes 4—5

mm. longus. Arillus ruber fimbriatus. Semen nitidum, griseum,

pallido-brunneo variegatum.

Tree 4-16 m. high. Bark reddish brown outside, of no great

thickness, slightly rough to nearly smooth with elongated or

297

‘ Gardens Bulletin, S.

TurMn Der

4 cm

Fig. 10. Knema communis J. Sinclair.

A, Leafy twig with male flowers. B, Male flower from above. C, Fruit.

A-B from Sinclair S.F.N. 39570. C from Sinclair S.F.N. 39474.

298

Vol. XVI. (1958).

rounded depressions 6—10 cm. long here and there where flakes

have dropped out, inner bark red; sap red, abundant. Twigs scurfily

rusty-tomentulose towards the apex, including the elongated leaf

bud, terete, further back glabrous, reddish-brown and finely

striate. Leaves coriaceous, dark green, glabrous and shining above,

drying greenish, glaucous beneath with a covering of rusty-stellate

scales when young, becoming glabrescent, oblong to oblong lanceo-

late or lanceolate, apex acute, base acute or slightly rounded,

sometimes unequal-sided; midrib raised on both surfaces; nerves

17—20 pairs, slender and slightly raised above, prominent below,

oblique or curving rather irregularly at times, line of interarching

often broken; reticulations forming a very close, fine network on

both surfaces, distinct when dry, faint or not visible when fresh;

length 6—25 cm.; breadth 2:5—7 cm.; petiole 7 mm.—2 cm. long, at

first rusty-scaly-puberulous, later glabrous. Male flowers sweet-

scented only when crushed, on axillary, woody, persistent tubercles;

pedicels 4-5 mm. long with a minute, obtuse, median bracteole;

perianth globose in bud, rusty-stellate-tomentulose outside, glabr-

ous and pink inside, fleshy, 3-4 mm. long, split nearly to the base

into 3 lobes; staminal column 1 mm. high with a white, 1-5 mm.

broad disc; anthers 7—9, horizontal when mature, not touching

each other, shortly stalked. Female flowers (immature) 3 mm.

long on a pedicel 1-5 mm. long; ovary 2 mm. long, pubescent;

stigma a 4-lobed disc. Fruit obovoid narrowed towards the base,

rusty-tomentulose, the circumferential line of suture visible; length

1-8 cm. and breadth 1-1-3 cm.; stalk 1-1-5 cm. long. Aril red,

fimbriate in the upper third. Seed shining, greyish, mottled with

pale brown. |

PERAK: Tapah, Burn-Murdoch 380 (BM, SING); Trolak F.R.,

Jaamat F.D. 43436 (KEP, SING); Ulu Bubong, King 10286 (CAL,

DD).

TRENGGANU: Compt 3B, Gunong Tebu F.R., Sinclair & Kiah S.F.N.

40417 (E, K, L, SING).

SELANGOR: Sungei Buloh F.R., Watson F.D. 17017 (SING).

SINGAPORE: South side of Mac Ritchie Reservoir, Sinclair S.F.N.

38913 (E, K, SING); Boundary Path above Fern Valley, Bukit Timah,

Sinclair S.F.N. 39474 (E, K, SING); Rock Path, Bukit Timah, Sinclair

S.F. Nos. 39570 (BKF, BM, BO, DD, Delhi Univ., E, K, L, P, PNH,

SAN, SING) and 40320 (BM, BO, DD, E, K, L, P, SING); Gardens’

Jungle, Sinclair 40522 (BKF, BM, BO, DD, Delhi Univ., E, K, KEP,

L, M, P, PNH, SAN, SING holotype).

DISTRIBUTION: Confined to Malaya.

_ The fruit is exactly as in K. malayana but the leaves are more

coriaceous, larger, not elliptic, and covered with rusty-stellate:

scales when young. The pedicels of the male flowers are shorter.

299

Gardens —_— ihe

(11) K. ena (Warb.) J. Sinclair, comb. nov.

Basionym: Gymunacranthera stenophylla Warb., Monog. My-

rist. (1897) 364 T. 20 Figs 1-2.

Synonym: Myristica stenophylla (Warb.) Boerl., Handl. FI.

Ned. Ind. 3, 1 (1900) 88 nom. alt.—Fig. 11.

Tree 8-10 m. high. Bark reddish brown, smooth except for a

few flakes here and there; sap red, copious. Twigs slender, much

branched, rusty-pubescent at the tips, soon glabrous, dark brown to

blackish, finally striate; terminal bud thin, slender. Leaves thinly

coriaceous, dark green, shining above, drying coffee-brown, glauc-

ous beneath, glabrous on both surfaces except the lower which is

puberulous when young, lanceolate to oblong-lanceolate, margins

somewhat undulate, apex acute, less often obtuse, base cuneate;

midrib raised on both surfaces; nerves 18-25 pairs, slender, faint

above, distinct below, indistinctly anastomosing at the margins;

reticulations faint below, scarcely or not at all visible above; length

7-17 cm.; breadth 1-5-4 cm.; petiole 1-1-3 cm. long, slender,

glabrous. Flowers on woody, axillary tubercles, fragrant, rusty-

pubescent outside, glabrous inside. Male: pedicels rusty-pubescent,

slender 4-5 mm. long with a minute median bracteole; perianth

2-2:5 mm. in diam., ovate, acute, triangular-globose in bud; disc

shortly stalked, reddish-brown, bearing 6—8 white, obtuse, slightly

erect anthers which do not touch each other; filaments very short.

Female: pedicels shorter and thicker than in the male; perianth

urceolate or oblong, 3 mm. long; ovary 1 mm. long, tomentose;

stigma sessile, very young in the material examined, seems to be

5—6 lobed, but is probably bifid with the main lobes further sub-

divided (more and older material should be again examined when

available). Fruit ridged, obovoid with rounded apex, minutely

pubescent, becoming glabrous, 1-3 cm. long and 1 cm. broad; the

stigmatic scar often persistent as an apiculus, stalk 5 mm.—1 cm.

long, broadest at the base of the fruit. Aril thin, covering the seed,

entire or slightly laciniate at the apex.

KEDAH: Enggong F.R., Sik, Kochummen K.F.N. 83809 (KEP).

KELANTAN: Bukit Temangon, Haniff & Nur S.F.N. 10256 (BRI, K,

KEP, SING, UC).

PENANG: Curtis 3679 (CAL, K, SING).

PERAK: Scortechini 1527 (BM, CAL, K); Kroh Reserve, Yeop F.D.

9352 (SING); Pulau Rumbia, Malacca Straits, Seimund F.M.S. Mus.

Herb., 10th March, 1926 (SING); Wyatt-Smith K.F.N. 76520 (KEP);

Parit Reserve, Kinta, D.F.O., K.F.N. 54802 (KEP).

TRENGGANU: Ulu Bendong, Kemaman, Corner S.F. Nos. 30088 (FI,

SING) and 20042 (FI, SING).

PAHANG: Rotan Tunggal F.R., Raub, Osman F.D. 29253 (KEP).

300

Vol. XVI. (1958).

Fig. 11. Knema stenophylla (Warb.) J. Sinclair.

A, Leafy twig with male flowers. B, Male flower. C, Male flower

showing staminal column. D, Fruit. A-C from Nur S.F.N. 10256.

D from Somerville F.D. 14588.

301

Gardens Bulletin, S.

SELANGOR: Ayer Hitam, Kajang, Somerville F.D. 14588 (KEP,

SING); Sungei Buloh F.R., Ja’amat F.D. 14913 (KEP); Klang, Keh-

ding 150 (FI Herb. Becc. 7714 holotype).

JoHORE: Gunong Pulai, King, Aug. 1879 (CAL, K); Nur & S.F.N.

7806 (K, SING); Sinclair S.F. Nos. 39518 (BO, E, K, L, P, SING)

and 39524 (E, K, L, SING).

DISTRIBUTION: Sarawak, Beccari 1709 (FI, K).

The most striking features about this species are the slender

twigs and the lanceolate leaves which, on the upper surface have

faint nerves and no visible reticulations. On account of the faint

nerves and the absence of reticulations it resembles a Gymnacran-

thera and was so described by Warburg. However, his figure clearly

shows the fruit to be that of a Knema and not a Gymnacranthera

since the aril is laciniate at the apex only. In Gymnacranthera it is

laciniate to the base. I am grateful to Dr. Pichi-Sermolli for the

loan of the type specimen, Kehding 150 which matches the Mala-

yan material quoted here.

Scortechini 1527 quoted by King as K. geminata is K. steno-

phylla and his plate 160, fig. 1 is most probably K. stenophylla.

Fig. 2 of the same plate is correct for K. glaucescens Jack, synonym

K. geminata. Two other specimens referred to by Gamble as K.

geminata are K. stenophylla. These are Curtis 3679, Penang and

King s.n., Gunong Pulai, Johore.

(12) K. glaucescens Jack in Mal. Miscellanies 7 (1821) 35; Warb.,

Monog. Myrist. (1897) 616 inter “species negligendas”.

Synonyms: Knema geminata (Miq.) Warb., Monog. Myrist.

(1897) 604 pro parte, (altera pars — K. stenophylla); Gamble,

Mat. F.M.P. 5, 23 (1912) 247 pro parte; Ridley, F.M.P. 3

(1924) 72 pro parte. Myristica. geminata Miq., Fl. Ind. Bat.

Suppl. (1861) 384; King in Ann. Roy. Bot. Gard. Calc. 3

(1891) 322 (excl. tab.) quoad specimina sumatrana tantum. K.

glauca (Bl.) Warb., Monog. Myrist. (1897) 594 T. 25 Figs

1-3 cum var. typica Warb. l|.c. 594 et var. sumatrana Warb. Lc.

594. M. glauca Bl., Cat. (1823) 111; Bijdr. 2 (1825) 576 et

Rumphia 1 (1835) 187 T. 60; King in Ann. Roy. Bot. Gard.

Calc. 3 (1891) 320 Pl. 157 pro parte. M. glauca var. sumatrana

Miq., Fl. Ind. Bat. Suppl. 1 (1861) 384. M. corticosa (Lour.)

Hk. f. et Th., Fl. Ind. (1855) 158 pro parte, non typica.

M. corticosa var. sumatrana Miq., Fl. Ind. Bat. Suppl. (1861)

384. M. corticosa var. lanceolata Miq., Fl. Ind. Bat. Suppl. 1

(1861) 384. M. glaucescens Hk. f. et Th., Fl. Br. Ind. 5 (1886)

111 pro parte, (una pars = K. malayana Warb.). K. palem-

banica (Mig.) Warb., Monog. Myrist. (1897) 592 T. 25. M.

palembanica Migq., Fl. Ind. Bat. Suppl. 1 (1861) 384. Knema

302

Vol. XVI. (1958).

wrayi (King) Warb., Monog. Myrist. (1897) 572 T. 24 Figs

1—2; Gamble, Mat. F.M.P. 5, 23 (1912) 243; Ridley, F.M.P.

3 (1924) 70. M. wrayi King in Ann. Roy. Bot. Gard. Calc. 3

tint) Sto Fl. 131,

Knema glaucescens is a polymorphic species with a wide distri-

bution and consequently has been given many names. Many of

the forms or varieties grade into each other and as far as I can see

the only distinct one is var. cordata which I think is not different

from K. pulchra. I would have considered K. glauca var. sumatrana

as a good variety, had I seen the type and nothing else, but other

Sumatran specimens show large and small leaves on the same

sheet and are not different otherwise. The same thing applies to

K. geminata which can also have large and small leaves on the

same plant. Here the fruit is more strongly ridged but I do not

think this character matters since the ridging in K. glaucescens

may be faint, strong or not at all. This applies to other species of

Myristicaceae as well. I, therefore, regard K. geminata Miq. as a

synonym of K. glaucescens. Malayan material quoted by King,

Warburg, Gamble and Ridley as K. geminata is not the same as

the Sumatran and is K. stenophylla. K. palembanica and K.

wrayi must also be considered as synonyms of K. glaucescens.

The most widely used name for K. glaucescens is K. glauca

(Bl.) Warb. but I discovered the type of K. glaucescens Jack

at Leiden, collected by Jack in Sumatra. The writing is on

blotting paper and this specimen was believed to have been burnt

along with others in the ship Fame on which Raffles embarked for

Europe in 1824 (see Warburg page 616 and Flora Malesiana vol.

1 page 257). Warburg relegated this plant to ‘Species Negligenda’

but Merrill, in Journ. Arnold Arboretum 33, 3 (1952) pointed out

that the name glaucescens ought to be used for glauca as the des-

cription fitted the plant. He, however, had not seen Jack’s type.

Certain numbers quoted by King under Myristica glauca in Ann.

Roy. Bot. Gard. Calc. 3 (1891) 320 are not entirely glaucescens

(glauca). Maingay 1280 consists of K. malayana and K. glauces-

cens var. patentinervia; Maingay 1280/2 is glaucescens var. paten-

tinervia, 1281 is intermedia and 1282 is globularia. Curtis, Penang,

is also not glaucescens. There are no records of this plant from

Penang. I have seen no specimens of glaucescens collected in

Malacca by Hervey or by Cantley in Singapore. I suspect Hervey’s

plant was globularia (see List of Collectors’ Numbers). Knema

glauca vars. andamanica and nicobarica and other sheets named

glauca from South Burma are K. glaucescens var. andamanica.

303

Gardens Bulletin, S.

var. glaucescens—Fig. 12 & Fig. 13, A.

Tree 6-20 m. high with a few stilt roots. Bark flaky, wood

white; sap red, not copious. Twigs glabrous, stout, often pale

straw-coloured, with reddish, blackish or greyish patches here and

there, coarsely striate. Leaves coriaceous, glabrous, dark green and

glossy above, whitish grey beneath, oblong er elliptic-oblong, apex

acute, base rounded, sometimes emarginate, less often bluntly

acute; midrib and nerves slightly raised above, prominent beneath;

nerves 13-20 pairs, slender, oblique, interarching at the margins;

reticulations fine but visible on both surfaces when dry, forming a

close network; length 10-24 cm.; breadth 4—9 cm.; petiole 1—-1:5

cm. long. Flowers on axillary tubercles which are sometimes bi- or

trifurcate. Male more numerous than the female in the clusters,

sub-globose in bud; pedicels 8 mm.—1 cm. long with a 1 mm. long,

obtuse bracteole above the middle or at the middle or occasionally

a little below the base of the flower (in this case flowers probably

immature and bracteole may appear lower down later); perianth

spreading, split nearly to the base, puberulous to sub-glabrous out-

side, glabrous and rose-red inside, 4 mm. long; staminal disc pur-

plish, sub-triangular, sub-convex, with a short, stout stalk; anthers

12-17, horizontal, shortly stalked, not touching each other. Female

(those observed young, not open) as in the male but urceolate on

stouter, shorter, 5 mm. long pedicels; ovary nearly glabrous, length,

including the 4—S-lobed stigma, 2 mm. Fruit pink to yellow, black-

ish-brown and slightly rough when dry, glabrous or tomentulose,

oblong or ovoid, 2:-5—4:5 cm. long and 1-2-5 cm. broad, obtuse

at both ends; pericarp 2 mm. thick; stalk 7 mm.—1 cm. long,

slender: Aril entire or fimbriate at the apex only, thin but tough.

KELANTAN: Sungei Ketil, Henderson S.F.N. 24817 (A, DD, K,

SING); Sungei Chalil, S. Lebir, Henderson S.F.N. 29531 (K, SING).

PeRAK: Larut, King Nos. 5299 (CAL, DD, K); 6514 (CAL, DD);

6521 (CAL, DD, K, SING, UPS) and 7475 (BM, CAL, Fi Gok

UC); Kuala Depang, King 8277 (CAL, E, K, L); Ulu Bubong, King

Nos. 10444 (BM, CAL, FI, G, K, L) and 10691 (CAL, MEL, SING);

Ulu Leding, Wray 2006 (CAL, K, MEL, SING); Kurau, 606 (SING);

Dengong, Telok Anson, Haniff S.F.N. 14310 (A, KEP, SING).

SELANGOR: Mahamud or Mat 3538 (K, SING); Ulu Selangor,

Goodenough 10519 (CAL, SING):

MALACCA: Griffith 4349 (A, K) sheet is labelled Tenasserim and

Andamans but I think in error; Batang Malacca, Derry 909 (SING).

JOHORE: Tempayan River, Ridley: 13264 (K, SING); Kuala Sem-

brang, Lake & Kelsall 4011 (SING); S. Sedili, Mawai, Corner S.F.

Nos. 28176 (SING) and 29275 (K, SING); Sungei Kayu, Kiah S.F.

Nos. 31951 (SING and 22384 (SING); Sungei Berassau, Mawai-Jema-

luang Road, Corner S.F.N. 28962 (SING).

SINGAPORE: Sungei Jurong, Ridley 3873 (K, SING).

304

ee i \-

Vol. XVI. (1958).

a tA AN

(if ATi ClO

ie NZ SANS f » °

\ NNR UR =< : xa PLS. * S

N W Ss; —ae SLN

Nt t

‘\ NS

5 cm

Fig. 12. Knema glaucescens Jack var. glaucescens.

A, Leafy twig with male flowers. B, Male flower. D, Ovary. E, Fruit.

A-D from Henderson S.F.N. 24817. E from Lake & Kelsall 4011.

305

Garileas Bulletin, Be

ius]

Seg

LLL L Lee

ha;

'\ Md

"ge

WS 4¢

& La

WZ {iF

A LL

@ CG

y CE

Fig. 13. Knema glaucescens Jack and its varieties.

A, var. glaucescens. B, forma rubens. C, var. patentinervia. D, var.

cordata.

306

Vol. XVI. (1958).

DISTRIBUTION: Siam, Banka, Sumatra,’ Borneo, Java, Sumbawa, Lom-

bok, Ceram and Celebes (the Celebes plant may be a variety).

Type MATERIAL: Knema glaucescens, Sumatra, Jack (L) holotype.

Myristica wrayi Perak, King and Wray’s numbers, syntypes. M. palem-

banica, Derma-enim, Palembang (U) holotype. M. glauca, Java, Blume

(L) type material. M. geminata Sumatra, Teijsmann (L, U) syntypes.

forma rubens J. Sinclair, f. nov.—Fig. 13, B.

-A typo foliis angustioribus, costa subtus nitida rubro-brunnea,

antheris paucioribus 8—11, pedicellis brevioribus distinguitur.

Arbor 5—16 m. alta. Cortex extus atro-fuscus, fere sub-laevis,

intus carneo-fuscus; latex roseus. Ramuli in partibus juvenilibus

pallido-brunneo puberuli, in senioribus striati, glabri, crassioribus.

Folia tenuiter-coriacea, glabra, supra nitida, atro-viridia, subtus

glauca, lanceolata, cum marginibus fere parallelis, apice acuta vel

obtusiuscula, basi cuneata, 9-20 cm., longa, 3—5 cm. lata; costa

utrinque elevata, subtus rubra nitida; nervi 17—20 pares utrinque

elevati, subtus prominentes, rubri, obliqui, abruptiuscule adscen-

dentes, indistincte anastomosantes; reticulationes supra vix dis-

tincte, subtus prominentes scalariformes; petioli 5 mm.—1 cm.

longi, pubescentes. Flores masculi extus ferrugineo-tomentosi, intus

carnei, glabri; pedicelli 3-4 mm. longi, medio bracteolati; perian-

thium in alabastro triangulare subglobosum, 2—3 mm. longum cum

lobis late-ovatis, sub-acutis; discus antheris 7-12 obtusis horizon-

talibus coronatus; stipes 1 mm. longus. Flores feminei ignoti.

Fructus oblongus utrinque obtusus, ferrugineo-tomentellis 2—2:2

cm. longus et 1-5—1-7 cm. latus cum stipite 3-5 mm. longo.

Tree 5—16 m. high. Bark dark brown, nearly smooth, inner layer

pinkish brown; sap pinkish. Twigs, the younger parts slender, light

brown, puberulous, the older glabrous, pale straw-coloured, stouter

and striate. Leaves lanceolate, the sides nearly parallel, the apex

acute or bluntly acute, the base cuneate; midrib and the 17—20

pairs of oblique nerves reddish brown and shining beneath; length

9-20 cm.; breadth 3-5 cm. Male flowers rusty-tomentose outside,

pink and glabrous inside; pedicel 3-4 mm. long with a median

bracteole; perianth triangular-globose in bud, 2—3 mm. long; an-

thers 7-12, well-spaced, stalked, horizontal, obtuse. Female flowers

not seen. Fruit oblong, obtuse at both ends, rusty-tomentulose,

2—2:2 cm. long and 1-5-1:7 cm. broad; stalk 3—5 mm. long.

PAHANG: 8 miles south of Kuala Lipis, Burkill & Haniff S.F.N. 17052

(A, SING); Sungei Lembing, Kuantan, Symington and Kiah S.F.N.

28755 (K, KEP, SING); Kemansul Reserve, Temerloh, Hamid F.D.

10583 (E, K, SING).

JoHORE: Kluang, Holttum S.F.N. 9425 (BRI, K, SING); S. Segun,

Gunong Panti, Mawai, Corner S.F.N. 29403 (A, K, SING); Rengam

F.R., 7 miles from Kluang, Cousens K.F. Nos. 65588 (KEP) and

69775 (KEP).

307

Gardens Bulletin, S.

SINGAPORE: Bukit Timah F.R., Ngadiman S.F.N. 35901 tree No. 197

(A, K, L, SING); Reservoir Woods, Ridley 4819 (DD, K, SING holo-

type); Mac Ritchie Reservoir off Lornie Road, Sinclair S.F.N. 40213

(BM, BO, E, K, KEP, L, P, SING).

DISTRIBUTION: Sumatra, Krukoff 4087 (A, BR, BRI, G, L, LE, NY,

SING). Bangka, Kostermans & Anta 922 (PNH).

I have not given a higher rank than forma to the above since

there are intermediates between it and K. glaucescens. It can be

best distinguished by the narrower lanceolate leaves, 3-5 cm.

broad with the lower midrib and nerves ‘reddish brown and often

shining. The student might not at first sight associate it with glau-

cescens if he has not seen the full range of forms of that species

and so it has been given a place in the key in order to avoid diffi-

culties. Among the intermediates between it and glaucescens are

Achmad Nos. 206 (L, SING, U) and 840 (L, U). In these the

leaves are broader as in typical glaucescens but they retain the

shining reddish brown nerves of forma rubens. I should however

name these two numbers K. glaucescens var. glaucescens.

var. patentinervia J. Sinclair, var. nov.—Fig. 13, C.

A typo fructibus minoribus, reticulationibus foliorum supra

valde obscuris, nervis magis horizontalibus differt. Etiam K. steno-

phyllae affinis, a qua foliis latioribus, antheris pluribus, floribus

fructibusque maioribus divergit.

Arbor 8-16 m. alta. Latex ruber. Ramuli apice minute ferru-

gineo-pubescentes, in partibus veteribus glabri, fusco-brunnei,

striati. Folia coriacea oblonga vel oblongo-lanceolata, supra in vivo

atro-viridia nitida, in sicco coffeo-brunnea, subtus glauca, basi

acuta vel rotundata, apice obtusa vel acuta, 7-23 cm. longa, 3—

7:5 cm. lata; costa utrinque elevata, supra in sulco laminae posita;

nervi 15—25 pares supra leviter prominuli vel interdum non cons-

picui, subtus prominentes, inter se valde approximati, a costa an-

gulo 70°—prope 90° abeuntes, marginem versus indistincte arcua-

to-conjuncti; reticulationes supra vix visibiles, subtus tenues;

petioli 7 mm.—1 cm. longi, primum pubescentes, deinde glabri.

Flores masculi coriacei, extus pallido-brunneo-tomentosi, intus

glabri, in alabastro triangulare sub-globosi, in lobis 3—5 mm.

longis ovatis sub-acutis terminati; pedicelli 5 mm.—1 cm. longi,

graciles, ad medium vel infra minute unibracteolati; discus 2 mm.

latus, stipitatus, antheris 10—14 stipitatis, inter se paulo dissitis,

coronatus. Flores feminei 5 mm. longi, in lobis longitudinaliter 4-

fissi; ovarium 2 mm. longum tomentosum verticaliter sulcatum;

stigma sessile, in duobus lobis breviter bifidis divisum. Fructus

308

Vol. XVI. (1958).

ellipsoideus ferrugineo-tomentellus, basin quam apicem versus

magis angustatus, 1-3—1:5 cm. longus, 1-1-3 cm. latus cum stipite

7 mm. longo suffultus.

Tree 8—16 m. high. Sap red. Twigs, the apices minutely rusty-

pubescent, the older parts glabrous, dark brown, striate. Leaves

coriaceous, oblong or oblong-lanceolate, dark green above, shin-

ing, coffee-brown when dry, glaucous beneath, base rounded or

acute, apex obtuse or acute; midrib raised on both surfaces, lying

in a groove on the upper; nerves 15-25 pairs, slightly raised or

faint above, prominent beneath, close together, nearly horizontal,

leaving the midrib at an angle 70°- to nearly 90°, interarching

in a faint line at the margins; reticulations scarcely or not visible

above, faint beneath; length 7—23 cm.; breadth 3—7-5 cm.; petiole

7 mm.—1 cm. long, pubescent, later glabrous. Male flowers light

brown-tomentose outside, glabrous inside, triangular to sub-glo-

bose in bud, the lobes 3—5 mm. long, ovate, sub-acute; pedicels 5

mm.—1 cm. long, slender with a minute bracteole at the middle or

below; disc 2 mm. broad, stalked, bearing 10-13 well-spaced,

stalked anthers. Female flowers elongate, 5 mm. long, cleft 3-way

down into 3 lobes; ovary tomentose, 2 mm. long, vertically

grooved; stigma sessile, bi-lobed, each lobe again minutely bifid.

Fruit ellipsoid, rusty-tomentulose, narrowed at the base and

slightly at the apex; 1-3-1-5 cm. long and 1-1-3 cm. broad on a

7 mm. long stalk.

' TRENGGANU: Bukit Kajang, Kemaman, Corner, 7th Nov., 1935

(SING) and Corner S.F.N. 30482 (SING); Ulu Bendong, Kemaman,

Corner, 1st Nov., 1935 (SING).

PAHANG: Karak Reserve, Mohamud F.D. 14960 (KEP, SING).

SELANGOR: Weld Hill, Hashim or Burn-Murdoch 10 (KEP, SING);

Hamid C.F. 561 (KEP, SING); Kuala Lumpur, Foxworthy C.F. 2363

(KEP, SING).

NEGRI SEMBILAN: Selaru, Ludin C.F. 1873 (K, KEP, SING holo-

type).

Maxacca: Maingay Nos. 1299 (L); Maingay 1280 (K) and

1280/2 (CAL, K); Ayer Panas, Derry 1038 (BM, CAL, K, SING).

JOHORE: Bukit Tinjau Laut, Ngadiman S.F.N. 36935 (BRI, K, KEP,

SING); Sungei Sedili below Mawai, Corner, 16th July, 1939 (SING);

Tanjong Kupang, Ridley 44 (SING); Bukit Soga, Ridley 11028

(SING); Gunong Panti, Corner, 24th Jan., 1937 (SING); Bukit Kuing,

Corner S.F.N. 28649 (SING).

SINGAPORE: Mandai Road, Corner, 7th Aug., 1940 (SING) and 24th

July, 1940 (SING).

DISTRIBUTION: Sumatra, Palembang, Grashoff 737 (L). Borneo,

Pulow Tekemeng, Main 2065 (A, L, SING).

At first, I thought this was a distinct species, but after seeing a

large amount of material of K. glaucescens, I am compelled to

309

Gardens Bulletin, S.

consider it only as a variety of that species. The leaves look dif-

ferent owing to their almost horizontal nerves and the lack of reti-

culations above but the flower structure is essentially the same.

There are two forms, one with the apices of the leaves obtuse and

the other with the apices acute. The leaves recall those of K..

stenophylla in that they dry the same colour above, a coffee-brown,

have indistinct or no reticulations above but are broader. The:

nerves are slightly more patent and much more patent than those

of K. glaucescens var. glaucescens as has been pointed out, often

leaving the midrib at an angle of 90°. The relationship with K.

stenophylla and with group 3b is more apparent in floral than in

leaf characters. This is seen in the stalked anthers, the bracteole

more or less median on the pedicel and the stigma sessile with few

lobes. The fruit is similar to that of K. malayana but is smaller

than in typical glaucescens.

var. cordata J. Sinclair, var. nov.—Fig. 13, D.

Synonyms: Knema pulchra (Miq.) Warb., Monog. Myrist.

(1897) 600. .

Myristica pulchra Mig. in Ann. Mus. Bot. Lugd.-Bat. 2

(1865) 51.

A typo foliis plerumque maioribus, basi cordatis, ramulis cras-

sioribus rubro-vel griseo-brunneis differt.

Arbor 6-13 m. alta, radicibus adventivis suffulta. Cortex brun-

nescens laevis cum paucis lenticellis elongatis, nec decorticans nec

pustulatus; latex rubescens exilis. Ramuli crassi, in innovationibus

laeves, rufo-tomentelli, in partibus vetustioribus glabri valde

striate. Folia coriacea, oblonga, supra atro-viridia, costa nervisque

albido-viridibus exceptis, subtus glauca sed in nervis et costa brun-

nea, glabra (folia iunvenalia caduco-tomentosa, fulva) apice acuta,

basi rotundata leviter cordata, ad petiolum aliquanto decurrentia;

33-50 cm. longa, 15-17 cm. lata; costa basi in pagina superiore

tantum plana, 4 mm. lata, in alteris partibus elevata; nervi 25

pares vel plures, utrinque prominentes basales inter se approximati,

alteri distantiores; reticulationes tenues dense dispositae, pro

maiore parte scalariformes; petioli crassi 1-8—2 cm. longi, deciduo-

pubescentes. Flores masculi: basi minute bracteolati, pedicelli 3—5

mm. longi, ferrugineo-pubescentes; perianthium 2 mm. longum; in

alabastro sub-globosum, sub-glabrum carnosum, extus ferrugineo-

pubescens, intus albo-viride glabrum; discus staminalis triangularis,

stipite 1 mm. longo suffultus, cum antheris 13 sessilibus horizon-

talibus aurantiacis vel brunneis distantibus. Flores feminei ignoti.

Fructus oblongus utrinque obtusus, minute fusco-brunneo-tomen-

tellus, 3-3-5 cm. longus, 2-2-5 cm. in diam.; pericarpium 2 mm.

310

Vol. XVI. (1958).

crassum; stipes 7 mm.—1:3 cm. longus, tenuiusculus ut in fructu

tomentellus. Arillus carneo-aurantiacus, crassiusculus, apice laci-

niatum. Semen laeve nitidum.

Tree 6—13 m. high with a few stilt roots. Bark brownish, a few

elongated lenticels present, not flaking or pustulate; sap reddish,

not very abundant. Twigs stout, smooth, rusty tomentulose at the

tips, glabrous and coarsely striate further back. Leaves coriaceous,

oblong, dark green with whitish-green midrib and nerves above,

rather dull when dry, glaucous beneath with brownish midrib and

veins, glabrous except for the young leaves which are covered with

fulvous, caducous tomentum, apex acute, base rounded, slightly

cordate and decurrent on to the petiole; midrib raised on both sur-

faces except at the base on the upper where it is flat, flush with the

leaf surface and 4 mm. broad; nerves 25 pairs or more, boldly

raised on both surfaces, crowded at the base of the leaf, well-

spaced higher up, oblique and interarching at the margins; reticu-

lations faint and close but visible on both surfaces, the principal

ones scalariform; length 33-50 cm.; breadth 15-17 cm.; petiole

stout, 1-8—2 cm. long, pubescent, becoming glabrous. Male flowers:

pedicels 6-8 mm. long with a minute bracteole slightly below the

base of the flower (not median), rusty-pubescent like the flowers;

perianth 2 mm. long, sub-globose in bud, fleshy, glabrous and

whitish green inside; disc triangular with 13 crowded, sessile (prob-

ably not mature and may become stalked later), horizontal, orange

or brownish anthers which touch each other, its stalk 1 mm. long.

Female flowers: not seen. Fruit oblong, obtuse at both ends, dark

brown-tomentulose, 3—3-5 cm. long and 2—2:5 cm. broad; pericarp

2 mm. thick; stalk 7 mm.—1-3 cm. long, rather slender, dark brown-

tomentulose. Avril pinkish-orange, rather thick, covering the seed,

laciniate at the apex. Seed smooth, shining.

TRENGGANU: 38th mile Trengganu-Besut Road, Sinclair & Kiah

S.F.N. 39959 (A, BM, BO, E, K, KEP, L, P, SING holotype); 34th

mile Kuala Trengganu-Besut Road, Sinclair & Kiah S.F.N. 40845 (E,

Kb, SING):

JouHorE: 14th mile Mawai-Jemaluang Road, Corner S.F.N. 29015

(K, SING); 134 mile Mawai-Jemaluang Road, Corner S.F. Nos. 29419

(K, SING) and 29436 (K, SING); S. Kayu Ara, 123 mile Mawai-Jema-

luang Road, Corner S.F.N. 29477 (SING).

DISTRIBUTION: Borneo and probably Sumatra, Grashoff 1019 (L).

At first sight this may appear to be a distinct species but I do not

think it is anything more than a variety of K. glaucescens, distin-

guished from it by the leaves cordate at the base and for the most

part larger. The fruit is tomentulose and may be ridged or not.

‘The bract is near the base of the flower while it is generally above

311

Gardens Bulletin, S.

the middle in typical glaucescens but this may be due to immaturity

of the flowers since their anthers also appear to be sessile, probably

elongating later.

(13) K. plumulosa J. Sinclair, sp. nov.

Synonym: Myristica cantleyi Hk. f. sensu King in Ann. Roy.

Bot. Gard. Calc. 3 (1891) 327 Pl. 168 (excl. specimen Cant-

leyanum = Knema laurina). Knema cantleyi (Hk. f.) Warb.,

sensu Warb., Monog. Myrist. (1897) 554 T. 24 Figs 1-2;

Gamble, Mat. F.M.P. 5, 23 (1912) 238; Ridley, F.M.P. 3

(1924) 68. K. intermedia var. dubia Warb., Monog. Myrist.

(1897) 561; M. glaucescens Wall. Cat. 6810 nom. nud.; M.

furfuracea Hk. f. et Th., sensu A. DC., Prodr. 14 (1856) 206.—

Fig. 14.

K. intermediae proxima sed innovationibus, petiolis iuvenibus,

floribus fructibusque lanosis (non tomentosis nec pubescentibus),

floribus maioribus, pedicellis crassioribus, perianthis persistentibus,

haec species differt. |

Arbor 5-16 m. alta. Ramuli novelli primo dense ferrugineo-

lanosi cum pilis dendroideis plumosis minute barbulatis, deinde

laeves, glabri rubro-brunnei. Folia coriacea oblonga vel oblongo-

lanceolata, primo ramuli eodem modo lanosa, mox glabra, supra

atro-viridia, subtus pallidiora non glauca, 14~30 cm. longa, 5—8-5

cm. lata, utrinque acuta vel interdum basi sub-rotundata; nervi

15-20 pares utrinque elevati ad marginem anastomosantes; reti-

culationes scalariformes et irregulares laxe intermixtae, utrinque

distinctae; petioli 1:5—2-5 cm. longi. Flores basi bracteolati, extus.

dense lanosi, intus glabri, flavido-albi vel pallido-virides vel carnei.

Flores masculi: pedicelli 8 mm.—1 cm. longi, 2—3 mm. crassi, pe-

rianthium coriaceum 6—7 mm. longum, 6—7 mm. latum, ad basim

in lobos tres late ovatos patentes fissum; discus staminalis 4 mm.

in diam., valde mammillatus, stipite 2 mm. longo praeditus; an-

thera 13-15 breviter stipitata. Flores feminei: sessiles; ovarium

globosum, sessile dense ferrugineo-lanosum; stylus brevis; stigma

stellatum in lobis angustis 12 terminatum. Fructus ellipsoideus, 4

cm. longus, 2—2-5 cm. in diam, ut in ramulis novellis lanosus, cum

perianthio et stigmate persistentibus, pericarpium 3 mm. crassum

cum stipite 1 cm. longo. Semen nitidum, brunneo-variegatum, arillo

carneo ad apicem leviter laciniato tectum.

Tree 5-16 m. high, sometimes with a few stilt roots. Bark pale

brown, sap not very copious. Twigs, when young densely covered

with rusty-brown wool, the individual hairs of which are minute

plumes with small barbules; when older reddish brown, smooth

and glabrous. Leaves coriaceous, oblong or oblong-lanceolate, at

312

Vol. XVI. (1958).

ere

- ap ieee eC

as FL / LB -

Fig. 14. Knema plumulosa J. J. Sinclair.

A, Leafy twig with male flowers. B, Male flower. C, Staminal column.

D, Ovary. E, Fruit. A-D from Wray 3126. E from Corner S.F.N.

29367.

313

Gardens Bulletin, S.

first tomentose with the same tomentum as the twigs, soon glabr- —

ous, dark green above, paler green but not glaucous beneath, apex

acute, base acute or less often sub-rotund; nerves 15—20 pairs,

raised on both surfaces, interarching near the margin; reti-

culations scalariform, interwoven with others to form a lax open

network, distinct on both surfaces; length 14-30 cm.; breadth 5—

8:5 cm.; petiole 1-5—2-5 cm. long. Flowers densely felted with

rusty wool outside, glabrous and creamy white or pale green to

pink inside. Male: pedicels 8 mm.—1.cm. long and 2—3 mm. thick

with the bracteole at the base of the flower; perianth coriaceous,

split down to the base by the broadly ovate lobes, 6-7 mm. long

and 6-7 mm. broad, lobes spreading, horizontal, a raised disc

present at their union near the base of the flower on which the

staminal disc rests; staminal disc, 4 mm. across, mammillate, pink,

bearing 13-15 sessile or almost sessile anthers, its stalk 2 mm.

long. Female: 1 cm. long, nearly sessile in few-flowered clusters,

the pedicels 3 mm. long and 4—5 mm. thick, bracteole at the base

of the flower; ovary globular, sessile, densely rufous-tomentose;

style short, stigma stellate with about 12 narrow lobes. Fruit ellip-

soid, 4 cm. long and 2-2-5 cm. broad, densely covered with the

same kind of wool as is found on the young twigs, pedicels and

flowers, the remains of the stigma and also the now much enlarged

perianth persisting; pericarp 3 mm. thick; stalk 1 cm. long. Seed

shining, mottled brown, covered by the carmine-pink aril which is

slightly lacinate at the apex.

PENANG: Wall. Cat. 6810 (A, BM, E, G, Prodr. & Boiss., K) type

of K. intermedia var. dubia; Herb. Hook (K).

PERAK: Scortechini s.n. (CAL); Larut, King Nos. 3350 (CAL, DD,

E, MEL); 5317 (CAL, FI, G, K, L, MEL, P); 5614 (CAL, DD, FI,

K, SING, UPS); 6569 (CAL, G, K, L); 6867 (BM, CAL, FI); 7290

(CAL, FI, K, L); Taiping King 8443 (CAL, G, K) and Wray 2700

(BM, CAL, K, SING); Assam Kumbang, Wray 3126 (CAL, SING);,

Trolak F.R., Ja’amat F.D. 43479 (KEP).

TRENGGANU: Path leading to Kampong Bukit off 203 mile Kuala.

Trengganu-Besut Road, Sinclair S.F.N. 40455 (B, BK, BM, BO, Delhi

Univ., E, K, L, P, PNH, SAN, SING holotype).

SELANGOR: Kuala Lumpur, Ridley 10227 (CAL, SING); Klang

Water-catchment Forest, Burkill S.F.N. 6850 (SING) and 9166 (K,,.

SING); Sungei Buloh F.R., Kiai F.D. 8288 (KEP, SING); Symington

F.D. 44720 (KEP) and Motan K.F.N. 70412 (KEP); Klang Gates,.

Burn-Murdoch 14247 (SING); Hashim or Burn-Murdoch 279 (K,

KEP); Klang, Keheding 192 (FI).

JouHorE: S. Kayu Ara, Mawai-Jemaluang Road, Corner S.F.N. 29367

(BM, K, KEP, SING); S. Kayu, Kiah S.F.N. 32130 (A, K, KEP,

SING); 6th mile Kota Tinggi-Mawai Road, Corner S.F.N. 30972 (A,

BM, BRI, DD, K, KEP, SING); 5th mile Kota Tinggi-Mawai Road,.

Corner S.F.N. 29944 (A, K, KEP SING); Sedili River, Panti F.R.,

Kota Tinggi, Motan K.F.N. 53930 (KEP).

DISTRIBUTION: Confined to Malaya.

314

Vol. XVI. (1958).

This species is in the group with K. intermedia and both are

distinguished from other species by the convex mammillate stami-

nal disc. Both have dark reddish brown, smooth twigs, seen best

when the tomentum is shed. The tomentum is lanose in this spe-

cies and very dense, while it is much shorter and never lanose in

intermedia. The present species is further distinguished from inter-

media by the larger flowers and thicker pedicels and by the perianth

persisting in fruit. Both have stilt roots if growing in wet soil.

Hooker’s description of Myristica cantleyi in Fl. Br. Ind. 5 (1886)

110 (Cantley, Singapore, sine numero, the type) is K. laurina and

has 6—8 anthers as Hooker points out. K. plumulosa has never

been found in Singapore. King confused the Perak specimens of

K. plumulosa with cantleyi and his plate No. 168 in Ann. Roy.

Bot. Gard. Calc. of cantleyi is plumulosa. Warburg, Gamble and

Ridley followed King including the present species under K. cant-

leyi. Cantley Nos. 195 and 6867 and one sheet without a number,

the type, Cantley, date 1885, Bukit Timah, Singapore, are all

laurina and were quoted by Warburg as cantleyi.

(14) K. intermedia (BI.) Warb., Monog. Myrist. (1897) 564 T.

25 Figs 1-2; Gamble, Mat. F.M.P. 5, 23 (1912) 239; Ridley,

F.M.P. 3 (1924) 68; Corner, Wayside Trees of Malaya 1

(1940) 477.

Basionym: Myristica intermedia Bl., Rump. 1 (1835) 187;

Hk. f. et. Th., Fl. Ind. 1 (1855) 158; A. DC., Prodr. 14, 1

(1856) 206; Miq., Fl. Ind. Bat. 1, 2 (1858) 70 et in Ann. Mus.

Bot. Lugd.-Bat. 2 (1865) 51; Hk. f., Fl. Br. Ind. 5 (1886) 112;

King in Ann. Roy. Bot. Gard. Calc. 3 (1891) 317 Pl. 154.

Synonyms: M. iteophylla Migq., Fl. Ind. Bat. 1, 2 (1858) 59.

M. corticosa Hk. f. et. Th. var. decipiens Miq., in Ann. Mus.

Bot. Lugd.-Bat. 2 (1865) 51. M. glabra de Vriese (non Bl.) PI.

Indiae Batavae Orientalis (Pl. Reinwardt.) (1857) 85.—Fig.

15. Plate IIB.

Tree 6-10 m. high, occasionally up to 20 m., stilt roots present.

Bark light rufous-brown, smooth to slightly rough, sap pinkish,

scant. Twigs, leaves and petioles when young, covered with very

short, stellate scurf and plumose hairs, the twigs smooth, angled

and reddish-brown. Leaves coriaceous, lanceolate to oblong-lan-

ceolate, dark green and shining above, glaucous beneath, apex

acute or acuminate, base cuneate; midrib and the 12—22 pairs of

nerves distinct and raised on both surfaces, the latter opposite or

alternate, curving and ascending slowly, interarching at the margin;

reticulations raised on both surfaces, mostly scalariform but not

315

Gardens Bulletin, S.

FI}

ae,

Fig. 15. Knema intermedia (Bl.) Warb.

A, Twig with male flowers. B, Leafy twig with male flowers. C & D,

Male flowers showing the staminal column. E, Fruit. F, Leaf with

obtuse base. G, Twig with female flowers. H, Female flower. I,.

Ovary. A-D from Sinclair S.F.N. 40591. E, G, H from Ridley 2107.

F from Alvins 34.

316

Vol. XVI. (1958).

straight; length 11-22 cm.; occasionally 30 cm. or over; breadth

3-4-5 cm., occasionally up to 9 cm. in sapling leaves; petiole 1:5—2

cm., rather slender, glabrous when mature. Flowers rusty, scurfy-

stellate or with plumose hairs outside, creamy white, glabrous and

striate inside. Male: numerous on woody tubercles; pedicels 8 mm.—

1 cm. long and 1 mm. or less thick and with a minute, 1 mm. long

bracteole below the base of the flower; lobes coriaceous, triangular-

ovate, spreading, sub-acute, 4-5 mm. long and 4-5 mm. broad;

staminal disc red, 2-5 mm. broad, mammillate with 12—15 (average

13) sessile, yellow anthers, its stalk very short, 1 mm. long.

Female: larger than the male, 7-8 mm. long but fewer in the um-

bels, pyriform, their pedicels 7-8 mm. long and 1:5—2 mm. thick;

ovary 2 mm. long, sessile, ovoid, rufous-tomentose; stigma sessile

with 6-8 or more narrow lobes. Fruit solitary or in a cluster of

2-3, ellipsoid, light yellow to orange when fresh but rusty when

dry, covered with stellate scurf which rubs off easily, 3—4 cm. long

and 1-8—2-2 cm. broad, the circumferential line of dehiscence pro-

minent; pericarp 3-4 mm. thick; stalk 8 mm.—1-2 cm. long. Seed

mottled brown, covered with the carmine-pink or red aril, laciniate

(sometimes nearly entire) at the top only.

KELANTAN: Sungei Keteh near Gua Nanak, Nur & Foxworthy S.F.N.

1212] (BRI, K, SING, UC).

PENANG: Porter, date 1840 (M); Cantley 449 (SING); Haniff Nos.

3839 (DD, NY) and 328 (DD, NY); Reserve line, West Hill, Curtis

1044 (K, SING); Government Hill, Curtis 3402 (CAL, K, SING);

Burkill Nos. S.F.N. 6261 (SING) and 4582 (SING); Waterfall Gar-

dens, Ridley 10786 (CAL).

PROVINCE WELLESLEY: Tassek Glugor, Curtis, April 1902 (SING).

PERAK: Scortechini Nos. 631 (CAL, MEL, SING) and 803 (BM,

CAL, SING); Wray 3010 (BM, CAL, K, SING); Larut, King Nos.

5419 (CAL, DD, E, FI, G, K, L); 6146 (BM, CAL, DD, E, FI, G, K,

L, SING); 6371 (CAL, DD, K, MEL, SING); 6704 (CAL, K); 7576

(CAL, K, SING, UPS); Batu Togoh, Wray 2130 (CAL, SING).

SELANGOR: Telok Reserve, Klang, Nur S.F. Nos. 34043 (A, K, L, S,

SING, UC) and 34046 (A, BM, L, S, SING, UC); Sungei Tinggi,

Kuala Selangor, Nur S.F.N. 34125 (A, K, L, S, SING, UC); Bukit

Lagong F.R., Lindong K.F.N. 55792 (KEP).

Maacca: Maingay 1288 (BM, FI, K) also numbered 1004; Maingay

1281 (A, BM, CAL, K, L); Alvins or Cantley 525 (K, SING); Griffith

4359 (A, CAL, K, M, P); Alvins 653 (SING); Merlimau, Derry 1149

(CAL, SING) and Alvins or Cantley 34 (K, SING); Tanjong Tuan

F.R. (Cape Rachado) Sow & Lindong K.F.N. 70489 (KEP); Sungei

Udang, Sinclair S.F.N. 40591 (BM, BO, E, K, L, P, SING).

JouoreE: S. Sedili, Mawai, Corner S.F.N. 29268 (K, SING); Mawai,

Corner S.F.N. 28453 (K, SING); S. Kayu, Mawai-Jemaluang Road,

Corner S.F. Nos. 29252 (A, BM, K, SING) and 21333 (A, BM, K,

SING).

SINGAPORE: Cantley 83 (K) and s.n. (CAL); Lobb, Herb. Hookeria-

num s.n. (CAL, K); Lobb 315 (BM, E); Changi, Ridley 1820 (CAL,

K, SING); Loyang, Changi, Ridley 2039 (BM, CAL, SING); Jurong,

Corner S.F.N. 26050 (A, BR, K, NY, SING); Tanglin, Hullett 5739

317

Gardens Bulletin, S.

(SING); Kranji, Mat 5972 (SING); Bukit Timah, Ridley Nos. 2107

(BM, CAL, SING) and 6734 (SING); Cantley 29 (K, SING); Liew

S.F.N. 36441 (A, BRI, DD, K, KEP, SING); Sinclair, 10th June, 1953

(E); Bukit Timah, Warburg 4853 (L, M); Rock Path, Bukit Timah,

Sinclair S.F.N. 39571 (E, SING); Tyershall, Ridley 5065 (SING);

Cluny Road, Ridley 4827 (SING); Mandai Road, Corner S.F.N. 34544

(A, K, KEP, SING); Burkrushauri Jungle, Hullett, 7th May, 1893

(SING); Gardens’ Jungle, Ridley 4818 (CAL, SING) and Sinclair,

26th Feb., 1953 (E); Arboretum, Furtado Gardens’ No. 1194 (SING,

UC); front of carpenter’s shed, Botanic Gardens, Ridley 1646 (SING);

Rockery, Botanic Gardens, Ridley 9222 (CAL, K, SING); South side

of Seletar Reservoir, Sinclair S.F.N. 39564 (E, SING).

DISTRIBUTION: Borneo, Sumatra, Banka, Java.

Type MATERIAL: Type localities are Salak and Tjampia but no col-

lector’s numbers quoted by Blume.

This species is nearest to K. plumulosa. For differences see under

that species. Flower buds were observed on the tree in the Singa-

pore Arboretum and they remained closed for six weeks before

opening. They had probably been in the unopened stage for quite

a considerable time prior to having been noticed. There is one

species K. rufa in Sarawak which resembles this one in the leaves

but the twigs are stouter, the pedicels much longer and the disc

does not appear to be mammillate. So far it is known from the type

collection only.

(15) K. retusa (King) Warb., Monog. Myrist. (1897) 612 T. 25;

Gamble, Mat. F.M.P. 5, 23 (1912) 249; Ridley, F.M.P. 3

(1924) 72.

Basionym: Myristica retusa King in Ann. Roy. Bot. Gard.

Calc. 3 (1891) 330 Pl. 171.

Tree 12-18 m. high. Twigs stout, striate, pale brown. Leaves

coriaceous, oblong, elliptic-oblong or obovate, obtuse or retuse at

the apex, rounded or sub-cordate at the base, upper surface glabr-

ous, shining, lower pale brown or whitish from a uniform layer of

dense, minute, cinnamomeous hairs, margins recurved; nerves 16—

24 pairs, spreading horizontally and straight near the base, gra-

dually more and more curved towards the apex where they inter-

arch; reticulations scalariform, numerous, sub-parallel, prominent

above; length 35-55 cm.; breadth 14-21 cm.; petiole 1-2 cm.

long. Flowers unknown. Fruits 2—3 on short, thick tubercles from

the axils of fallen leaves, ovoid, apiculate, gibbous on one side at

the base, 5-6 cm. long and 5 cm. broad; pericarp thick, minutely

chocolate-tomentose. Aril very small, merely embracing the base

of the seed, thin, much fimbriate. Seed narrowly obovoid, smooth.

PERAK: Gunong Bubu, King 7690 (CAL, G, K, L, P).

DISTRIBUTION: Only known from the type collection.

318

Vol. XVI. (1958).

It seems strange that this striking species has never been collected

again. The description of the aril (from King) does not seem

quite correct for a Knema but it may have shrunk on drying. I

suspect that this species is related to K. mandaharan and to glau-

cescens var. cordata but one cannot be sure until flowers are avail-

able. Every effort should be made to find out more about it, es-

pecially as the flowers are unknown.

(16) K. mandaharan (Miq.) Warb., Monog. Myrist. (1897) 553

Ee 25.

Basionym: Myristica mandaharan Migq., Fl. Ind. Bat. Suppl. 1

(1861) 384 et in Ann. Mus. Bot. Lugd.-Bat. 2 (1865) 48.—

Fig. 16.

Tree up to 20 m. high. Bark greyish-brown, drab, longitudinally

fissured, flaky, inner pinkish-brown; sap reddish, watery, scant.

Twigs stout, greyish-brown with a few faint striations here and

there, at first densely covered with rusty scurf which soon rubs off,

finally glabrous. Leaves coriaceous, dark green above, shining, dry-

ing a rich, medium brown, glaucous beneath, glabrous except when

young when they are covered on both sides with a short, rusty-

stellate tomentum, oblong-lanceolate, rounded at the base but not

or only slightly cordate, acute at the apex; midrib stout, raised on

both surfaces, 5-6 mm. broad at the base, where on the lower

surface of the leaf it is often minutely pustular for a short distance

upwards; nerves 22—26 pairs, raised on both surfaces, curving and

ascending to anastomose at the margin; reticulations scalariform

with a finer network interspersed among the scalariform series; faint

above, distinct beneath; length 30-50 cm.; breadth 8-5—14 cm.;

petiole 2—2:5 cm. long, very stout, at first scurfy-tomentose, later

glabrescent. Male flowers: (rather young) few-flowered, obovoid,

3—4 mm. long (probably longer later), on a pedicel 3 mm. long

with a minute bracteole at the base of the perianth; perianth densely

rusty-tomentose outside, the hairs minute, plumose barbules like

those on the fruit, glabrous inside but colour not known; staminal

disc flat or slightly convex on a 1 mm. long, striate stalk, anthers

16-17, horizontal, very crowded, without intervening spaces, ses-

sile. Female flowers 1 cm. long and 6 mm. broad, on a short pedi-

cel; ovary ovoid, tomentose, 5 mm. long; stigma a sessile, many-

lobed disc, the lobes linear. Fruit nearly as broad as long, the base

broad and horizontal, the apex rounded and somewhat flattened,

the circumferential ridge of dehiscence prominent, rusty-tomentose,

the hairs, plumose barbules about 1 mm. long; length 3-4-5 cm.;

319

Gardens Bulletin, S.

twig. B, Male flower, immature. C, Staminal column, imma-

ture. D, Fruit. A from Corner 28-8-1932. B, C from Corner 7-11-

1935. D from Corner S.F.N. 30486.

Leaf

Fig. 16. Knema mandaharan (Miq.) Warb.

320

Vol. XVI. (1958).

breadth 3-3-5 cm.; pericarp 8 mm. thick; stalk 5-7 mm. long and

5 mm. thick. Aril thick, covering the seed, laciniate at the apex.

PERAK: Larut, King 6515 (CAL, FI, K).

TRENGGANU: Ulu Bendong, Kemaman, Corner S.F.N. 30284 (A,

K, SING); Bukit Kajang, Kemaman, Corner, 7th Nov., 1935 (SING);

10th Nov., 1935 (SING); 12th Nov., 1935 (SING); Ulu Kajang,

Corner S.F.N. 30486 (A, DD, SING); Belara F.R., Kuala Trengganu,

Wood K.F.N. 76058 (KEP, SING); 23rd mile Kuala Trengganu-

Besut Road, Sinclair and Kiah S.F.N. 40753 (E, K, SING).

JOHORE: S. Sedili, Corner, 28th Aug., 1932 (SING).

DISTRIBUTION: Sumatra. Type from Province Priaman, De Vriese

(A, CAL, L, LE, S, W); Sarawak, Purseglove Nos. 4549 (SING) and

5367 (SING) and North Borneo, G.H.S. Wood SAN 16709 (L, SING).

This species recalls K. furfuracea but the resemblances are

superficial; they apply to the leaves and twigs and not to the flow-

ers. The leaves differ from those of furfuracea in not having a cor-

date or only slightly cordate base, less narrowed towards the base,

the nerves fewer and more widely spaced. The twigs have a darker

coloured tomentum which does not persist for so long on the tips

of the twigs and petioles. The bark on the twigs does not tend to

crack and is not blackish. The fruit is less elongated, being not

much broader than long and with a broad horizontal base. The

bracteole is at the base of the perianth and not at the middle,

while the flowers appear to be smaller with more numerous,

crowded anthers which touch each other.

K. mandaharan has not been recorded from Malaya and the spe-

cimens quoted here were doubtfully labelled K. furfuracea in the

Singapore Herbarium.

(17) K. oblongifolia (King) Warb., Monog. Myrist. (1897) 586

T. 24 Figs 1-2; Gamble, Mat. F.M.P. 5, 23 (1912) 246; Rid-

ley, F.M.P. 3 (1924) 71.

Basionym: Myristica oblongifolia King in Ann. Roy. Bot.

Gard. Calc. 3 (1891) 313 Pl. 148 Figs 1-7, 10 and 11 exclud-

ing the Penang specimens which are K. laurina.

var. oblongifolia—Fig. 17, A-B.

A shrub or small tree 10-13 m. high. Bark, no details available.

Twigs when young, rusty-stellate-tomentulose at the extreme tips

which are angled, later smooth and reddish brown. Leaves thickly

membranous, dark glossy green above, glaucous beneath, glabrous

except when young when there is some stellate scurf on the lower

surface, oblong, oblong-lanceolate or oblanceolate, base rounded,

often cordate, less often bluntly acute, apex acute or obtuse; mid-

rib raised on both surfaces; nerves about 30 pairs, oblique, parallel,

321

Gardens Bulletin, S.

z\)

XS SX

peso

ben

We = s

UES ZR

KSIPS pe

Be le AW

Kp yyy)

PSS ey)

px ‘\

gp IN

73S

secs

= Peay

‘aes _———,, a RETR

NES : ~ Z

Pest =

Fig. 17. Knema oblongifolia (King) Warb. .

A, var. oblongifolia, leafy twig with male flowers. B, the same with

male flowers. C, var. monticola, leafy twig with fruit. D, the same

with male flowers. E, the same with male flower. F, the same with

staminal column. G,. the same, ovary. A-B from King 4534. C

from Haniff S.F.N. 21024. D-—G from Purseglove 4246 & 4275.

322

Vol. XVI..(1958).

interarching near the margins, distinct and impressed above, pro-

minent beneath; reticulations scalariform, distinct above, prominent

beneath; length 20-43 cm., average 33 cm.; breadth 9—16 cm.,

average 11 cm.; petiole 1-5—2-3 cm. long, stout, scurfy-tomentose

at first, later glabrous. Male flowers in 6—8 flowered umbels from

short’ axillary tubercles, buds 4 mm. in diam., oblong-obovoid;

pedicels 6-8 mm. long, the bracteole minute, sub-orbicular, close

to the base of the perianth; perianth lobes obovate, blunt or sub-

acute, thick, minutely rufous-stellate-hairy outside, glabrous inside;

anthers 6—10, sessile, sub-erect on a stalked, slightly concave, or-

bicular disc. Female flowers sessile; buds ovoid, lobes of the

perianth ovate, blunt; ovary sessile, rusty-tomentose, style present

but shorter than the ovary, cylindric, glabrous; stigmas of 2

rounded, compressed erect lobes. Fruit in pairs or solitary, shortly

stalked, sub-obovoid or ovoid, pointed at the apex, often ridged,

2—2:2 cm. long and 1-5-2 cm. broad; pericarp thin, densely covered

with harsh, very short, rusty tomentum which tends to rub off when

old; stalk 5 mm. long. Aril thin. Seed mottled.

PERAK: Gopeng, King Nos. 5434 (CAL, E, SING) and 5983 (CAL,

E, FI, K, L, MEL, P, SING, UPS) both type material; Sungei Rajah

King 835 (CAL, FI, K) type material.

SELANGOR: Ulu Gombak, Hume F.M.S. Mus. Herb. Nos. 9241

(SING) and 9261 (SING); Bukit Hitam, Ridley, May 1896 (SING).

DISTRIBUTION: Lowland forests in Malaya.

Not nearly so common nor as often collected as its mountain

variety. There are no notes on bark characters nor on the colour of

the flower.

var. monticola (King) Warb., Monog. Myrist. (1897) 587; Gamble,

Mat. F.M.P. 5, 23 (1912) 246; Ridley, F.M.P. 3 (1924) 71.

Basionym: Myristica oblongifolia var. monticola King in Ann.

Roy. Bot. Gard. Calc. 3 (1891) 314 Pl. 148 Figs. 8-9.—

Fig. 17, C-G. |

Tree about 10 m. high. Bark greyish brown, the outer layers

thin, flaking slightly. Leaves proportionately smaller than in the

type, 15-25 cm. long and 4—7 cm. broad, also tending to be more

coriaceous; petioles and twigs more slender; apex of leaf acute or

acuminate, base usually acute, sometimes rounded; nerves 18—25

pairs. Male flowers: 5 mm. long, similar to those of the type but

sessile or on 1 mm. stalks, their pedunculate tubercles occasionally

bifurcate; staminal disc on a slightly longer stalk (2:5 mm. long);

anthers 6-10, erect, sessile. Female flowers 6—8 mm. long, sessile

or on a stout 1 mm. long pedicel with bract at the base of the

323

Gardens Bulletin, S.

perianth, light-buff stellate-tomentose outside and red inside, urceo-

late and split down about 1/3, the three lobes spreading; ovary

rusty-stellate-tomentose, 3 mm. in diam.; style 2 mm. long, slender,

stigma 2—3 lobed, lobes unequal. Fruit as in type, 2-2:3 cm. long

and 1-5-2 cm. broad.

KEDAH: Bukit Langot, Gunong Bintang, Haniff S.F.N. 21024 (K,

SING).

KELANTAN: Chaning, Ridley 4th Feb., 1917 (K).

PERAK: Scortechini s.n. (CAL, FI, G, K, L) type material; King

10953 (CAL, DD, FI, G, L) type material; Larut, King Nos. 3582

(CAL, FI, K); 3810 (BM, CAL, DD) and 6330 (BM, CAL, DD, K,

MEL) all type material; Taiping, King 8322 (CAL, K, SING) type

material; lower camp, Gunong Batu Puteh, Wray Nos. 993 (CAL,

SING); 1077 (CAL, SING) and 1087 (CAL, DD, K, L) all type

material; Maxwell’s Hill, Curtis 2050 (SING); Sinclair S.F.N. 38704

(BM, E, K, SING).

PAHANG: Ulu Serau, Osman F.D. 28358 (SING); Fraser’s Hill,

Corner, 12th Aug., 1987 (SING); Nur S.F.N. 11249 (K, SING); Hen-

derson F.M.S. Mus. Herb. 11545 (SING); Sungei Yet, Fraser’s Hill,

Nur S.F.N. 11126 (K, SING); Richmond, Fraser’s Hill, Purseglove

Nos. 4246 (BKF, BM, BO, E, K, L, P, SING); 4275 (BM, BO, E, K,

L, P, SING) and 4276 (BM, BO, E, K, L, SING); Sungei Lemoi,

Ja amat F.D. 28191 (K, SING); Boh Plantations, Cameron Highlands,

Nur, 20th April, 1937 (SING).

aces Gunong Moyang, Ulu Selangor, Symington F.D. 56703

DISTRIBUTION: Confined to mountains in Malaya, altitude 1,000—

1,330 metres.

K. oblongifolia is quite a distinct species but I have sorted out

from herbarium covers a mass of sheets which do not belong here

at all. The species most commonly confounded with it is Knema

laurina, especially the large leaved, nearly glabrous forms with the

large fruit. The Penang specimens quoted by King and included in

his type material must be excluded as they are this form of K.

laurina. The typical lowland form of K. oblongifolia has large,

thickly membranous leaves which are often cordate at the base

and have about 30 pairs of oblique parallel nerves. In the mountain

form, var. monticola the leaves are smaller, more coriaceous, with

fewer nerves and the base is usually acute, less often rounded. The

shortly-stalked obovoid fruit, pointed at the apex is a good specific

character for both forms. Another good diagnostic character is in

the twigs; the tips are angled and the younger parts are smooth,

shining and reddish brown.

K. oblongifolia seems to be allied to globularia as well as to

laurina but the similarity to globularia is not so apparent at first

sight. It is in the flowers that the relationship may be seen. The

anthers are sessile in both; similarly the bracteole is at the base of

the perianth and the style elongated or at least present with a 2—3-

lobed stigma.

324

Vol. XVI. (1958).

(18) K. globularia (Lamk) Warb., Monog. Myrist. (1897) 601.

Basionym: Myristica globularia Lamk, Mém. Ac., Paris (1788)

162 (non BI. in Rumphia).

Synonyms: K. corticosa Lour., Fl. Cochinch. (1790) 742 [in-

cluding var. tonkinensis Warb.]; Warb., Monog. Myrist. (1897)

593 T. 25 Figs 1-4; Lecomte, Fl. Gén. de’ L’Indo-Chine 5 (1914)

105 Figs 10 and 14. M. corticosa (Lour.) Hk. f. et Th., Fl. Ind.

(1855) 158 pro parte (see notes); A. DC., Prodr. 14, 1 (1856)

205 pro parte (see notes). M. glaucescens Hk. f. et Th., Fl. Ind.

1 (1855) 157 quoad sp. Mal. et Burm. pro parte, (altera pars

= K. malayana Warb.). M. sphaerula Hk. f., Fl. Br. Ind. 5

(1890) 859. K. sphaerula (Hk. f.) Airy-Shaw in Kew Bull.

(1939) 545. M. lanceolata Wall. Cat. 6794 nom. nud. M. mis-

sionis King in Ann. Roy. Bot. Gard. Calc. 3 (1891) 321 Pl. 158.

K. missionis (King) Warb., Monog. Myrist. (1897) 602 T. 24

Figs 1-3; Gamble, Mat. F.M.P. 5, 23 (1912) 247; Ridley,

F.M.P. 3 (1924) 71; Corner, Wayside Trees of Malaya 1

(1940) 417 Text Figs 159 and 161. K. petelotii Merr., in

Journ. Arnold Arb. 23 (1942) 164.—Fig. 18.

Tree 6—20 m. high. Bark flaking in small, thin scales; sap red,

copious. Twigs longitudinally striate, at first rusty-tomentulose,

later glabrous and dark brown to black. Leaves membranous or

thinly coriaceous, lanceolate to oblong-lanceolate, dark green,

glossy, glabrous above and drying olive-green or with a blackish

metallic lustre, glaucous and slightly pubescent beneath when

young, later glabrous, apex acute or acuminate, base acute; midrib

on upper surface lying in a groove, lower raised and finely striate

when dry; nerves 13-18 pairs, slender, raised on both surfaces,

drying reddish-brown, gradually ascending and nearly parallel, line

of interarching faint, often not visible; reticulations faint on both

surfaces, a primary set forming a lax network, interwoven into a

closer secondary set; length 8—17 cm.; breadth 1-5-5 cm.; petiole

8 mm.—1 cm. long. Flowers numerous with rusty-stellate or den-

droid tomentum outside, glabrous and cream-coloured inside, aris-

ing on axilliary, 8 mm.—1 cm. long, woody tubercles which are

occasionally bifurcate. Male flowers trigonous in bud, 3 mm. long,

split half-way down into 3 lobes; pedicels 5 mm. long, rusty-tomen-

tose, striate, thickened just below the base of the flower where the

minute bracteole is situated; staminal disc sometimes convex when

young, later flat or sub-concave, bearing 10—13 sessile, dark red

or reddish-purple, slightly erect anthers; stalk 0-5 mm. long. Female

flowers stouter than in the male, 5 mm. long, urceolate, on pedicels

3 mm. long; ovary conoidal, rusty-tomentose, 3 mm. long, including

325

Gardens Bulletin, S.

Fig. 18. Knema globularia (Lamk) Warb.

A, Leafy twig with male flowers. B, Male flower. C, Male flower dis-

sected to show staminal column. D, Staminal column from above.

E, Ovary. F, Style. G, Fruit. A-D from Ridley 1541. E-F from

Curtis 700. G from Sinclair S.F.N. 39240.

326

Vol. XVI. (1958).

the style and the bifid stigma, the lobes of which are sometimes

again bifid. Fruit orange, sub-globose, rusty-pubescent, finally nearly

glabrous, the longitudinal circumferential line of dehiscence often

prominent; length 1-5-2 cm.; breadth 1-3—1-5 cm.; pericarp 1-5-2

mm. thick; stalk 5 mm.—1 cm. long with a small collar at the base

of the fruit. Aril red, entire or with a few fimbriations at the apex.

Seed smooth, sub-globose.

Lower SIAM: Renong, Haniff 276 (K, SING); Hamid C.F. 2958 (K,

SING); Koh Prap, Franck 473 (SING); Kaw Deng, Annandale, 28th

Jan., 1916 (CAL, K, SING); Telok Udang, Terutau, Hanif & Nur

S.F.N. 7500 (BRI, CAL, K, SING); Kopak, Ban Krap, Bukit Tinggi,

Haniff & Nur S.F.N. 2736 (CAL, K, SING); Puket, Curtis 2925 (CAL,

K, SING): Setul, Ridley 14959 (CAL, K, SING).

PERLIS: Chupeng, Ridley 14958 (BM, CAL, K, SING): Bukit Ketri,

Henderson S.F.N. 22978 (A, CAL, SING).

KepAH: Pulau Chupak, Langkawi, Haniff & Nur 7569 (A, SING,

UC); Terutau, Langkawi, Robinson 6383 (K, SING); Kuala Kuah,

Langkawi, Haniff 15497 (BM, K, SING); Pulau Seluang, Langkawi,

Corner, 22nd Nov., 1941 (KEP, SING); Langkawi F.R., Abdulla

K.F.N. 51737 (KEP); behind the sanatorium, Langkawi, Wyatt-Smith

K.F.N. 71198 (K, KEP); Sungei Petani, Meh F.D. 10176 (SING); Pu-

lau Langsong, Curtis, June 1890 (SING); Bukit Tanjong, Haniff S.F.N.

10401 (K, SING).

KELANTAN: Gimlette, no data (SING).

PENANG: Wall. Cat. 6788 (K) type of M. missionis; Batu Feringi

and Muka Head, Curtis 700 (BM, CAL, K, SING): Pulau Boetong,

Curtis 935 (K, SING); Telok Bahang, Curtis 1559 (CAL, K, SING).

TRENGGANU: Kuala Trengganu, Holttum S.F.N. 15156 (BRI, SING);

Bukit Chenering, Sinclair & Kiah S.F.N. 40390 (SING).

PAHANG: Pulau Manis, Ridley 2262 (BM, CAL, K, SING); Rompin,

Mahamud F.D. 14999 (KEP, SING); Pulau Chibeh, P. Tioman, Cor-

ner S.F. Nos. 29824 (K, SING) and 298328 (K, SING); Telok Paya,

P. Tioman, Nur S.F.N. 21740 (A, DD, SING); Joara Bay, P. Tioman,

Burkill 877 and Juné 1915 (BRI, CAL, K, SING); Kuantan, Lambak

C.F. 4169 (BM, SING); Pengorok, Daud F.D. 8153 (SING); Kuala

Pahang, Kuala Perawas, Ridley 1541 (BM, MEL, SING); Pulau Du-

chong, Corner S.F. Nos. 29855 (K, KEP, SING) and 29859 (A,

SING): Kuala Bok, Ridley 2263 (CAL, K, SING).

Matacca: Maingay 1282 (CAL, E, K); Cuming 2315 (A, BM, C, G

& Boiss, K, L, M, UPS): Griffith 4344 (A, BM, C, CAL, DD, FI, G&

Boiss., K, M, P, S); Hervey, Aug. 1886 (CAL); Miller, no data

(CAL, DD, E, L, M, MEL); Alvins, 16th Jan., 1886 (K, SING);

Pulau Undan, Alvins 33 (K, SING); roadside, Malacca, Corner S.F.N.

30754 (KEP, SING); Bukit Bruang. Cantley 31 (K, SING) type of M.

sphaerula; Pulau Besar, Goodenough 7122 (CAL, K, SING) and Fox,

Dec. 1894 (SING).

JoHorE: Pulau Aor, Henderson S.F.N. 18244 (A, K, SING); Pulau

Tinggi, Burkill S.F.N. 944 (K, SING); Pulau Sauh, Sinclair S.F.N.

40612 (BM, BO, E, K, L, P, SING).

SINGAPORE: Wall. Cat. 6794, date 1822 (A, BM, BR, G & Prodr.,

K. M) type of M. lanceolata nom. nud.; Pulau Ubin, Ridley 4817

(CAL, K, SING): Changi, Ridley 13342 (SING); Pulau Sakijang

Pelepah, Sinclair S.F.N. 39240 (E, L, SING); Cluny Road, Ridley 4816

(CAL, K, SING).

327

Gardens Bulletin, S.

DISTRIBUTION: Burma (Mergui and Tenasserim), Indo-China, Yun-

nan?, Siam.

Type: The type of M. globularia Lamk is Sonnerat (P). I have seen

the isotypes at Geneva and Copenhagen. The type of K. corticosa Lou-

reiro is Loureiro s.n. (BM) from Cochinchina.

In Malaya this small tree usually occurs on the rocky seashores

of small islands. It is easily identified from the vegetative charac-

ters. The narrow leaves, the lower midrib, striate when dry and

the rusty, longitudinally striate twigs which become glabrous and

dark brown or black are good distinguishing features. The bracteole

at the base of the flower and the sessile anthers place it with laurina

and oblongifolia.

It is also allied to K. attenuata from South India but in that spe-

cies the flowers and fruit are much larger and the leaves broader.

The twigs, leaf-texture, venation and colour on drying are very

similar.

I have seen the isotypes of Myristica globularia, Sonnerat s.n.

at Geneva and Copenhagen. They certainly agree with the Malayan

material given under various synonyms and with the type of K.

corticosa Loureiro at the British Museum. The type locality is not

Java which has been given by mistake. Myristica sphaerula and

missionis are later synonyms.

In making the new combination M. corticosa, Hk. f. and Th.

included a number of specimens which are specifically different from

the type. I have examined the specimens of M. corticosa deposited

in the DC. Prodromus Herbarium at Geneva and found only one

of them, M. lanceolata, Wall. Cat. 6794, collected at Singapore

to be K. globularia. The others are as follows:—Herb. Hook. s.n.,

Chittagong is K. angustifolia; Cuming 844, Philippines, is K. glo-

merata; Zollinger 2650, Java, is K. glaucescens Jack; Labillardiére

s.n., Philippines named K. cinerea is K. glomerata. True globularia

does not extend further north than Tenasserim and Mergui and

there are no authentic records of it from Chittagong. King, under

Knema (glauca) glaucescens, quotes Maingay 1282 and certain other

Malayan specimens which are globularia while the rest are mala-

yana and glaucescens Jack. K. wangii H. H. Hu from Meng-la,

Jenn Hsein, Yunnan, Wang 80634 (K, P) does not seen different

from K. globularia but more material should be collected and ex-

amined as K. globularia is a tree of rocky sea-coasts and not of an

inland habitat. K. wangii and a species K. yunnanensis given with

it in the Kew Index are nomina nuda. They are merely mentioned

in an observation in the Royal Hort. Soc. Journ. 63 (1938) 387

without a description.

328

Vol. XVI. (1958).

(19) K. laurina (BI.) Warb., Monog. Myrist. (1897) 606 T. 24

Figs 1-3; Gamble, Mat. F.M.P. 5, 23 (1912) 248: Ridley,

F.M.P. 3 (1924) 72.

' Basionym: Myristica laurina Bl., Rumphia 1 (1835) 139 T.

61; A. DC., Prodr. 14, 1 (1856) 206; Mig., Fl. Ind. Bat. 1, 2

(1858) 71 et Suppl. (1861) 385 et in Ann. Mus. Bot. Lugd.-

Bat. 2 (1865) 51; Hk. f., Fl. Br. Ind. 5 (1886) 112; King in

Ann. Roy. Bot. Gard. Calc. 3 (1891) 319 Pl. 156.

Synonyms: M. tomentosa Bl., Bijdr. (1825) 577 non Thb.

M. cantleyi Hk. f., Fl. Br. Ind. 5 (1886) 110 non sensu King.

M. furfurascens Gandoger in Bull. Soc. Bot. France 66 (1919)

226 in clavi. K. oblongata Merr. in Journ. Roy. As. Soc. Str.

Br. 85 (1922) 190. K. obovoidea Merr. in Univ. Calif. Publ.

Bot. 15 (1929) 75.—Fig. 19.

Tree 4-20 m. high with rather slender branches. Bark reddish

brown, nearly smooth or very slightly cracked, not furrowed, of no

great thickness, fairly hard; sap copious, red, watery. Twigs when

young, also young petioles and the under-surface of young leaves,

harsh to the touch owing to the rusty, scurfy, stellate tomentum,

the twigs later slowly becoming darker and glabrous. Leaves mem-

branous to coriaceous, very variable in shape even on the same

tree, oblong, oblong-obovate or oblanceolate, broadest-above the

middle, acute at the apex and gradually narrowed to the acute,

rounded or cordate base (often acute or rounded on the same tree),

dark green, shining and glabrous above, except the midrib which

usually becomes glabrous later, sub-glaucous and stellate-pubescent

to tomentose beneath (the amount of tomentum varying consider-

ably); nerves 12—24 pairs, oblique, interarching near the margins;

reticulations fine, faint above, distinct beneath; length 9-28 cm.;

breadth 3—8 cm.; petiole stout, 1 cm. long, rough with stellate hairs.

Flowers axillary on short, woody, 2—3 mm. long tubercles, tawny-

stellate-tomentose outside and red and glabrous inside; bracteoles

1 mm. long and at the base of the perianth. Male on rusty-stellate-

tomentose, 3—5 mm. long pedicels; perianth 3-5 mm. long, 3-lobed,

papillose inside, and with guide-lines, the apices of the lobes thick

and forming incurved protuberances over the staminal column;

column red, 2 mm. long, the stalk 1-5 mm. long and the disc 5 mm.

across, slightly depressed in the centre, anthers 6-11 (average

number 7 or 8, rarely 11) white, sessile. Female fewer, sessile,

larger, 7 mm. long, the lobes reaching half-way down; stigma

green, 2-lobed, raised on a style, which together with the stigma is

1-5 mm. long; ovary about 3 mm. long, sub-globose, tawny-tomen-

tose. Fruit ovoid or ellipsoid, 25-3 cm. long and 1-5-2 cm.

329

Gardens Bulletin, S.

TURMAMI DEL-

Fig. 19. Knema laurina (Bl.) Warb.

A, Leafy twig with male flowers. B, Leaf with rounded base. C, Leaf

with slightly cordate base. D, Female flower dissected to show

ovary. E, Male flower dissected to show staminal column. F, Fruit.

G, Fruit opened to show aril and seed. A, B, E from Sinclair

S.F.N. 39475. F, G from Corner S.F.N. 28615. )

330

Vol. XVI. (1958).

broad; pericarp densely covered with coarse, rufous-tomentum

which varies in quantity; stalk 2-3 mm. long. Aril thin, crimson,

fimbriate at the apex. Seed ellipsoid.

KepaH: Langkawi, Doliman F.D. 21488 (SING); Gunong Raya,

Langkawi, Wyatt-Smith K.F.N. 71228 (KEP); behind the sanatorium,

Wyatt-Smith K.F.N. 71200 (KEP); Gunong Lang, Kiah S.F.N.

35070 (A, BM, DD, K, KEP, SING); Gunong Baling, Kiah S.F.N.

35398 (A, K, KEP, SING); Jerai Reserve, Mat F.D. 17946 (KEP,

SING); Bukit Balut, F.D. 17885 (SING).

KELANTAN: Perangin, Kuala S. Lakah, Sokor, Wyatt-Smith K.F.N.

68301 (K, KEP); Bukit Batu Papan, S. Lebir, Henderson S.F.N. 29519

(K, SING).

PENANG: King 1372 (BM, CAL, K, L); Curtis Nos. 1044 (BM, CAL,

SING) and 2842 (K, SING); Waterfall Gardens, Haniff S.F.N. 9062

(SING); Moniot Road, Curtis 2457 (CAL, DD, SING); Pulau Boe-

tong, Curtis Nos. 1191 (BM, K, SING) and 2770 (CAL, K, SING).

PERAK: Scortechini Nos. 546 (CAL, FI, K, SING); 820 (CAL, DD,

K) and 831 (CAL, K, L); Larut, King Nos. 5092 (BM, CAL, FI, G,

K, L); 7452 (BM, CAL, FI, G, K, L, SING, UC); Gopeng, Kinta,

King 4307 (CAL, DD, K); Gunong Pondok, Henderson S.F.N. 23795

(SING); Waterloo, Curtis 2728 (CAL, SING).

TRENGGANU: Ulu Brang, Moysey and Kiah S.F.N. 33839 (SING);

Bukit Kajang, Kemaman, Corner 14th Nov., 1935 (SING).

PAHANG: Belengo F.R., Henderson F.M.S. Mus. Herb. 10767

(SING); Pulau Chengei, Ridley 2264 (CAL, K, SING); Tembeling,

Henderson S.F.N. 21810 (A, K, SING); Kemansul F.R., Temerloh,

Hamid F.D. 10584 (E, K).

SELANGOR: Sungei Lalang, Kajang, Symington Nos. 22775 (SING)

and F.D. 40627 (KEP); Weld Hill, F.D. 7866 (SING) and C.F. 815

= 98 (SING); Hamid C.F. 182 (SING); Omar F.D. 8529 (K, SING);

Ahmad C.F. 3024 (K, SING); Ridley, March 1915 (K); Ulu Gombak,

Hume F.M.S. Mus. Herb. 9929 (SING); Ginting Bidai, Ridley 7627

(CAL, SING); Sungei Lagong F.R., Wyatt-Smith K.F.N. 6607 (KEP).

NEGRI SEMBILAN: Ulu Bendol F.R., Holttum S.F.N. 9867 (K, SING);

Gunong Angsi, Nur S.F.N. 11668 (A, BRI, DD, SING, UC); Pasu

F.R., Kuala Pilah, Ja’amat 46051 (KEP); Sungei Menyala F.R.,

Wyatt-Smith K.F.N. 64095 (KEP).

Matacca: Maingay 1294 (A, CAL, K, L); Maingay 1288 (A) Kew

specimen is K. intermedia; Griffith, no data (CAL); Alvins 782

(SING); Sungei Ujong, Cantley 2261 (K); Bukit Sidenau, Derry 139

(SING); Bukit Panchur, Derry 485 (CAL, SING); Batu Tiga, Derry

979 (CAL, SING); Bukit Naning, Alvins 901 (SING); Sungei Udang,

Goodenough or Ridley 1800 (CAL, P, SING).

JoHorE: Gunong Panti, Corner, 10th March, 1935 (SING); Sungei

Kayu, Kiah S.F. Nos. 32334 (A, BM, BRI, DD, K, KEP, SING) and

32379 (A, BM, K, KEP, SING); Bukit Tinjau Laut, Corner S.F.N.

37078 (A, BRI, DD, K, KEP, SING); Ngadiman S.F.N. 36927 (BM,

BRI, KEP, SING); Sedili Kechil, Corner S.F.N. 28615 (A, K, SING).

SINGAPORE: Cantley 195 (K) type of K. cantleyi; Bukit Timah, Cant-

ley 3083 (K, SING); Sinclair S.F.N. 39475 (BO, DD, E, K, L, P, SAN,

SING); Ngadiman S.F. Nos. 36146 (A, BM, BRI, DD, KEP, SING)

and 34966 (A, K, SING); Ridley, date 1892 (SING); Lornie Road,

Reservoir Jungle, Corner S.F.N. 36951 (SING); Botanic Gardens

331

Gardens Bulletin, S.

Jungle near Potting Yard, Sinclair S.F.N. 39491 (BO, E, K, L, P,

SING); near the bandstand, Botanic Gardens, Ridley 2044 (BM, CAL,

K, SING); near the Botanic Gardens, King 8th Sept., 1879 (CAL).

DISTRIBUTION: Indo-China, S. Burma (Mergui), Nicobars, Siam

(Kerr 2120), Java, Sumatra, Pulau Berhala (Wyatt-Smith), Borneo.

‘Type: Blume, Java (L).

A variable, polymorphic species with a wide distribution. The

leaves may be acute, rounded or cordate at the base with a varying

amount of rather harsh, stellate indumentum. The leaves vary too,

in the number of veins and in width and in thickness; some are

thin while others are coriaceous. Sometimes the leaves may be

acute and rounded at the base on the same tree. Warburg divides

this species up into four forms and states that forma typica comes

from Java and Sumatra while he assigns the name malayana to the

Malay Peninsula material. However I have seen forms from the

Malay Peninsula agreeing exactly with his typica. The remainder

of the Malayan material (he does not specify a type), would be a

rather heterogeneous mixture and can hardly be classified as one

form, namely malayana. The Penang specimens have a cordate

leaf-base, the leaves are more coriaceous and the indumentum

thinner and might be given some other trinomial, but on account of

the numerous intermediates and Warburg’s vagueness as regards

malayana, \ do not see any usefulness at this stage in dividing this

‘species up into unsatisfactory forms.

The general characters of K. laurina are quite definite. It has been

confused with K. conferta in herbaria and superficially it closely

resembles this species although it does not fall in the conferta

group on account of the structure of the flower and the stamens.

It is more closely related to oblongifolia and globularia, as has

been pointed out under these species, but if sterile, the relation to

globularia will not be so apparent to a beginner who will tend to

look at the grosser morphological characters such as leaves, twigs

and fruit and will readily confuse it with conferta.

The leaves of conferta are more distinctly reticulate and more

coriaceous and the veins are raised above (sunk) in Jaurina);

both male and female flowers are stalked; the bracteole is median

on the pedicel and not at the base of the perianth as in Jaurina.

The anthers are more numerous and the stigma is sessile. The fruit

has less tomentum (but there are forms of laurina in which the

tomentum is thin) and is glabrescent when ripe and is usually

larger. Again some of the northern forms of /aurina have an equally

large fruit. Curtis 2770 is K. laurina and not conferta as stated by

Gamble, page 244 of the Materials.

I think K. elegans Warb., [Pierre 5432, holotype, Cambodia

{CAL, K, LE, P)] is only one of the large leaved forms of laurina.

332

Vol. XVI. (1958).

2. MYRISTICA Boehmer in Ludwig, Definitiones Generum

Plantarum (1760) 513; Houttuyn, Handleid. Hist. Nat. 2, 3

(1774) 333 genus conservatum (excl. M. sylvestris); Rottb.,

Acta. Univ. Hafn. 1 (1778) 281; L-f., Suppl. (1781) 40 and

265; Bl., Rumphia 1 (1836) 180 sect. Myristica; Hk. f. et Th.,

Fl. Ind. (1855) 162 sect. Eumyristica excl. M. superba et M.

horsfieldii, A. DC., Prodr. 14, 1 (1856) 189 sect. Eumyristica

excl. M. eugeniifolia et 192 sect. Caloneura; Mig., Fl. Ind. Bat.

1, 2 (1858) 53 pro majore parte et in Ann. Mus. Bot. Lugd.-

Bat. (1864) 205 pro majore parte; Hk. f. Fl. Br. Ind. 5 (1886)

102 sect. Eumyristica; King in Ann. Roy. Bot. Gard. Calc. 3

(1891) 286 sect. Eumyristica; Warb., Monog. Myrist. (1897)

374; Gamble, Mat. F.M.P. 5, 23 (1912) 226; Ridley, F.M.P. 3

(1924) 62; Markgraf in Engl. Bot. Jahrb. 67 (1935) 154;

Smith, Bull. Torr. Bot. Club 66 (1941) 397.

Synonym: Comacum Adans. Famil. 2 (1763) 345.

Trees 5—40 m. high. Dioecious (M. fragrans occasionally mo-

noecious and M. crassa paroecious as well as dioecious). Sztilt

roots sometimes present. Bark slightly rough, sometimes longitudi-

nally fissured, greyish brown to coal black; sap watery, red or pink.

Twigs striate, often shining, bark sometimes tending to crack, in

some New Guinea species 2 or 4 lines present extending from the

base of one petiole to the base of the one above, in such cases the

twigs are sometimes swollen, hollow and inhabited by ants. Leaves

membranous, chartaceous or coriaceous, usually shining above

when dry, often whitish or glaucous beneath, glabrous or with

some tomentum or scales beneath, base acute, rounded or less

often cordate, apex acute, acuminate or rarely obtuse, mesophyll

without reticulate sclerenchyma; nerves usually distinct, nearly al-

ways impressed above, rarely raised, prominent beneath and inter-

arching at the margin; reticulations few, usually scalariform or

forming a lax network, never dense as in Knema, best seen when

dry, more abundant than in Horsfieldia. Inflorescence a branched

panicle of varying lengths with the flowers in cymes or sub-umbels,

axillary or in the axils of fallen leaves, or a short woody knob un-

branched or branched with 2-3 short ramifications as in Knema,

the scars of fallen flowers very close to each other and distinct or

less often a combination of the two types when the basal part is

smooth and bears 1—3 short, woody, scarred knobs at its extremity;

bracts absent, bracteoles persisting, sub-orbicular and embracing

half the base of the flower. Flowers urceolate or campanulate, less

often tubular, pedicelled, often fragrant, glabrous, puberulous,

333

Gardens Bulletin, S.

pubescent or tomentose outside, glabrous inside, rusty-brown or

cream-coloured, 3-toothed (2—4 teeth rarer and abnormal) the

teeth reflexed; female flowers more swollen than the male; androe-

cium a stalked column to which the 8—30 linear anthers are com-

pletely fused by their backs and to each other by their edges, the

extreme apex of the column often sterile; ovary globose or sub-

globose, glabrous or with tomentum; style absent, stigmas connate,

minutely bilobed. Fruit usually large, 1-10 cm. long, globose, sub-

globose, ovoid, pyriform or oblong, orange, yellow or rusty brown,

glabrous, tomentose or less often lanose; pericarp thick, coriacecus,

fleshy but firm. Aril red, orange or yellow, much fimbriate at the

apex and cleft to the base or nearly to the base. Seed shining, large

and conform to the fruit, testa hard, brown or brownish black;

albumen ruminate with a fixed oil and starch; cotyledons connate

at the base and edges, divaricate.

TYPE OF GENUS: Myristica fragrans Houtt., Handleid. Hist. Nat. 2

(1774) 333.

DISTRIBUTION: South India, Ceylon, Burma, the Nicobars and An-

damans, Siam, Indo-China, Malaya, Malay Islands to North Australia,

Polynesia and Micronesia.

Myristica differs from the other Malayan genera in the structure

of its androecium (see page 29). From Knema and Horsfieldia it

differs in the aril being laciniate to the base or nearly to the base.

From Gymnacranthera it differs in its striate twigs, though these are

not always, as closely striate as in Knema and Horsfieldia. Starch is

present in the albumen; this is not found in Gymnacranthera and

Horsfieldia. It further differs from these two in the presence of a

bracteole at the base of the flower, half embracing it. The bract

is absent. From Knema it differs in the cotyledons being connate at

the base and in the fewer, laxer and less conspicuous reticulations

of the leaves. These reticulations often occur on both surfaces and

may form an open network or a scalariform one at right angles to

the main nerves. From Horsfieldia it differs in the leaves being less

brittle and more glossy above when dry and the under-surface

usually whitish or glaucous. One can usually distinguish the leaves

of the two genera after some initial practice. Those of Horsfieldia

sometimes have a parchment-like or granular appearance on the

upper surface when dry. Horsfieldia has fewer reticulations than

Myristica but both genera have them raised on the upper surface

of the leaf and sunk on the lower. The nerves, however are nearly

always sunk on the upper surface in Myristica but raised or sunk

in Horsfieldia. The leaves of H. wallichii and H. sylvestris may be

mistaken for those of a Myristica as they are somewhat similar.

Their reticulations are just as numerous as in Myristica and their

334

Vol. XVI. (1958). -

nerves are sunk above but the gloss of their dried leaves is less

striking.

There are more species in Myristica than in the other three

Malayan genera. New Guinea is the main centre of distribution of

Myristica. From what I have seen, there are still a few undescribed

species in collections from that island. There are two main types

of inflorescence, the branched panicle with a smooth axis, and the

contracted, woody, Knema-like tubercle with the numerous pedicel-

scars all over it. Both may be useful in classification or in the con-

struction of a key. Like Warburg, I -have used these inflorescence

characters in the key, but they are not entirely satisfactory and are

unhappily complicated by the fact that we get a type of inflores-

cence which is a combination of the two. In such a combination,

the main axis is smooth like that of the first but it very soon bears

one or more woody protuberances, covered with the Knema-like

pedicel-scars as in the second, e.g. M. ceylanica. Again the panicle-

type may be very short and thick and one might confuse it with the

second type but the axis is usually smooth without the scars. As

far as the Malayan species are concerned the matter is simple as

only one species has this Knema-like inflorescence, namely M.

crassa. In New Guinea there are many species with this type of

inflorescence. Of these M. hollrungii is the most striking. Its axis

may be several centimetres long and is usually 2-3-fid at the

apex. Warburg, in his key, groups his species into twenty-

one series, undoubtedly too many. He has successfully grouped

together most of the related species but there are a few which

should not be in his particular groups. He and others have had to

use, in their keys, the sizes of flowers, fruits, pedicels, bracts and

leaves as the chief diagnostic characters. The construction of a key

in the case of Myristica is not an easy matter on account of the

great number of the species and of the uniformity of the flowers.

The keys tend to be cumbersome and one may be side-tracked and

not arrive at the correct identity of a species. It. would be easier

and there would be less danger of going astray if one key was used

for each geographical area, country or island. The Malayan species

usually have large oblong or lanceolate leaves but there are several

species in New Guinea and the Eastern Moluccas with much

smaller leaves which are usually elliptic. M. fragrans, not native,

but now naturalized in Penang is a good example of this type.

Others with small, elliptic leaves are /epidota, tubiflora and globosa.

These might be grouped together, though not necessarily in one

series.

335

Gardens Bulletin, S.

REY Novd

a. Inflorescence a branched panicle or branched cyme with

flowers in umbellate or sub-umbellate fashion at the ends of

the branches, the axis often flattened, scarcely woody and

without the scars of fallen flowers conspicuous

b. Adult leaves very large, 10-16 cm. broad; nerves 23-30

pairs. Male inflorescence much branched, 10-18 cm. long.

Male pedicels 1—1-5 cm. long (1) M. maxima

b. Adult leaves less than 12 cm. broad; nerves less than 23

pairs. Male inflorescence less than 10 cm. long. Male pedi-

cels less than 1-1-5 cm. long except in M. fragrans

c. Inflorescence axis and flowers densely rusty-tomentose or

woolly-tomentose

d. Leaves glabrous beneath. Fruit tomentose or becoming

glabrous

e. Young twigs slender, 2-3 mm. thick at the apex, the

portion below the apical bud not rusty-tomentose

when young. Leaves 7-10 cm. long and 2-3-5 cm.

broad, spathulate, apex obtuse; nerves 12—18 pairs.

Male flowers 4-5 mm. long, bracteole 1 mm. long.

Fruit minutely tomentose (2) M. gigantea

e. Young twigs stout, 5-6 mm. thick at the apex, the

portion below the apical bud rusty-tomentose (best

seen when young leaves are present). Leaves 12—30

cm. long and 45-9 cm. broad, apex acute; nerves

12-30 pairs. Male flowers 6 mm. long, bracteole 3-5

mm. long. Fruit woolly-tomentose, glabrous or be-

coming glabrous

f. Leaves when dry blackish brown and extremely

glossy above, deep rusty brown beneath; petioles

4—5 cm. long; the portion of the twig from the

apical bud downwards to a distance of 12 cm.

rusty-tomentose and striate. Fruit densely woolly-

tomentose (3) M. lowiana

f. Leaves when dry olive green above and less glossy

to dull, pale yellowish brown beneath; petioles

shorter, about 2:5 cm. long; the portion of the

twig from the apical bud downwards rusty-tomen-

tose and striate for a distance of about 3 cm only.

Fruit with reddish brown scurf, soon glabrous

(4) M. maingayi

336

Vol. XVI. (1958).

d. Leaves rusty or silvery-brown tomentulose beneath with

scales and hairs, the tomentum tending to rub off.

Fruit sub-globose, rusty-tomentose.

(5) M. guatteriifolia

c. Inflorescence axis glabrous, puberulous or pubescent

g. Leaves drying pale yellowish brown above, and whitish

or glaucous beneath, fragile, tending to break in

herbaria. Twigs pale straw-coloured. Inflorescence

branches often deflexed. Male flowers elongate, slightly

oblique. Fruit slightly gibbous, glabrous, apex turned

to one side (6) M. elliptica

g. Leaves not drying pale brown above or below, not

fragile. Twigs not pale straw-coloured. Inflorescence

branches not deflexed. Male flowers not oblique nor

elongate. Fruit not gibbous nor the apex turned to one

side

h. Bracteoles of flowers ciliate at the edges. Reticula-

tions of lower surface of leaf often distinct

(7) M. malaccensis

h, Bracteoles of flowers not ciliate at the edges; reticula-

tions of lower surface of leaf mostly invisible except

in M. fragrans

i. Leaves covered with cinnamon-brown or whitish-

brown scales beneath. Fruit covered with minute

rusty scales which tend to rub off

(8) M. cinnamomea

i. Leaves not covered with cinnamon-brown scales be-

neath. Fruit glabrous minutely scurfily pubescent

when young

j. Leaves broadly elliptic, 6-13 cm. long and 3-5-

6:5 cm. broad, the smallest in size of Malayan

species; nerves 8—11 pairs. Flowering pedicels

1-1-5 cm. long. Flowers up to 1 cm. long. Fruit

pyriform (9) M. fragrans

j. Leaves lanceolate, larger, 12—20 cm. long and 3-6

cm. broad; nerves 12-17 pairs. Flowering

pedicels 4-8 mm. long. Flowers 5—6 mm. long.

Fruit oblong (10) M. iners

a. Inflorescence axis a short, thick, woody, elongated tubercle,

conspicuous with numerous scars of fallen flowers as in

Knema, sometimes with 2—3 short branches (11) M. crassa

337

Gardens Bulletin, S.

KEY No. 2 (Sterile material)

a. Adult leaves very large, 10-16 cm. broad; nerves 23-30 pairs

(1) M. maxima

a. Adult leaves fess than 12 cm. broad; nerves less than 23 pairs

b. Terminal bud rusty-tomentose or woolly-tomentose —

c. Leaves glabrous beneath

d. Young twigs slender, 2-3 mm. thick at the apex, the

portion immediately below the apical bud not rusty-

tomentose when young. Leaves 7-10 cm. long and 2—

3-5 cm. broad; apex obtuse; nerves 12—18 pairs

7 (2) M. gigantea

d. Young twigs stout, 5—6 mm. thick at the apex, the por-

tion below the apical bud rusty-tomentose. Leaves 12—

30 cm. long and 4:5—9 cm. broad, apex acute; nerves

12-30 pairs.

e. Leaves when dry blackish brown and extremely glossy

above, deep rusty brown beneath; petioles 4-5 cm.

., long. The portion of. the twig from the apical bud

downwards to a distance of 12 cm. rusty-tomentose

and striate. Stilt roots present. (Fruit densely woolly-

tomentose) | (3) M. lowiana

e. Leaves when dry olive green above and less glossy to

dull, pale yellowish brown beneath; petioles shorter,

about 2:5'cm. long. The portion of the twig from

the apical bud downwards to a distance of about 3

cm. rusty-tomentose and striate. Stilt roots not pre-

sent. erat with reddish brown scurf, soon glabrous)

! (4) M. maingayi

c. Leaves rusty-tomentulose indivessinnes tomentulose be-

neath with scales and hairs, the tomentum thin and tend-

ing to rub off ) (5) M. guatteriifolia

b. Terminal bud glabrous, ‘diese or adpressed-pubescent

- not woolly-tomentose

f. Leaves cinnamon brown or whitish brown beneath due to

minute scales (8) M. cinnamonea

f. Leaves green, pale green or glaucous beneath, not cinnamon |

brown

g. Reticulations of leaves distinct beneath, lax, few

h. Leaves 15-25 cm. long and 4—9 cm. broad, oblong

or elliptic oblong, abruptly contracted just below the

middle; nerves 15—20 pairs. (Bracteoles of flowers

ciliate) (7) M. malaccensis

338

Vol. XVI. (1958).

h. Leaves 6-13 cm. long and 3-5—6-5 cm. broad, the

smallest in the Malayan species, elliptic or oblong-

lanceolate, not abruptly contracted just below the

middle; nerves 8—11 pairs. (Bracteoles of flowers not

ciliate) (9) M. fragrans

g. Reticulations indistinct or absent beneath

i. Twigs pale straw-coloured, slender, 3 mm. thick at the

apex. Leaves glaucous or pale yellowish to whitish

brown beneath when dry. Stilt roots present

(6) M. elliptica

i. Twigs not pale straw-coloured, stout nor slender at the

apex. Leaves not pale beneath when dry. Stilt roots

present or not

j. Nerves of leaf 15—22 pairs, prominent beneath.

Twigs very stout, 6 mm. thick at the apex. A few

weak stilt roots present (11) M. crassa

j. Nerves 12—17 pairs, rather fine at times, broken or

indistinct. Twigs stout or not. Stilt roots present

or not

k. Leaves lanceolate, 12-20 cm. long and 3-6 cm.

broad. Twigs slender at the apex, 3 mm. thick

there. Stilt roots sometimes present

(10) M. iners

k. Leaves oblong, elliptic-oblong or panduriform,

17-35 cm. long and 6-11 cm. broad. Twigs

stout, S—6 mm. thick at the apex. No stilt roots

(7) M. malaccensis

(1) M. maxima Warb., Monog. Myrist. (1897) 385; Gamble, Mat.

F.M.P. 5, 23 (1912) 228; Ridley F.M.P. 3 (1924) 63.

Synonym: M. bracteata King (non DC. = M. philippensis

Lamk) in Ann. Roy. Bot. Gard. Calc. 3 (1891) 286 Pl. 106 et

107.—Figs 20-21 Plate. III-IVA.

Tree 15-30 m. high, pyramidal in outline with numerous hori-

zontal and descending branches. Bark light reddish-brown, rough

with some flaking here and there but scarcely fissured, pinkish white

inside; wood white; sap red, copious, watery. Twigs glabrous, stout,

greyish brown, striate in the older parts. Leaves coriaceous, dark

green and slightly bullate above, pale whitish green beneath,

glabrous, apex rounded and then shortly and abruptly acuminate,

base rounded; midrib flat above, very prominent beneath; nerves

23-30 pairs, parallel, oblique, sunk above, raised beneath, inter-

arching near the margin; reticulations faint; length 25-40 cm.;

339

Gardens Bulletin, S.

(/|3

Fig. 20. Myristica maxima Warb.

A, Leafy twig with female inflorescence. B, Female inflorescence. C,

Female flower dissected to show ovary. A-C from female tree on

Lawn Z, Botanic Gardens, Singapore.

340

Po OVE? (1958).*.’

breadth 10-16 cm.; petiole 2—3 cm. long, stout, deeply channelled.

Male inflorescence an axillary panicle, 10-18 cm. long, the main

axis flattened, the branches opposite, bearing the flowers in sub-

umbellate racemes; pedicels slender, 1—-1:5 cm. long; bracteoles

sub-orbicular, 2mm. long, embracing half the base of the flower.

Male flowers 5—8 mm. long and 4 mm. broad, coriaceous, tomen-

tulose outside, glabrous inside, urceolate, the mouth with 3 trian-

gular, acute teeth; androecium cylindric, 4-5 mm. long on a short,

1-2 mm. long stalk; anthers 12—20, narrowly elongate. Female

inflorescence much shorter, stouter and contracted, 3—5 cm. long;

pedicels stout, 3 mm. long; bracteole 3 mm. long, obtuse, sub-

orbicular, embracing the base of the flower. Female flowers

fragrant, coriaceous, creamy yellow, urceolate, with 3 reflexed,

acute teeth which extend half way down, minutely puberulous out-

side, glabrous inside, 8-9 mm. long; ovary sub-globose, minutely

rusty-pubescent, 5 mm. long including the bilobed, sessile stigma.

Fruit oblong, blunt at the apex, slightly narrowed into the woody

stalk, minutely rusty-pubescent, much wrinkled on drying, 7—9 cm.

long, 3-5-5 cm. broad; pericarp thickly coriaceous. Aril thin,

fleshy, bright red, laciniate. Seed shining, filling the carpel.

PENANG: Government ‘Hill, Curtis 1497 (K, SING) type material;

Waterfall Garden, Sinclair S.F.N. 39338 (SING).

PERAK: Scortechini 1872 (BM, CAL, FI, K) type material; Larut,

King Nos. 5513 (CAL, E, FI, G, K, L, MEL, SING, UPS) and 6960

(CAL, K, P) both type material; Kinta, D.F.O., K.F.N. 54805 (KEP).

PAHANG: Temerloh, Awang Piah F.D. 40372 (KEP).

SELANGOR: Ulu Gombak F.R., Strugnell F.D. 27906 (KEP, SING);

10th mile Ulu Gombak, Strugnell F.D. 12132 (SING); Weld Hill,

Abdul Rahman C.F. 1802 (SING); Ginting Simpah, Nur S.F.N. 34307

(KEP, SING); along pipe line, F.R.I., Kepong, Wyatt-Smith, K.F.N.

66540 (KEP):

JOHORE: Mawai, Ngadiman S.F.N. 34740 (BM, DD, K, KEP,

SING).

SINGAPORE: Kurz, no data (CAL); Bukit Timah, Ridley 3363 (CAL,

K, SING); pipe line beyond No. 1 rifle range, Nee Soon, Seletar

Forest, Sinclair S.F.N. 40304 (E, K, L, SING).

DISTRIBUTION: Borneo, Beccari 1556 (FI, K, P) ‘type material.

Sumatra.

A fine tree, not likely to be confused with any Malayan Myristica.

It is, however, closest to M. philippensis but differs from it in the

smaller bracteoles the less hairy flowers and almost glabrous in-

florescence axis. There is a female tree, planted on Lawn Z in the

Singapore Botanic Gardens. This originally came from Gunong

Pulai in Johore. The female flowers, hitherto unknown, are describ-

ed here from it and its fruit was exhibited in the ““Ridleyana Exhi-

bition” held at the Gardens on 10th December, 1955 to mark the

100th birthday of Henry Nicholas Ridley.

341

Gardens Bulletin, =S.

umn. D, Fruit.

Fig. 21. Myristica maxima Warb.

A, Male inflorescence. B,, Male flower. C, Staminal col

E, Seed with aril. A-C from Ridley 3363. D-E from female tree

on Lawn Z, Botanic Gardens, Singapore.

342

Vol. XVI. (1958).

(2) M. gigantea King in Ann. Roy. Bot. Gard. Calc. 3 (1891)

288 Pl. 110; Warb., Monog. Myrist. (1897) 400; Gamble, Mat.

F.M.P. 5, 23 (1912) 229; Ridley, F.M.P. 3 (1924) 64.—Fig.

22.

Tree 30—40 m. high. Twigs rusty puberulous at the apex, lower

down glabrous, dark reddish black with shining bark which is angled

and fissured and tends to separate from the wood. Leaves coriace-

ous, shining above, sub-glaucous beneath, narrowly oblong-lanceo-

late or lanceolate, margins revolute, base acute, apex obtuse; midrib

lying in a furrow above, raised beneath; nerves 12—18 pairs, oblique,

depressed above, raised beneath, the line of anastomosing broken

or indistinct; reticulations faint or invisible; length 7-10 cm.;

breadth 2—3-5 cm.; petiole 1-8 cm. long, slender, glabrous. Male in-

florescence an axillary panicle about 2:5 cm. long, rusty-tomentu-

lose, the branches opposite and short, 4-5 mm. long with pedicels

1—2 mm. long. Male flowers 4-5 mm. long with a minute, 1 mm.

long, orbicular, basal bracteole embracing half of the flower;

perianth rusty-tomentulose, ovoid with short, blunt or sub-acute

teeth; androecium about half the length of the perianth, the stalk

short, pubescent, the column cylindric, blunt at the apex, the

anthers linear, about 10, covering the top of the column. Female

flowers unknown. Fruit ovoid, narrowed towards the apex, 5-5 cm.

long and 4 cm. broad, minutely rusty-tomentose; stalk short, thick

0-5—2-5 cm. long. Seed shining, the aril thin, extending to the apex

of the seed, fimbriate nearly to the base.

PERAK: Scortechini 1949 (BM, CAL, FI, G, K, L) type material;

and Scortechini s.n. (BM, CAL, G, L, SING); Gopeng, King Nos. 5866

(CAL, K, MEL, SING, UPS) and 6050 (CAL, E, FI, K, KEP, P,

SING) both type material; Kampong Sayong F.R., Kinta, Amat K.F.N.

65903 (KEP); Bintang Hijau, Kuala Kangsar, Abdul Majid K.F.N.

68860 (KEP); Parit F.R., Kinta, Abidi F.D. 54697 (KEP) and K.F.N.

54653 (KEP); Chikus F.R., Batang Padang, Paduka bin Abdulla

K.F.N. 65054 (KEP).

TRENGGANU: Pulau Tenggol, Dungun, Kochummen K.F.N. 80585

(KEP, SING).

PAHANG: Bukit Beserah F.R., Wyatt-Smith K.F.N. 68679 (KEP).

SELANGOR: Rantau Panjang F.R., Flemmich 24917 (SING); Ayer

Ham F.R., Ngah F.D. Nos. 41778 (KEP, SING); 41784 (KEP, SING)

and 41779 (KEP, SING); Bangi F.R., Ulu Langat, Salleh bin Isa

K.F.N. 63609 (KEP); Bukit Tunggul F.R., Ulu Langat, Bidin K.F.N.

53659 (KEP); Bukit Cheraka, Ulu Langat, K.F.N. 71373 (KEP);

Subang, Bukit Cheraka F.R., Md Jaya K.F.N. 53688 (KEP); 15th

mile Ginting Simpah, Strugnell F.D. 12117 (SING).

NEGRI SEMBILAN: Sendayan, Yakim C.F. 505 (K, SING); Sungei

Menyala F.R., Jidin F.D. Nos. 18409 (KEP) and 18410 (KEP).

DISTRIBUTION: Sumatra, Borneo and Labuan. Motley 145 from

Labuan is this species and not M. maingayi (Warburg page 400).

343

Gardens Bulletin, S.

5 cm

JRA DEL

Fig. 22. Myristica gigantea King.

A, Leafy twig with male inflorescence. B, Male inflorescence. C, Male

flower. D, Staminal column. E, Fruit. A from King 6050, B—D from

Yakim C.F. 505. E from Meijer 4221.

344

Vol. XVI. (1958).

This is the tallest Malayan Myristica. Female flowers have never

been collected. The tree is distinguished from M. maingayi by its

greater height, smaller leaves with obtuse apices, smaller flowers

and bracteoles and by the minutely rusty-tomentose fruits. (Those

in M. maingayi are covered with some rusty scurf when young only).

It is also distinguished from M. maingayi by its twigs, the apical

portions of which are much more slender, 2—3 mm. thick as against

the 5—6 mm. thick ones in maingayi; the terminal bud is rusty-

tomentose but very slender and the portion of twig below it glabrous

or almost so. In maingayi the terminal bud is much stouter and

densely rusty-tomentose to almost lanose while a portion of the twig

up to 3 cm. below the bud is densely rusty-tomentose, seen best

after the formation of new leaves.

(3) M. lowiana King in Ann. Roy. Bot. Gard. Calc. 3 (1891)

293 Pl. 120 Figs 2, 3, 4 tantum. [Figs 1, 5, 6, 1 = M. crassa};

Warb., Monog. Myrist. (1897) 496; Gamble, Mat. F.M.P. 5,

23 (1912) 235; Ridley, F.M.P. 3 (1924) 66. — Fig. 23 Plate

IVB.

Tree 10—25 m. high with a few stilt roots. Bark coal black with

narrow longitudinal fissures, hard, brittle; inner soft, pink, slightly

laminated; sap watery, red; sapwood soft, white. Twigs stout, rusty-

tomentose from the tips down to a distance of 12 cm. (this portion

becoming striate on drying), below this glabrous and the bark be-

coming rough and tending to crack; apical bud elongate, rusty-

tomentose with 1 mm. long hairs. Leaves stoutly coriaceous when

fresh, dark green and dull above, paler green beneath with the

midrib and nerves paler and often yellowish green beneath, when

dry the blade dark blackish brown and extremely glossy above, deep

rusty brown beneath, glabrous, lanceolate to oblong-lanceolate with

slightly recurved margins, acute at the apex and base; midrib in a

furrow flush with the upper surface, very stout beneath; nerves 17—

20 pairs, impressed above, prominent beneath, the line of interar-

ching faint or invisible; reticulations faint above, fainter or invisible

beneath; length 19-30 cm., breadth 5—9 cm.; petiole 3-4-5 cm.

long, tomentose when young, becoming glabrous. Male inflores-

cence axillary, horizontal or pendulous, 3—5 cm. long, bearing 2—3

short branches with the flowers fascicled at their ends, rusty-

stellate-tomentose as are the 4-5 mm. long pedicels and the

3 mm. long, obtuse bracteoles. Male flowers scented when

crushed, rusty-tomentose outside, glabrous inside, 5-6 mm.

long, the three teeth cream-coloured, slightly reflexed, acute,

extending down about 4 length of the perianth; androecium

345

Gardens Bulletin, S.

Fig. 23. Myristica lowiana King.

A, Leafy twig with male inflorescences. B, Male cal, C, Staminal

column. D, Fruit. E, Seed with aril. pi. from ‘Sinclair S.F.N.

40319. D from Nur S.F.N. 34001.

Ee a oe

346

Vol. XVI. (1958).

4 mm. long, the basal, 1-2 mm. long portion ‘sterile and rusty-

tomentose, the upper with about 14 anthers, the extremity ending

in a sterile, sub-obtuse point. Female flowers unknown. Fruit

6-8 cm. long and 4 cm. broad, ovoid, pointed at both ends,

the base gradually narrowed into the short, stout, 5-8 mm. long

stalk, densely covered with rusty-woolly tomentum (1 mm. long

hairs) which tends to rub off with age; pericarp 8 mm.—1 cm. thick.

Aril laciniate nearly to the base, the fimbriations numerous and

narrow at the apex. Seed 4-5 cm. long, smooth, shining.

PERAK: Port Weld, Scortechini 1855 (K); Scortechini 1851 (CAL,

K lectotype).

PAHANG: Kuantan, Soh F.D. 15130 (K, SING).

SELANGOR: Sungei Tinggi, Kuala Selangor, Nur S.F.N. 34124 (BM,

K, L, S, SING); Telok F.R., Klang, Hamid F.D. 46212 (KEP) and

Aziz F.D. 46238 (KEP); Bukit Changgong Sawmill, Klang, Nur

S.F.N. 34001 (K, KEP, S, SING); Bukit Badong, Kepong, Jantan F.D.

26347 (KEP, SING); Cheraka F.R., Ulu Selangor, Lelah bin Khamis

K.F.N. 65569 (KEP).

MaALacca: Merlimau, Ridley or Goodenough 1621 (CAL, SING).

JOHORE: Bukit Piatu, Sungei Sedili, Ngadiman S.F.N. 36880 (BM,

BRI, K, KEP, SING); Muar, Ridley 3755 (CAL, SING); Pengkalan

Raja, Pontian, Ngadiman S.F.N. 36622 (K, KEP, SING) and Corner

& Henderson S.F.N. 36603 (BRI, K, KEP, SING); Pontian, Abdulla

bin Md. Sharif K.F.N. 69957 (KEP) and Bidin bin Samad K.F.N.

70260 (K, KEP); Benut, Ibrahim bin Hussin K.F.N. 72752 (KEP).

SINGAPORE: Jurong, Corner S.F.N. 28131 (BM, BRI, DD, K, KEP,

SING) and 19th Oct., 1932 (SING); Kranji, Ridley 6451 (CAL, K,

SING); cultivated, Arboretum, Botanic Gardens, Sinclair S.F.N. 40319

(BM, BO, E, K, L, P, SING).

DISTRIBUTION: Sarawak, Borneo, Sumatra and Banka.

This Myristica is an inhabitant of the peat forest. If growing in

drier situations it will still develop a few weak stilt roots. It is dis-

tinguished from the others by its woolly fruits. Sterile material is

easily confused with M. maingayi but the leaves when dry are

blackish brown, extremely glossy above and deep rusty-brown

beneath whereas in maingayi they dry olive green, are much less

glossy above and the lower surface is pale yellowish brown. The

petioles are longer, 3—5 cm. long as against 2:5 cm. in maingayi.

Further the portion of the twig from the apex downwards to a dis-

tance of 12 cm. is rusty-tomentose. The tomentum in maingayi

extends only some 3cm. downwards. Female flowers have not been

seen.

King Nos. 5537 and 7258, flowering material, belong to M. crassa

and have been excluded from the type. Scortechini 1851, fruiting

material, agrees with King’s description of M. lowiana and has been

made the lectotype of this species.

347

Gardens Bulletin, S.

(4) M. maingayi Hk. f., Fl. Br. Ind. 5 (1886) 104; King in Ann.

Roy. Bot. Gard. Calc. 3 (1891) 294 Pl. 114 excl. Curtis 2455;

Warb., Monog. Myrist. (1897) 398; Gamble, Mat. F.M.P. 5, 23

(1912) 228; Ridley, F.M.P. 3 (1924) 64.—Fig. 24. Plate. VB.

Tree up to 36 m. high. Bark much fissured longitudinally and

stratified, greyish black, brittle; wood reddish; sap pink, watery,

copious. Twigs with reddish brown, shining bark which tends to

crack, rusty-tomentose for a short distance, 3 cm. from apex down-

wards, best seen when young leaves appear; terminal bud rusty-

tomentose with adpressed, 1 mm. long hairs. Leaves coriaceous,

linear-oblong, dark green above with paler midrib and nerves, paler

green beneath, drying olive green above and yellowish brown

beneath, base acute, apex acute, edges slightly revolute when dry;

nerves about 20 pairs, not very straight, interarching rather indis-

tinctly at the margin; reticulations faint; length 12—28 cm.; breadth

4:5-8:5 cm.; petiole stout, channelled, 2—2:5 cm. long. Male

inflorescence solitary, axillary, drooping, branched, 10-16 cm.

long, rusty-tomentose, bearing the flowers in umbellate cymes;

pedicels 3 mm. long; bracteoles half-orbicular, 5 mm. long, embrac-

ing half base of the flower. Male flowers 4—6 mm. long, ovate,

rusty-tomentose, the teeth ovate, acute; androecium 5 mm. long,

fusiform, shortly apiculate on a short, rufous-villous stalk; anthers

7-8, linear. Female inflorescence stout, up to 3 cm. long with

2—3 branches, rusty-tomentose; pedicels stout, 4-5 mm. long;

bracteole 5 mm. long and 6 mm. broad. Female flowers

fragrant, ovoid, 9 mm. long with 3-4 acute lobes, rusty-

tomentose outside, creamy-yellow inside; ovary ovoid, rusty-

pubescent, 4-5 mm. long, tapering gradually into the bilobed

stigma. Fruit pale yellow, with some reddish-brown scurf, soon

becoming entirely glabrous, oblong-ovoid, 10:5 cm. long and

6—6:5 cm. broad, the groove of dehiscence prominent; pericarp

2—2:5 cm. thick; stalk stout, 2 cm. long. Aril at first orange-red,

then scarlet, split down # or more of its length several times, much

fimbriated at the apex. Seed 6 cm. long, blackish brown, shining,

smooth.

PERAK: Bukit Hijau F.R., Lenggon, A. Manap bin A. Rahman

K.F.N. 51208 (KEP).

TRENGGANU: Bukit Gemok Kijal, Kemaman, Forest Ranger F.D.

44164 (KEP); Bukit Bauk F.R., Dungun, Ahmad bin Ibrahim K.F.N.

53361 (KEP); Bukit Kajang, Kemaman, Corner, 14th Nov., 1935

(SING).

PAHANG: Bukit Beserah, F.R., Kuantan, Tambey Chik K.F.N. 65681

(KEP).

SELANGOR: Sungei Lalang F.R., Md. Yatim K.F. Nos. 74452 (KEP);

and 53610 (KEP); Bukit Lagong F.R., Compt. 17, Wyatt-Smith

K.F.N. 65531 (KEP).

348

—E —

Fig. 24. Myristica maingayi Hk. f.

A, Leafy twig with female inflorescences. B, Female flower. C, Female

flower dissected to show ovary. D, Fruit. E, Seed with aril. F, Sta-

minal column. A—E from Sinclair S.F.N. 40599. F from King’s

Plate 114.

Gardens Bulletin, S.

NEGRI SEMBILAN: Compt. 19 Serting Ext. Res., Kuala Pilah, Moha-

med bin Poz K.F.N. (KEP).

MALACCA: Maingay 1289 (CAL, K holotype, L).°

JoHoRE: Bukit Naga Mengalir, S. Sedili, Ngadiman S.F.N. 36900

(SING); 73 miles Mersing-Endau Road, Mersing, A. Latiff bin Domal,

K.F.N. 70063 (KEP); 6th mile Tanjong Labok, Batu Pahat, Suleiman

bin Manja K.F.N. 70179 (KEP); Renggam F:R., Cousens K.F.N.

69783 (KEP); Burgess K.F.N. 69909 (KEP) and Wyatt-Smith K.F.N.

71271, CREP).

SINGAPORE: MacRitchie Raaon Corner SEN. 83555 (SING);

‘Reservoir Jungle, Corner, 20th July, 1932 (SING); cultivated in Arbo-

retum, Botanic Gardens, Henderson S.F.N. 37918 (BM, L, SING);

Sinclair, 21st Jan., 1953-(E,-K, L).and- Sinclair S.F.N. 40599 (BM,

BKF, BO, E, K, L, P, SAN, SING). ;

DISTRIBUTION: Confined to Malaya. .

The bark of this species and of M. gigantea, iners and lowiana is

of a similar type, brittle, blackish-or greyish black with numerous

longitudinal fissures and in this respect, if one is familiar with its

appearance, there should be no difficulty in separating this group

from the other Malayan -species.. The individual species of this

group, however, cannot be distinguished from each other by the

bark characters alone which are so similar. M. maingayi, when

sterile may be readily confused with any of these species. Its leaves

are variable in size and its fruit is very similar to that of iners. King

confused Penang specimens of iners, Curtis 2455, with it. From

iners. it is distinguished by the more coriaceous leaves with the

veins more distinct on the lower surface, the thicker twigs and the

larger, tomentose terminal-bud. From lowiana it is distinguished by

the absence of stilt roots, the shorter hairs of the terminal bud, the

apical portions of the twigs less tomentose or very soon becoming

glabrous, the smaller leaves with shorter petioles, the different

colour of the leaves on drying and the glabrous fruit. From

gigantea it is distinguished by the larger leaves which are not or

scarcely obtuse at the apex, the stouter twigs and thicker terminal

bud, the larger flowers and the glabrous fruit. For other differences

see notes under these species.

(5) M. guatteriifolia A. DC., in Ann. Sc. Nat. 4, 4 (1855) 20 T.

4 et Prodr. 14, 1 (1856) 193; Miq., Fl. Ind. Bat. 1, 2 (1858)

61; Rolfe in Journ. Linn. Soc. 21 (1888) 298; Vidal, Revis.

Pl. Vasc. Filip. (1886) 220; Warb., Monog. Myrist. (1897)

412 T. 13 Figs 1-4; Gamble, Mat. F.M.P. 5, 23 (1912) 235;

Ridley, F.M.P. 3 (1924) 63; Corner, Wayside Trees of Malaya

1 (1940) 478.

Synonyms: M. cookii Warb., Monog. Myrist. (1897) 414 T.

15 Figs 1-5. M. littoralis Miq., Fl. Ind. Bat. 1, 2 (1858) 37;

350

Vol. XVI. (1958).

Koorders en Valeton in Mededeel. Lands. Plantentiun 4 (1896)

173; Warb. Monog. Myrist. (1897) 418. M. riedelii Warb.,

Monog, Myrist. (1897) 417 T. 15 Figs 1-2.—Fig. 25.

Tree 13—20 m. high. Bark a warm brown to greyish brown, not

fissured but flaky in rather small irregular pieces; sap abundant,

dark red. Twigs densely rusty-tomentose at the apex, slowly becom-

ing glabrous further back, reddish and striate. Leaves coriaceous,

glabrous above, thinly tomentose beneath to nearly glabrous with a

rusty-stellate whitish or brown scaly tomentum, ellipitic or elliptic-

oblong, acute at the base and apex; nerves 15-20 pairs, distinct

on both surfaces, oblique and nearly parallel, the line of interar-

ching faint; reticulations faint on both surfaces, not always visible

if the tomentum is thick; length 15—30 cm.; breadth 3-12 cm.;

petiole 2—2:5 cm. long, rusty tomentose, becoming glabrous. Male

inflorescence axillary (all parts including the flowers, pedicels and

bracteoles densely rusty-tomentose) variable in length and in the

number of branches, 2—8 cm. long, branches 5 mm.—1 cm. long;

pedicels 5 mm. long; bracteoles 3 mm. long, obtuse, caducous, Male

flowers 4—7 mm. long, glabrous inside, split down to about 1/3 by

3 reflexed teeth; androecium 4 mm. long with a rather truncate

apex, stalk tomentose, 1 mm long; anthers about 10. Female flowers

5S—7 mm. long on 4 mm. long pedicels, more coriaceous and swollen

than in the male; ovary sub-globose, rusty-tomentose, 4 mm. long

with a 2-lipped stigma. Fruit rusty-tomentosé, 5 cm. long and 4 cm.

broad; stalk 1:5—1-8 long; pericarp 7 mm. thick. Aril fimbriate

nearly to the base, the fimbriations narrow and numerous at the

apex. Seed dark brown, smooth, shining, 3-8 cm. long.

TRENGGANU: Pulau Tenggol, Dungun, Kochummen K.F.N. 80586

(KEP, SING).

PAHANG: Pulau Chibeh, P. Tioman, Corner S.F.N. 29841 (K, SING).

JoHoRE: Pulau Selindan, Corner, 30th Jan., 1937 (SING); S. Rhu

Reba, Jason Bay, Corner S.F.N. 28512 (SING); Telok Lundang, Ba-

kau, 10th miles south of Jason Bay, Corner, 20th June, 1934 (SING).

SINGAPORE: Cultivated, Botanic Gardens’ Potting Yard Corner S.F.N.

28133 (CAL, K, NY, SING) and Nov. 1934 (K, SING); Sinclair, 24th

Feb., 1953 (DD, E, L, P, PNH, SAN); Botanic Gardens, Palm Valley,

Furtado, 6th Jan., 1928 (SING); ditto Burkill, 10th Sept., 1921

(SING); Ridley Nos. 2040 (CAL, SING) and 3705 (BM, SING);

Botanic Gardens, Aroid Rockery, Furtado, 6th Jan., 1928 (SING);

Fort Canning, Ridley 10922 (CAL, K, SING).

DISTRIBUTION: Cochinchina, North Borneo, Labuan, Philippines,

Billiton, Java.

Tyre: Cuming 1582 holotype of M. guatteriifolia in G. Isotypes in

BM. K, LE, M. P.

Native on the rocky east coast of Malaya. A somewhat poly-

morphic species, not likely to be confused with any other Malayan

351

Gardens Bulletin, S.

Fig. 25. Myristica guatteriifolia A. DC.

A, Leafy twig with female inflorescence. B, Female flower. C, Fruit.

D, Seed with aril. E, Portion of male inflorescence. F, Male flower.

G, Staminal column. H, Scales from undersurface of leaf. A, B, H

from tree in Potting Yard, Botanic Gardens, Singapore. C-D from

Corner 20-6-1934, Jason Bay. E-G from Cuming 1582.

352

Vol. XVI. (1958).

Myristica. The leaves are very variable in length and breadth as

well as in the amount of tomentum. Some of the Philippine material

has almost glabrous leaves. The inflorescence too, varies in length

and in the amount of branching. Philippine material tends to have

the longest inflorescence while the Javan (M. littoralis) is the least

branched. The flowers vary slightly in size but I cannot admit such

variable forms to pass as different species. I have to reduce M.

cookii from Pulau Condor to guatteriijolia as it agrees exactly with

the Malayan material and so does M. riedelii from Billiton.

(6) M. elliptica Hk. f. et Th., Fl. Ind. 1 (1855) 162; A. DC., 14

1 (1856) 190; Mig., FL Ind. Bat. 1, 2 (1858) 58; Hk. £, FL.

Br. Ind. 5 (1886) 102; King in Ann. Roy. Bot. Gard. Cal-. 3

(1891) 295 Pl 113; Warb., Monog. Myrist. (1897) 435 T. 16;

Gamble, Mat. F.M.P. 5, 23 (1912) 231; Ridley, F.M-P. 3

(1924) 65; Corner, Wayside Trees of Malaya 1 (1940) 477

Text Figs 195, 161.

Synonyms: M. calocarpa Migq., Fl. Ind. Bat. 1, 2 (1858) 71

et in Ann. Mus. Bot. Lugd.-Bat. 2 (1865) 48. M. sycocarpa

Mig., FL Ind. Bat. 1, 2 (1858) 68 et in Ann. Mus. Bot. Lugd-

Bat. 2 (1865) 48.

var. elliptica—Fig. 26. Plate VIA-B.

Tree 8-33 m. high; stilt roots present when growing in wet

ground. Bark reddish brown, slightly rough but not fissured, of no

great depth; sap red, copious. Twigs hanging down, glabrous, paic,

their bark whitish; terminal bud 2 cm. long, narrow, slender,

puberulous. Leaves chartaceous, rather brittle when dry, glabrous.

dark green and shining above, drying pale yellow, paler or glaucous

beneath, elliptic or elliptic-oblong, base acute or rounded and then

acute, apex acute; nerves 12—17 pairs, curving, slender, distinct,

slightly raised above, somewhat decurrent on to the midrib above,

the line of their anastomosing broken or indistinct; reticulations in-

distinct, not visible in dried material; average length 12-18 cm.;

average breadth 4-8 cm. but 28 cm. long and 12 cm. broad in

sapling leaves in shade; petiole 1-3 cm. long, stout, 4-sided. Male

inflorescence axillary, short, 2—2-5 cm. long, little branched, the

flowers in sub-umbellate clusters; pedicels 5 mm. long; bracteoles

sub-orbicular, 1 mm. long. Male flowers cream-coloured, 8—9 mm.

long, glabrous or adpressed-puberulous, elongate and narrowly

tubular, sub-urceolate, 3-toothed, the teeth 3 mm. long: staminal

column 7 mm. long on a very short, glabrous stalk, the apex

bluntly apiculate, anthers about 10. Female inflorescence shorter

and less branched. Female flowers urceolate, about the same

353

Gardens Bulletin, S.

length as the male but broader and more swollen, the teeth

reflexed and the bracteoles 1-2 mm. long, ovary 5 mm. long,

strigose, narrowed gradually into the 2-cleft, sessile stigma. Fruit

7-8 cm. long and 3-5-4 cm. broad, glabrous, orange-yellow, oblong,

Fig. 26. Myristica elliptica Hk. f. et Th. var. elliptica.

A, Leafy twig with female inflorescences. B, Female flower. C, Female

flower dissected to show ovary. D, Fruit. E, Seed with aril. F, Male

flower. G, Staminal column. A—C from Sinclair S.F.N. 39661. D-E

from Sinclair S.F.N. 39503. F—G from Corner S.F.N. 30243.

5 : »'s

aEnsiatetediinmeeh thane aden hediaetieitinns:odcanaees a eee

354

Vol. XVI. (1958).

slightly gibbous, the apex bluntly narrowed and turned to one side,

ridged along the sutures; pericarp 1 cm. thick; stalk 2 cm. long.

Aril scarlet, fleshy, narrowly fimbriate with numerous segments,

cleft down 1-1-5 cm. from the base. Seed olive brown, 5 cm. long.

LOWER SIAM: Chawng, Kiah S.F.N. 24381 (K, SING).

KEDAH: Enggong F.R., 5th mile Sik, Rose K.F.N. 73782 (KEP).

PENANG: Porter, Wall. Cat. 6798a (BM, E, G, K holotype, M, NY);

King 5198 (CAL, DD, FI, G, L); Government Hill and Penara Bukit,

Curtis 1122 (K, SING); Government Hill, Burkill S.F.N. 1536 (BM,

CAL, SING); Sungei Pinang, Nauen S.F.N. 35852 (BRI, SING); sine

loc., Nauen S.F.N. 38091 (BM, BRI, K, KEP, SING); Mount Olivia,

Haniff, April 1901 (SING).

PERAK: Scortechini 79b (CAL, DD, FI, G, K) and 1964 (CAL, DD,

E, K); Tupai, Wray 2345 (CAL, SING); Waterfall Hill, Wray 1736

(CAL, SING); Gopeng. Kinta, King Nos. 4276 (BM, CAL, MEL);

4426 (CAL, DD, FI, P) and 4703 (CAL, FI, K, L); Ulu Bubong,

King Nos. 10321 (CAL, FI, K) and 10583 (CAL, DD, E, UPS);

Larut, King Nos. 3732 (CAL, DD, FI) and 5288 (CAL, FI); Bikum,

Sungei Perak, Burn-Murdoch 373 (K); Dindings, Mat Hassan K.F.N.

69417 (KEP).

TRENGGANU: Ulu Kajang, Kemaman, Corner S.F.N. 30489 (DD, K,

KEP, SING); Ulu Ayam swamp, Bukit Kajang, Corner S.F.N. 30243

(BRI, DD, KEP, SING); 21st mile Kuala Trengganu-Besut Road, Sin-

clair & Kiah S.F.N. 40480 (BM, BO, E, K, L, P, SING).

PAHANG: Kemansul Reserve, Temerloh, Hamid F.D. Nos. 10888 (K,

SING) and 10693 (KEP); Bruas Valley, Bentong, Burkill & Haniff

S.F.N. 16448 (SING).

SELANGOR: Ulu Selangor, King 8559 (BM, CAL, DD, K); 12th mile

Ginting Simpah, Strugnell F.D. 12391 (KEP); 15th mile Kanching Re-

serve, Strugnell F.D. 13953 (KEP).

NEGRI SEMBILAN: Sendayan F.R., Md. Yakim C.F. 515 (SING).

Matacca: Maingay 1296 (CAL, K, L); Selandor, Alvins 262 (SING)

and 530 (SING); Batang Malaka, Derry 909 (SING) Knema glauces-

cens is mounted on the same sheet; Bukit Sedanan, Goodenough 1471

(CAL, SING).

JoHorE: Sungei Pao, Fox 11275 (CAL, SING); Macao Kankor,

Kuala Tebing Tinggi, Ridley, no number nor date (SING); Temeo River,

Kota Tinggi, Ridley 15344 (K, SING); Gunong Pulai, Mat, date 1892

(SING); 16th mile Gunong Pulai Road, Best S.F.N. 8292 (K, SING);

Sungei Segum, Gunong Panti, Mawai, Corner S.F.N. 29402 (SING);

Mawai, Ngadiman S.F.N. 34729 (K, KEP, SING); Mawai, Sungei

Sedili, Corner S.F.N. 28322 (K, NY, SING); S. Sedili, Corner S.F.

Nos. 26055 (K, NY, SING) and 25954 (K, SING); Kangka, Sedili

Kechil, Corner S.F.N. 28584 (K, SING); Castlewood, Ridley, date

1909 (SING); Batu Pahat, Ridley, Nov. 1900 (SING); S. Kayu, Kiah

S.F.N. 32004 (BRI, DD, K, KEP, SING); Mersing Road, Holttum

S.F.N. 36392 (KEP, SING).

SINGAPORE: Bukit Timah, Ngadiman S.F. Nos. 34952 (CAL, K,

SING) and 35923 (K, SING); Ridley 6920 (CAL, SING) and 28th

Jan., 1891 (SING); Bukit Mandai, Goodenough 1632 (SING); Corner,

24th July, 1940 (SING); Ridley, date 1892 (SING) and Ridley, date

1893 (SING); Chan Chu Kang, Ridley 331 (CAL, SING) and 274

(SING); Seletar Reservoir, Corner S.F.N. 37275 (BM, BRI, K, SING);

Sinclair, 7th Jan., 1950 (DD, E, L); viaduct, north side of Peirce Re-

servoir, Sinclair S.F.N. 89661 (B, BK, BO, DD, Delhi Univ., E, K, )

Gardens Bulletin, S.

P, PNH, SAN, SING); 123 miles Mandai Road, Sinclair S.F.N. 39503

(B, BE M, SING).

DISTRIBUTION: Siam, Borneo, Sumatra, Billiton.

The Swamp Nutmeg, a common species in Johore and Singapore,

is perhaps one of the most easily distinguished Malayan species.

The flowers are fragrant when crushed. The fruit is a beautiful

object, orange red with a scarlet aril. I here reduce Myristica

celebica Miq. to a variety of M. elliptica, the only important differ-

ences being the more tomentose or pilose flowers and the stalk of

the staminal column also pilose. M. simiarum from the Philippines

is also merely a variety of elliptica but the differences here are more

considerable. The inflorescence is much more branched and the

fruit smaller.

(1) M. elliptica Hk. f. et Th. var. celebica (Migq.) J. Sinclair, stat. nov.

Basionym: M. celebica Miq., Ann. Mus. Bot. Lugd.-Bat. 2

(1865) 47.

(2) M. elliptica Hk. f. et Th. var. simiarum (A. DC.) J. Sinclair, stat.

nov.

Basionym: M. simiarum A. DC. in Ann. Sci. Nat. 4,4

(1855) 29; Prodr., 14,1 (1856) 192.

(7) M. malaccensis Hk. f., Fl. Br. Ind. 5 (1886) 104; King in

Ann. Roy. Bot. Gard. Calc. 3 (1891) 287 Pl. 107 bis; Warb.

Monog. Myrist. (1897) 411; Gamble, Mat. F.M.P. 5, 23 (1912)

230; Ridley, F.M.P. 3 (1924) 64. non M. malaccensis Gando-

ger — Ardisia teijsmanniana Scheft.

Synonyms: M. pandurifolia H. Winkler in Engl. Bot. Jahrbuch.

49 (1913) 367. M. borneensis Warb., Monog. Myrist. (1897)

401 T. 14 Figs. 1-3. non M. borneensis Gandoger = M. villosa

Warb. M. wyatt-smithii Airy-Shaw in Kew Bull. (1948) 251.—

Fig. 27.

Tree 5-13 m. high. Twigs rather pale, angled, glabrous. Leaves

oblong or elliptic-oblong, widest at the middle and abruptly con-

tracted just below the widest part, glabrous, shining above, slightly

glaucous beneath; nerves 15—20 pairs, impressed above, raised and

distinct beneath, the interarching at the margin very distinct; reti-

culations visible above, more distinct beneath, forming a loose

network; length 15—25 cm.; breadth 4—9 cm., petiole 1:5 cm. long.

Male inflorescence an axillary panicle, 7-10 cm. long, bearing the

flowers in sub-umbellate cymes; bracteoles 1 mm. long, reniform,

membranous, glabrous with ciliate margins. Male flowers 2:5 mm.

long, sub-globose, coriaceous, glabrous with three blunt teeth;

androecium nearly sessile, 1:5 mm. long, anthers 7 with sub-acute

tips. Female inflorescence very short, 7 mm.—2 cm. long, little

branched. Female flowers and bracteoles as in the male but the

356

Vol. XVI. (1958).

Fig. 27. Myristica malaccensis Hk. f.

A, Leafy twig with fruit. B, Female inflorescence. C, Female flower.

D, Ovary. E, Male inflorescence. F, Male flower. G, Staminal

column. H, Ciliate bracteole. A from Alvins 2071. B—D from Ahmad

5081. E-H from King’s Plate 107 bis.

357

Gardens Bulletin, S.

perianth a little longer, 3-4 mm. long; ovary glabrous with sessile

stigma. Fruit oblong, glabrous, 5 cm. long and 2:8 cm. broad on a

rather slender, 4 cm. long stalk. Aril slightly fimbriate with gaps

down to the middle. Seed 4-3 cm. long, dark brown when dry.

KEDAH: Bukit Tanjong, Symington F.D. 57034 (KEP).

PERAK: Ulu Bubong, King 10749 (CAL, DD).

PAHANG: Bentong, Ahmad C.F. 5081 (K, SING).

SELANGOR: Sungei Buloh Reserve, Kiai F.D. 8263 (SING); Bukit

Lagong Forest Reserve, Wyatt-Smith K.F.N. 63176 (K, KEP, SING);

Wyatt-Smith and Sow K.F.N. 52149 (K) holotype of M. wyatt-smithii;

Sow K.F.N. 64427 (K).

Matacca: Maingay 1305 (CAL, K holotype, L); Sungei Ujong,

Alvins 2071 (SING) and 2074 (SING).

DISTRIBUTION: Sumatra, Boschpr. E. 883 (L). Borneo, Winkler 2405

(BRSL, L, SING) isotypes of M. pandurifolia. Beccari Nos. 652 (FI,

G, K, M, P); 666 (FI, G, K, M, P); 1270 (FI); 1574 (FI, G, K, P);

1575 (FI) and 2328 (FI) all syntypes of M. borneensis.

A rare and little collected species. Bark characters and colour

notes on leaves, flowers, fruit aril. and seeds are wanted. It is

nearest to M. malabarica but differs from it in its larger leaves and

in the androecium which in mialabarica has 10-15 anthers and ends

in a conical process. The best distinguishing guide to separate it

from the other Malayan species is the ciliate bracteole. The rather

pale, angled, glabrous twigs and the leaves widest at the middle

with rather numerous veins may be useful with sterile material.

M. pandurifolia and M. borneensis from Borneo are not different.

It is with great reluctance that I reduce M. wyatt-smithii to malac-

censis. I found that in wyatt-smithii that the bracteoles were also

ciliate and this led me to see the connection between these two

species. The leaves of wyatt-smithii show no distinct reticulations on

the lower surface while those of malaccensis usually have a few. I

do not now think that this character is of diagnostic value. It was

while at Leiden that I saw several sheets of a sterile Myristica from

Sumatra and could not decide whether to class it with M. wyatt-

smithii or malaccensis. I left them unnamed as I did not, at that

time, realize that the two species might be the same.

(8) M. cinnamomea King in Ann. Roy. Bot. Gard. Calc. 3 (1891)

292 Pl. 116; Warb., Monog. Myrist. (1897) 435; Gamble, Mat.

F.M.P. 5, 23 (1912) 232; Ridley, F.M.P. 3 (1924) 63; Corner,

Wayside Trees of Malaya 1 (1940) 477.—Fig. 28. Plate VII A.

Tree 18—24 m. high. Bark blackish-brown, finely rugose-fissured,

the outer bark hard, rather brittle and readily breaking off from the

pinkish inner-bark; sap copious, red. Twigs rather slender, dark

coloured, minutely rufous-puberulous at the tips, glabrous further

358

ny [ary ly .

Vol. XVI. (1958).

Fig. 28. Myristica cinnamomea King.

A, Leafy twig with male inflorescences. B, Male flower. C, Staminal

column. D, Fruit. E, Seed with aril. A from Goodenough 5572.

B-C from Corner S.F.N. 30388. D—-E from Corner 2-4-1934, S.

Sedili Kechil (spirit).

359

Gardens Bulletin, S.

back. Leaves thinly coriaceous, lanceolate to oblong-lanceolate,

dark green and shining above, uniformly covered with minute,

stellate, silvery-brown scales beneath; midrib in a furrow above,

raised beneath; nerves 14—20 pairs, slender, at times faint, reddish

beneath when dry, line of anastomosis indistinct, secondary nerves

often present between the primary, reticulations not visible; length

15-20 cm.; breadth 3-6-5 cm.; petiole 1:2—2:2 cm. long, slender,

minutely pubescent, becoming glabrous. Male inflorescence of

rufous-tomentose, axillary, umbellate cymes, the main axis 5 mm.—

1 cm. long with about 3 branches 5 mm.—1 cm. long; pedicels 5

mm. long with a minute, ovate, acute bracteole at the base of the .

flower. Male perianth brown, 6 mm. long, minutely pubescent out-_

side, glabrous inside, split down half-way by the 3, broadly

triangular reflexed, sub-acute teeth; androecitum 4 mm. long with

a pubescent, 2 mm. long stalk and about 10 anthers, the extreme

apex sterile and slightly obtuse. Female flowers not seen. Fruit

reddish yellow, minutely rusty-scaly, the scurf tending to rub off,

ellipsoid, broadest at the base and narrowed to the rounded apex,

6-9 cm. long and 4°5 cm. broad; pericarp 1-5 cm. thick; stalk stout,

6 mm.—1.2 cm. long. Aril cleft 8 mm.—1:5 cm. from the base, finely

fimbriate at the apex. Seed 5 cm. long with thin shining testa.

PERAK: Larut, King Nos. 3534 (CAL, DD, K); 5170 (CAL, K, L,

MEL, SING); 5355 (CAL, E, FI, K, P); 5458 (CAL, DD) Fi Gime

UPS); 6696 (CAL, DD, FI, G, K, L); 7474 (BM, CAL, FI, K, L, P);

Gopeng, King 1057 (CAL, E, G, K); Simpang, Wray 3026 (CAL, DD,

MEL, SING); 44th mile Bruas, Burn-Murdoch 265 (SING); Dindings,

Fox 12649 (K, SING).

TRENGGANU: Bukit Kajang, Kemaman, Corner S.F.N. 30388 (DD,

k, KEP, SING); Ulu Bendong Kemaman, Corner S.F.N. 30181 (KEP,

SING); Belara F.R., Wood K.F.N. 76070 (KEP, SING); off 23rd mile

Kuala Trengganu-Besut Road, Belara F.R., Sinclair & Kiah S.F.N.

408538 (E, K, SING).

PAHANG: Bukit Kajang Reserve, Raub, Kalong F.D. 20313 (SING);

Cherai Hill, Rompin, Foxworthy F.D. 3215 (SING); Sungei Rompin,

Yeop F.D. 3233 (K, SING); Bukit Goh F.R., Kuantan, Yeop bin Ab-

dul Rahim F.D. 3103 (SING); Kuantan Road, Mahamed C.F. 881

(SING); Kuantan, Mahamed C.F. 2724 (SING); Benchah F.R., Kuala

Lipis, Mat Nong C.F. 4010 (K, SING).

SELANGOR: Sungei Buloh Reserve, Kiai F.D. 8265 (KEP); 14th mile

Kanching Reserve, Strugnell F.D. 12472 (KEP); 21st mile Ginting

Simpah, Strugnell F.D. 12726 (KEP).

JOHORE: 10th mile Mawai-Jemaluang Road, Corner S.F.N. 28671

(K, SING); Sungei Sedili, Corner, 16th July, 1939 (SING); Nam

Heng, Sinclair S.F.N. 40362 (BK, BM, BO, DD, E, K, L, P, SING).

SINGAPORE: Changi, Ridley 3377 (CAL, SING); Sumbawang, Good-

enough 5572 (BM, CAL, K, MEL, SING); Bukit Mandai, Ridley

3581 (a) (CAL, K, SING); Nee Soon Seletar Forest, Sinclair S.F.N.

40374 (BO, E, K, L, SING); Sungei Buloh, Ridley 6266 (K, SING).

DISTRIBUTION: Borneo and Sumatra, Philippines.

Type MATERIAL: Wray and King’s numbers.

360

Vol. XVI. (1958).

The silvery-brown undersurface of the leaves and the absence of

reticulations should distinguish this species from the others. Ac-

cording to Ridley, the aril and seed have a spicy odour.

(9) M. fragrans Houtt., Handleid. Hist. Nat. Linn. 2, 3 (1774)

333; Blume, Rumphia 1 (1835) 180 T. 55; A. DC., Prodr. 14, 1

(1856) 189; Migq., Fl. Ind. Bat. 1, 2 (1858) 53; Hk. f., Fl. Br. Ind.

5 (1886) 102; King in Ann. Roy. Bot. Gard. Calc. 3 (1891)

287 Pl. 108; Warb., Monog. Myrist. (1897) 458; Gamble, Mat.

Pave. 3, 25 (1012) 233; Ridley, F.M.P. 3.( 1924) 63.

Synonyms: M. offcinalis L. f., Suppl. (1781) 265. M. mos-

chata Thunb. in Act. Holm. (1782) 45. M. aromatica Lamk. in

Act. Paris (1788) 155 T. 5—7; Roxb., Plants of Coromand. 3

(1819) 274 T. 267. M. amboinensis Gandoger in Bull. Soc. Bot.

France 66 (1919) 225 in clavi. M. laurella Gandoger l.c. 226 in

clavi. M. philippinensis Gandoger in Bull. Soc. Bot. France 66

(1919) 226 in clavi.i—Fig. 29. Plate VII B.

Tree 4-5 m. high with all parts aromatic, usually dioecious but

sometimes male and female flowers on the same tree. Bark greyish-

black, slightly fissured longitudinally in older trees. Twigs glabrous,

slender, greyish brown. Leaves coriaceous, glabrous, medium to

dark green above and shining, light green or sub-glaucous beneath

with the lower midrib yellowish green, both it and the nerves

beneath reddish when dry, elliptic or oblong-lanceolate, base acute,

apex acute or slightly acuminate; nerves 8—11 pairs, slender, curv-

ing and running out to the edge with indistinct anastomosis, reticul-

ations not usually visible above but fairly distinct beneath, even

when fresh, forming a lax network; length 6-13 cm.; breadth

3:-5—6:5 cm.; petiole 1 cm. long. Male and female inflorescences

similar, glabrous, axillary, the flowers in umbellate cymes; main

axis 1-1-5 cm. long, not branched or rarely branched more than

twice; pedicels 1-1-5 cm. long, glabrous with a minute caducous

bracteole at the base of the flower. Flowers fragrant, creamy-yellow,,

glabrous on both surfaces, waxy, coriaceous, up to 1 cm. long,

ellipsoid, sub-urceolate, 3-toothed, teeth reflexed. Male androecium

up to 7 mm. long, acute at the apex, the stalk 2 mm. long, glabrous;

anthers 8-12. Female flowers: ovary puberulous, 7 mm. long with

2-lipped stigma. Fruit broadly pyriform, yellow, glabrous, often

drooping, 6—9 cm. long and nearly as broad with a circumferential

longitudinal ridge and persistent remains of the stigma; stalk

thickest at the base of the fruit; pericarp yellow, succulent. Aril

red and much laciniate. Seed purplish brown, broadly ovoid.

PENANG: Wallich 6785 (G Boiss., K) and 6785b, c, d, e (K); King

1677 (CAL); hillside above Tanjong Bunga, Sinclair S.F.N. 39258

(K, L, SING).

361

Gardens Bulletin, S.

Fig. 29. Myristica fragrans Houtt.

A, Leafy twig with female inflorescence. B, Male flower. C, Staminal

column. D, Female flower. E, Ovary. F, Fruit. G, Fruit showing

seed and aril. A-—G from tree behind office, Botanic Gardens,

Singapore.

362

Vol. XVI. (1958).

PERAK: Kuala Kangsar, Haniff S.F.N. 14917 (SING).

SELANGOR: Batu Caves, Hamid C.F. 6445 (BRI, K, SING).

MALACCA: ce 4352 (K); Maingay 1285 (CAL, K, L); Derry,

date 1892 (SING

SINGAPORE: re ee 11 (BM, BR, E, MEL, NY); Cuming 2418

(CAL, K); Murton, date 1878 (SING): Botanic Gardens, Lawn Z,

Furtado S.F.N. 34860 (BRI, KEP, SING); ditto, Nur, 4th July, 1928

(SING); ditto, Office Lines, Nur, 17th June, 1924 (SING); ditto,

Arboretum, Nur, 16th June, 1924 (SING).

DISTRIBUTION: Moluccas.

TyPE MATERIAL: Houttuyn s.n., Banda (C, L). The oldest specimen

but not type material, is from Banda, in Herb. Breyne, date 1682 (W).

I believe this specimen was burnt during the war as I have not seen

it among material on loan from Vienna.

The Commercial Nutmeg is a native of the east Moluccas and

was formerly cultivated in Singapore and Penang. In Penang it is

still to be seen on hillsides and in abandoned clearings giving one

the impression that it is native. It has smaller leaves than any of

the native Malayan species and is not likely to be confused with

them.

(10) M. imers Bl., Bijdr. (1828) 575; Rumphia 1 (1835) 184 T.

58; A.DC., Prodr. 14, 1 (1856) 190 excl. sp. Roxb. et Cum.,

Mig., Fl. Ind. Bat. 1, 2 (1858) 57 excl. sp. Roxb. et Cum.;

Warb., Monog. Myrist. (1897) 521 T. 16 Fig. 2; Gamble, Mat.

F.M.P. 5, 23 (1912) 230; Ridley, F.M.P. 3 (1924) 64.

Synonyms: M. sublanceolata Migq., Fl. Ind. Bat. 1, 2 (1859)

58. M. vordermannii Warb., Monog. Myrist. (1897) 525 T.

14 Figs 1-3. M. heritieriifolia Pierre ex Lecomte in Flor. Gén.

de L’Indo-Chine 5, 2 (1914) 98—Fig. 30. Plate VIII A.

Tree 10-33 m. high; stilt roots sometimes present. Bark brittle,

greyish black, longitudinally fissured; wood reddish, sap pink,

watery, copious. Twigs with dark reddish brown bark which tends to

crack; youngest parts slender with an elongate, slender, minutely

pubescent, terminal bud. Leaves chartaceous, medium green and

dull above, paler green beneath, often drying slightly black on the

upper surface and reddish brown beneath, lanceolate, apex acute,

base acute; midrib lying in a groove flush with the upper surface,

flat, Imm. broad, paler green than the rest of the leaf, raised

beneath and yellowish green, nerves 12—15 pairs, distinct above but

faint beneath, not very straight, curving near the margin, the line of

anastomosis broken and indistinct; reticulations faint or indistinct;

length 12—20 cm.; breadth 3-6 cm.; petiole 1-5 cm. long, slender.

Male inflorescence a short, axillary panicle, 2—2:5 cm. long, bearing

the flowers in sub-umbels on the short, 2—3 mm. long branches;

pedicels 4 mm. long; bracteoles at the base of the flower, 1 mm.

363

Gardens Bulletin, S.

TueRMM! Der.

Fig. 30. Myristica iners Bl.

A, Leafy twig. B, Fruit. C, Fruit, immature, showing seed and aril.

D, Ripe fruit. E, Male inflorescences. F, Male flower. G, Staminal

column. A—D from fresh material of tree in Arboretum, Botanic

Gardens, Singapore, Sinclair S.F.N. 40048. E-G from Corner

S.F.N. 36282.

364

Vol. XVI. (1958).

long, deciduous. Male flowers ovoid, narrowed towards the apex,

rusty-puberulous, 7 mm. long and 6 mm. broad, 3 toothed; androe-

cium 6 mm. long, stalk 3 mm. long, sterile at the extreme top; anthers

9-10, about 3 mm. long. Female inflorescence 1-4 cm. long. Female

flowers campanulate; ovary sub-globose, glabrous. Fruit glabrous,

pale yellow, oblong to oblong-ovate, line of dehiscence faint, 6:-5—8

cm. long, 4:5 cm. broad; pericarp 1-3 cm. thick; stalk rather

slender, 1-5—3-5 cm. long. Aril bright red, divided nearly to the base

or down to licm. from it, the divisions more numerous and

narrower towards the apex. Seed shining, brownish black, 4-2—5

cm. long.

KEDAH: Gunong Raya F.R., Langkawi, Wyatt-Smith K.F.N. 71188

(KEP); Teloi F.R., Baling, Shariff bin Omar K.F.N. 66396 (KEP);

Bukit Lada Achel, Hassan 7571 (K).

PENANG: Government Hill, Curtis 2455 (BM, CAL, K, KEP, SING)

not M. maingayi; Highland Hill above rifle range, Haniff S.F.N. 9072

(SING); near Lily Pond, Waterfall Garden, Das s.n. 30th May, 1955

(SING).

PERAK: Scortechini s.n. (K); Gunong Batu Puteh, Wray 1214 (CAL,

K).

TRENGGANU: Kemudi, Dungun, Ahmad bin Ibrahim K.F.N. 55352

(KEP).

PAHANG: Pulau Tioman, Burkill S.F.N. 1014 (K, SING).

SELANGOR: Semenyih, Hume F.M.S. Mus. Herb. 7952 (SING);

Bukit Lagong F.R., Compt. 17, Wyatt-Smith K.F.N. 65536 (KEP).

Matacca: Bukit Jus, Derry 1033 (CAL, SING).

JoHORE: Kulai Young Estate, Corner, 18th June, 1939 (SING);

Sungei Sedili Corner, 2nd March, 1938 (SING); Mawai, Corner S.F.N.

28186 (SING) and Ngadiman S.F.N. 36863 (BRI, DD, K, KEP,

SING); Gunong Banang F.R., Batu Pahat, Suleiman K.F.N. 70220

(KEP).

SINGAPORE: Mandai Road, Corner S.F. Nos. 36282 (BM, DD, K,

KEP, SING); 37134 (SING) and 28097 (K, SING); Arboretum Bo-

tanic Gardens, Sinclair S.F.N. 40048 (BM, BO, DD, E, K, L, P,

SING).

DISTRIBUTION: Siam, Indo-China, Sumatra, Java, Island of Mendanu

near Billiton. Probably also Banka.

Type MATERIAL: Blume, Java (CAL, L, NY, S). Pierre 5435 (P)

holotype of M. heritieriifolia, near Baria, Cochinchina. Vordermann 19

(L) holotype of M. vordermannii, and isotype (CAL), Island of

Mendanu, near Billiton.

The fruit is very beautiful, pale yellow with a red aril and an

almost black seed, and is similar to, but slightly smaller than that

of M. maingayi. M. iners has been much confused with M. maingayi

in herbaria chiefly because little material of M. maingayi has been

collected until recently. The leaves in iners are chartaceous and not

coriaceous; they are usually narrower and the reticulations less dis-

tinct or invisible. Other points of difference are the slender apical

portions of the twigs with a puberulous, slender terminal bud, the

365

Gardens Bulletin, S.

presence sometimes of stilt roots, the puberulous flower and the

more slender fruiting stalks. The bark is similar. The slender twigs

and the glabrous fruits recall those of M. malaccensis.

(11) M. crassa King in Ann. Roy. Bot. Gard. Calc. 3 (1891) 293

Pl. 117; Warb., Monog. Myrist. (1897) 495; Gamble, Mat.

F.M.P. 5, 23 (1912) 234; Ridley, F.M.P. 3 (1924) 66.

Synonym: M. suavis King in Ann. Roy Bot. Gard Calc. 3

(1891) 295 Pl. 121; Warb., Monog. Myrist. (1897) 441;

Gamble, Mat. F.M.P. 5, 23 (1912) 232; Ridley, F.M°P. 3

(1924) 65. M. lowiana King in Ann. Roy. Bot. Gard. Calc. 3

(1891) 293 Pl. 120 Figs. 1, 5, 6, 7 flowering material King Nos.

5587 et 7258 tantum.—Fig. 31. Plate VIII B.

Tree 12-18 m. high with spreading branches; a few week stilt

roots present. Bark brown, slightly fissured longitudinally, the

fissures very shallow. Twigs stout, even at the apices, striate,

glabrous except the minutely rusty-pubescent buds. Leaves cori-

aceous, glabrous, dark green and dull above, faintly glaucous

beneath, dark brown when dry, elliptic-oblong to oblong-lanceolate,

acute at the apex and base; nerves 15—22 pairs, impressed above,

prominent beneath, secondary nerves sometimes present between

the primary, line of interarching, faint or invisible; reticulations

rarely visible; length 18-40 cm.; breadth 5—12 cm.; petiole

1-5-3 cm. long, stout, channelled above. Inflorescence axillary, the

axis very short, 5 mm.—1 cm. long, woody with scars of former

flowers as in Knema, containing male and female flowers mixed or

dioecious; pedicels 4-8 mm. long, 1 mm. or less thick; bracteole

deciduous, 2 mm. long, embracing the base of the flower on one

side. Flowers coriaceous, pale yellow, with a faint, sweet odour,

puberulous outside, glabrous inside, urceolate with 3 triangular,

reflexed teeth, extending about 1/3 way down; length 4-7 mm.;

breadth 3 mm. in the male, the female 4 mm. broad and more

swollen; staminal column 5 mm. long, its stalk 2-5 mm. long,

pubescent, the apex obtuse and sterile; anthers 12-18, linear, 2

mm. long; ovary 5 mm. long, pubescent, ovate with a 2—lipped

stigma. Fruit ovoid-globose grooved longitudinally, minutely rusty-

puberulous, tending to become glabrous, the scurf easily rubbed

off, 2:5—4-5 cm. in diameter; pericarp 1 cm. thick, coriaceous; stalk

stout, 5 mm.—1 cm. long. Aril fleshy, fragrant, edible, yellow,

laciniate from the middle or below to the apex. Seed ovoid.

Perak: B.P.D., King 7756 (CAL, DD, MEL, P, UPS) fruiting, type

material; Gopeng, King Nos. 4475 (CAL, FI, SING) fruiting, type

material and 6061 (CAL, DD, E, K, SING) fruiting, both type mate-

rial; Larut, King Nos. 2758 (BM, CAL, FI, G, K, UPS) flowering,

type material; 5065 (CAL, FI, G, K, L) fruiting, type material; 5537

366

Vol. XVI. (1958).

Fig. 31. Myristica crassa King.

A, Leafy twig with male and female flowers and fruit. B, Inflorescence

of male and female flowers. C, Flower dissected to show staminal

column. D, Staminal column, cross section. E, Female flower from

above. F, Female flower dissected to show ovary. G, Fruit. A-G

from Sinclair S.F.N. 39490.

367

Gardens Bulletin, S.

(CAL, K, L) flowering, not lowiana as stated and 7258 (BM, CAL, G,

K, P) has been wrongly named lowiana; Ulu Bubong, King 19038

(BM, CAL, K) fruiting, type material; Waterfall Hill, Wray 646 (K,

SING) type material; Pondok Tanjong Reserve, Taiping, Mat Gani,

F.D. 9777 (SING).

Matacca: Cantley 35 (K) type material; Bukit Danan, Alvins 575

(SING); Merlimau, Alvins, 14th Dec., 1885 (SING); Selandor, Cant-

ley (CAL) holotype of M. suavis.

JOHORE: 134 mile Mawai-Jemaluang Road, Corner S.F.N. 34912

(SING); Sungei Sedili, Ngadiman S.F.N. 36919 (SING).

SINGAPORE: MacRitchie Reservoir off Lornie Road, Sinclair S.F.N.

39490 (B, BK, BM, BO, DD, Delhi Univ., E, K, L, M, NY, PNH,

SAN, SING).

DISTRIBUTION: Confined to Malaya.

I have found male and female flowers in the same inflorescence

in the Singapore and Johore material quoted here, an unusual

occurrence in Myristicaceae. King’s type material was mostly fruit-

ing but in King 2758 flowering material, male, there is no trace of

female flowers and King states “female flowers unknown”. However

in M. suavis which I have reduced to M. crassa there are rudimen-

tary anthers present in the female flowers. King’s flowering material,

Nos 5537 and 7258 of M. lowiana is M. crassa. M. crassa is

extremely like M. teijsmannii from Java but I must keep them

separate for the time being, since I have not seen sufficient material

of teijsmannii to make a decision. Here are small differences which

may count for not uniting them. In M. teijsmannii, the leaves are less

coriaceous, the petioles not so stout and the twigs more slender,

3 mm. thick at the apex as against S—6 mm. in crassa. The flowers

are obtuse in bud, ovoid with a persistent bract 5 mm. long, where-

as in crassa the flowers are pointed in bud, urceolate and the bract

is deciduous and 2 mm. long.

3. HORSFIELDIA Willd. Sp. Pl. 4 (1806) 872 (non Bl. =

Harmsiopanax Warb.); Pers. Symb. 2 (1807) 635; Warb.,

Monog. Myrist. (1897) 130 and 262; Gamble, Mat. F.M.P. 5,

23 (1912) 206; Ridley, F.M.P. 3 (1924) 54.

Synonyms: Myristica sec. Pyrrhosa Bl., Rumphia 1 (1836)

190; Hk. f. et Th., Fl. Ind. 1 (1855) sections Irya 159, Pyrrhosa

160, Eumyristica 162 pro parte et Fl. Br. Ind. 5 (1886) 105 sect.

Pyrrhosa pro parte; A.DC., Prodr. 14 (1856) sects. Caloneura

194 pro parte, Horsfieldia 200, Irya and Pyrrhosa 202; Miq.,

Fl. Ind. Bat. 1, 2 (1858) sects. Horsfieldia 63, Irya and Pyrrhosa

64; King in Ann. Roy. Bot. Gard. Calc. 3 (1891) sects. Eumyri-

stica subsect. Horsfieldia 282, Pyrrhosa pro parte 282 and Irya

284. Pyrrhosa Endl. Gen. 830 (1839).

368

Vol. XVI. (1958).

Trees 5—30 m. high. Dioecious, very rarely monoecious (H.

canarioides). Stilt roots sometimes present. Bark usually reddish-

brown, smooth or more often striate or rough with circular or

irregular dents, sometimes flaking but mostly not; sap red. Twigs

striate, sometimes lenticellate, two raised ridges extending from

petiole to petiole in two Malayan species. Leaves membranous,

chartaceous or coriaceous, often becoming brittle in herbaria,

mesophyll without reticulate sclerenchyma, nerves prominent or

faint, reticulations often present but few and lax, never forming a

dense close network as in Knema; lower surface not glaucous or

whitish. Inflorescence axis laxly branched, often several times,

rarely condensed as in H. iryaghedhi which has sessile flowers;

flowers numerous in sub-umbellate cymules, pedicelled, rarely

sessile; bracts sometimes large, early caducous; bracteoles absent.

Flowers waxy-yellow, often sweet-scented, small, globose, sub-

clavate or laterally compressed, perianth 2-lobed or 3-lobed some-

times (a 2—lobed species may have a few 3-lobed flowers in the

same inflorescence or a 3-lobed species 4 and 5 lobes), the lobes

rarely spreading; female flowers usually larger than the male;

androecium globose, sub-globose, annular, oblong, cylindric or

trigonous and triangular in cross-section, sometimes forming a cup

or just a shallow depression at the top; anthers 6—30, adnate to the

column by their backs and to each other by their edges or their

edges free for some distance, the apices incurved and rolled back

into the cup or the apices erect, or free from the column for some

short distance downwards and easily separated from each other

in which case the column is trigonous. Style absent. Stigmas

minutely bilobed or connate, a groove between the two. Ovary

glabrous, less often pubescent. Fruit globose or sub-globose, nearly

always glabrous; perianth leathery, rarely succulent. Aril completely

covering the seed, entire or with a few lobes at the apex. Seed with

a thin woody testa; albumen ruminate, a fixed oil present in the

albumen but no starch; cotyledons connate at the base.

TYPE OF GENUS: Horsfieldia odorata Willd. Sp. Pl. 4 (1806) 872

= H. iryaghedhi (Gaertn.) Warb., Monog. Myrist. (1897) 332.

DISTRIBUTION: Ceylon, India, Burma, Andamans, Nicobars, Siam,

Indo-China, Hainan, Yunnan, Malay Peninsula, Malay Islands including

Philippines and New Guinea, Solomon Islands, N. Australia, Polynesia

and Micronesia.

Horsfieldia differs from the other Malayan genera in the structure

of its androecium which may be cup-shaped with the anthers rolled

back into the cup. The cup may be deep or just a mere depression

or the depression may be absent as in H. fulva, the resulting column

being like that a Myristica. In Horsfieldia species 17-19, there is

however, a close approach to the type of androecium found in

369

Gardens Bulletin, S.

Gymnacranthera. In these three species, the anthers are free at

their apices and even for a short way down from the apices, but

the apices are not inflexed; further the column is triangular in cross-

section and not circular. Horsfieldia also diflers from the other

Malayan genera in its leaves not being glaucous or whitish beneath

and in its perianth being 2—lobed in one group (Bivalves). It differs

from Knema in the paucity of the reticulations of the leaves, the

branched inflorescence, the ebracteolate pedicels, the stigma never

more than 2—lobed, the connate cotyledons and in the absence of

a Style and in the absence of starch in the albumen. It differs from

Gymnacranthera in the aril not being laciniate to the base and in

the striate twigs. From Myristica, it differs in the more closely

striate twigs, more flowers in the inflorescence, the ebracteolate

pedicels, the non-lacinate aril and the absence of starch in the

albumen.

The genus has a wide distribution throughout Malaysia, but with

the greatest number of species in two centres, Malaya and New

Guinea. There are some 19 in Malaya and between 20 and 30 in

New Guinea, if certain undescribed species from the latter are in-

cluded. The Malayan species are not difficult to identify but they

have been much confused and wrongly named in herbaria. In fact

most of them can be indentified from sterile material, provided that

a sufficient length of twig is present. Two keys are given here, one

based chiefly on the structure of the flower and the androecium.

The other is for sterile material but fruit and flower characters are

sometimes added as an extra aid in the key, if they are considered

helpful. I have not at this stage grouped the species into named

sections nor have I adopted those sections and series of Warburg

since they are rather artificial. He has three sections /rya, Pyrrhosa

and Orthanthera. The more natural classification, in my opinion

would be to make two sections Bivalves and Trivalves and to divide

these up into sub-sections and series. I cannot do this accurately at

present without having studied all the Malaysian species, especially

those from New Guinea. Many of the New Guinea species fall into

bivalves and some of these are difficult to identify. There are only

three species with a bi-valvate perianth in Malaya and one of these,

H. irya, is put into section Jrya, sub-section Euirya by Warburg. It

should, I think, go with the other bivalves in series globularia or in

a series closely related to globularia, but at least under bivalves.

H. sucosa should not be placed with irya and crassifolia but should

go with amygdalina, kingii, bracteosa etc. in sub-section Eupyr-

rhosa. He has, I think, correctly placed superba, flocculosa, fulva,

grandis and tomentosa in sub-section Eupyrrhosa and canarioides

(now H. macrocoma var. canarioides) in sub-section Papillosae.

370

Vol. XVI. (1958).

H. polyspherula, subglobosa and ridleyana go together on account

of the triangular androecium and he has also grouped them together

but has placed them in a sub-section Trivalves when, as is pointed

out above, Trivalves would be happier name for a section. The three

species in Section Orthanthera have the flowers densely packed

together but only one of these has a bilobed perianth.

In the following scheme I have attempted to group the Malayan

species under bivalves and trivalves using brackets to group the

more closely related ones.

Bivalves Trivalves

(H. bivalvis androecium H.macrocoma var. Inasub-section.

H. irya cup-shaped canarioides Androecium turbinate

H. crassifolia to annular with a

depression.

H. superba

ft fulva

(H. flocculosa In one sub-section divi-

< H. grandis ded into series or less

| H. tomentosa preferrably in more

than one sub-section.

H. wallichii Androecium varying,

H. penangiana cylindrical oblong, or

disc-shaped with or

H. glabra without a depression.

Cr subalpina

H. punctatifolia

H. sucosa

| H. bracteosa

(H. polyspherula In a sub-section.

H. subglobosa Androecium triangular

_H. ridleyana in cross section.

Sect. Orthanthera In a _ sub-section; no

example in Malaya.

Androecium sub-clavate,

apex very Slightly

depressed.

SCIENTIFIC KEY TO HORSFIELDIA

a. Perianth bivalved—Bivalves

b. Male perianth large, 3-3-5 mm. long and 4 mm. broad,

transversely flattened; androecium transversely elongate;

anthers 20-30, their apices rolled back into the androecium

cup and reaching nearly to its bottom and more or less

free at the extreme apex; male pedicels 2 mm. long

(1) H. bivalvis

371

Gardens Bulletin, S.

5. Male perianth smaller, sub-globose or slightly flattened, 1—

1:5 mm. in diam.; androecium sub-globose or slightly

transversely elongate; anthers less than 20, average 6-10,

their apices rolled back, obtuse and free but scarcely reach-

ing to bottom of the cup; male pedicels 0-5-1 mm. long

c. Twigs with 2 raised ridges from petiole to petiole. Leaves.

membranous or thinly. coriaceous, covered with white

marks when dry, acute at the apex, glabrous beneath.

Perianth not persisting in fruit (2) H. irya

c. Twigs without 2 raised ridges. Leaves thickly coriaceous

and covered with minute brown scales beneath; obtuse

or emarginate at the apex. Perianth persisting in fruit

(3) H. crassifolia

a. Perianth 3-valved—Trivalves

d. (Monoecious but male and female flowers not in the same

inflorescence). Male flowers papillose, split almost to the

base and the lobes spreading widely. Female inflorescence

a long, lax, spreading panicle; 13—23 cm. long with oblong-

ellipsoid fruit 6-8 cm. long and 3—4 cm. in diam.; stalks

3—4 cm. long (4) H. macrocoma var. canarioides

d. (Dioecious). Male flowers papillose or not, not spilt almost

to the base nor the lobes widely spreading. Female in-

florescence never so long, less branched, fruit not ee

oblong-ellipsoid

e. Androecium a globose, depressed column or elongated and

oblong, not triangular in cross-section, a shallow de-

pression or not present at the apex; anthers not free at

the sides and not or very slightly at the apices, not

easily separated from each other; apices mostly obtuse,

sometimes slightly acute, incurved and not erect

f. Twigs with pale brown or pink lenticels. Androecium

longer than broad, oblong or elongated, with a shallow

depression at the top or not. Flowers coriaceous or

fleshy

g. Leaves very large, 25-70 cm. long ah 10-22 cm.

broad, nerves 15—20 pairs, lower surface with scurfy

tomentum which is easily rubbed off with the finger.

Flowers larger than any of the other Malayan

Horsfieldia species, male 7-8 mm. long and 5 mm.

broad, female 9 mm.—1:2 cm. long and 8 mm. broad.

Androecium 4—5 mm. long. Fruit 7-9 cm. long and

5:5—6:8 cm. in diam. (5) H. superba

SiS

Vol. XVI. (1958).

g. Leaves much smaller, 13-24 cm. long and 4—11 cm.

broad; nerves 12-14 pairs, lower surface glabrous.

Male flowers 3-5 mm. long, female 5—6 mm. long.

Androecium 2—3 mm. long. Fruit 2-5-3 cm. long

and 2:5 cm. in diam. (6) H. fulva

f. Twigs without pale brown or pink lenticels (except in

punctatifolia). Androecium longer than broad or

otherwise. Flowers coriaceous or not

h. Leaves tomentose beneath. Perianth membranous,

often with hyaline or black dots, best seen with a

lens or on soaking dried material. Male pedicels

very slender

i. Tomentum dense and woolly, not easily rubbed off

with the finger. Leaves large 25—45 cm. long and

10—20 cm. broad, coriaceous to thinly coriaceous,

nerves 15-20 pairs

j. Tomentum soft; bark of the twigs cracking. Leaves

coriaceous and rough above but not scabrid or

harsh when felt with the finger; reticulations

mostly invisible, except a few scattered ones on

the lower surface. Male flowers 3 mm. long,

obovoid; androecium longer than broad, 2 mm.

long and 1:5 mm. broad; anthers 10; male

pedicels 3-4 mm. long (7) H. flocculosa

j. Tomentum harsh, bark of the twigs not cracking.

Leaves thinly coriaceous, harsh above when

felt with the finger; reticulations distinct

on both surfaces, the upper surface almost bul-

late. Male flowers 1—2 mm. long, globose;

androecium 1 mm. long and 1-1-5 mm. broad;

anthers 13-15; male pedicels 1 mm. long

(8) H. grandis

i. Tomentum short and scurfy beneath, not very dense,

easily rubbed off with the finger. Leaves smaller,

10-25 cm. long and 5—12 cm. broad, membranous,

nerves 12—15 pairs (9) H. tomentosa

h. Leaves glabrous beneath, never tomentose. (H. wallichii

is pubescent on the lower midrib when young, soon

glabrous). Perianth membranous or coriaceous. Male

pedicels slender or not

373

Gardens Bulletin, S.

k. Male flowers sessile or nearly so, the pedicels thick

and up to 1 mm. long; perianth succulent

and coriaceous, 3—4 mm. long, the female larger;

anthers 15-16. Sides of leaf more or less parallel;

nerves 16—22 pairs

(10) H. wallichii

k. Male flowers on slender pedicels 1 mm. long or

over; perianth usually membranous, 1-3 mm.

long; anthers 5-10. Sides of leaf not parallel, the

leaf broadest at the middle, gradually narrowed

from there to apex and base

1. Leaves 7-13 cm. long and 2-:8—3:5 cm. broad;

nerves very fine on the lower surface, not raised

and cannot be felt with the finger

(11) H. penangiana

1. Leaves 9-22 cm. long and 3-5—6 cm. broad; nerves

raised on the lower surface

m. Androecium ionger than broad, an elongated

column without any depression at the apex.

Leaves obtuse or bluntly acute at the apex;

coriaceous. Mountain trees (mostly acute in

H. subalpina)

n. Veins of leaf sunk above (perhaps sometimes

level with surface, never raised). Fruit large,

8 cm. x 6:5—7 cm. Inflorescence axis at first

puberulous, later glabrous

(12) H. subalpina

n. Veins of leaf level with surface above. Fruit

smaller, 2:-5—3-5 cm. x 2 cm. Inflorescence

axis always glabrous (13) H. glabra

m. Androecium not longer than broad, sub-

globose or a flattened disc with a depression

at the apex or centre. Leaves acute at the

apex, coriaceous or membranous. Lowland

forest trees

o. Twigs reddish brown, slender. Leaves

punctate beneath. Perianth not persistent

in fruit (14) H. punctatifolia

o. Twigs pale straw-coloured, stout. Leaves

not punctate beneath. Perianth persistent

in fruit in bracteosa |

374

Vol. XVI. (1958).

p. Leaves crowded round the terminal bud,

coriaceous, narrowed to the base and

decurrent on to the petiole, 3-5—4-5 cm.

broad at the middle; petiole 2—2:5 cm.

long; terminal bud 4-5 mm. thick

(15) H. sucosa

p. Leaves not crowded round the terminal

bud, membranous, less narrowed at the

base and not or only slightly decurrent

on to the petiole, 5-8 cm. broad at the

middle; petiole 1-1-5 cm. long; ter-

minal bud 2 mm. thick

(16) H. bracteosa

e. Androecium trigonous, triangular in cross-section; anthers

easily separated from each other, acute, free and erect

at their apices, their backs adnate to the column at the

base, the edges more or less free

q. Leaves 10-28 cm. long and 4-10 cm. broad; nerves

10-21 pairs, distinct on both surfaces; reticulations

faintly visible beneath

r. Inflorescence axis rusty-tomentose. Twigs rusty-fur-

furaceous at the apex, slender and about 2 mm.

thick there. Leaves, average length 10-14 cm.;

occasionally up to 21 cm.; breadth 4-5-(8) cm;

midrib at first rusty-pubescent beneath, later

glabrous, not broadening above the petiole on the

upper surface (17) H. pelyspherula

r. Inflorescence axis puberulous or glabrous. Twigs

» glabrous, stouter, 3—4 mm. thick at the apex. Leaves

larger, 15—28 cm. long and 4—10 cm. broad; midrib

glabrous beneath, broadening above the petiole on

the upper surface

s. Twigs without 2 raised lines from petiole to petiole

(18) H. subglobosa var. subglobosa

s. Twigs with 2 raised lines from petiole to petiole

(18) H. subglobosa var. brachiata

q. Leaves 5-5-—13 cm. long and 2-3-5 cm. broad; nerves

7-10 pairs, faint on both surfaces; reticulations in-

visible (19) H. ridleyana

375

Gardens Bulletin, S.

KEY TO HORSFIELDIA

(Based mainly on sterile material and fruit; flower characters are

occasionally employed when species are close to each other)

a. Leaves tomentose beneath

b. Tomentum dense, woolly

c. Tomentum soft, light brown, bark of twigs cracking. Leaves

coriaceous and rough above but not scabrid and harsh

when felt with the finger; reticulations mostly invisible

except a few scattered ones on the lower surface

(7) H. flocculosa

c. Tomentum harsh, dark brown, bark of twigs cracking.

Leaves thinly coriaceous, harsh above when felt with the

finger; reticulations distinct on both surfaces, the upper

surface almost bullate (8) H. grandis

b. Tomentum a thin, rusty scurf which tends to rub off

d. Leaves coriaceous, very large, 25—70 cm. long and 10-26

cm. broad; nerves 15-30 pairs. Fruit 7-9 cm. long and

5:5—6:8 cm. in diam. (5) H. superba

d. Leaves membranous, 12—25 cm. long and 5—12 cm. broad;

nerves 12-15 pairs. Fruit much smaller, 2-2-5 cm. long

and 1:5 cm. broad (9) H. tomentosa

a. Leaves glabrous beneath

e. Lower leaf surface punctate with black or brown dots

(14) H. punctatifolia

e. Lower surface not punctate with black or brown dots

f. Twigs glabrous, thick, pale straw-coloured

g. Leaves crowded round the terminal bud, coriaceous,

narrowed to the base and decurrent on to the petiole;

3-5—4:5 cm. broad at the middle; petiole 2—2:5 cm.

long; terminal bud 4—5 mm. thick. Fruit 6-7-5 cm.

long and 4—5 cm. in diam. Perianth not persistent in

fruit (14) H. sucosa

g. Leaves not crowded round the terminal bud, mem-

branous, less narrowed at the base and not or only

slightly decurrent on to the petiole; petiole 1-1-5 cm.

long; terminal bud 2 mm. thick. Fruit 4-5 cm. long

and 3-5 cm. in diam. Perianth 3-lobed, persistent in

fruit (15) H. bracteosa

f. Twigs not pale straw-coloured, thick or not

376

Vol. XVI. (1958).

h. 2 raised lines present on the twigs from petiole to petiole,

sometimes not always visible if only the apical portion

of the twigs is examined

i. Leaves often covered above with white spots when dry,

sides nearly parallel. Lenticels very numerous. Fruit

globose, 2—2-5 cm. in diam.; flowers if present (the

majority) with a 2-lobed perianth (2) H. irya

i. Leaves without white spots above when dry, sides

curving. Lenticels fewer. Fruit sub-globose, slightly

larger, the apex slightly pointed; flowers if present

with a 3-lobed perianth

(18) H. subglobosa var. brachiata

h. 2 raised lines on the twigs absent (but species 2 and 18

again included in this key in case the twig is an apical

portion or short and does not happen to show them)

j. Apex of leaf obtuse or emarginate, occasionally bluntly

acute, (most acute in H. subalpina)

k. Lower surface of leaf covered with minute rusty

scales. Swamp forest trees with a few weak stilt

roots. Flowers if present with a 2-lobed perianth

(3) H. crassifolia

k. Lower surface of leaf not covered with rusty scales.

Mountain trees without stilt roots. Flowers with

a 3-lobed perianth

I. Veins of leaf sunk above (perhaps sometimes level

with surface, never raised). Fruit large, 8 cm.

x 6:5—7 cm. Inflorescence axis at first puber-

ulous, later glabrous (12) H. subalpina

I. Veins of leaf level with surface above. Fruit smaller,

2:5-3-5 cm. x 2 cm. Inflorescence axis always

glabrous (13) H. glabra

j. Apex of leaf acute

m. Veins of leaf fine and faint on both surfaces, reti-

culations invisible, texture membranous, size small,

55-13 cm. long and 2-3-5 cm. broad

n. Leaves drying olive green above, length 5:5—10-5

cm. and breadth 2—3 cm. Anthers acute and free

at the apex; androecium traingular in cross-

section. Fruit subglobose, 1:2 cm. long and 1—

1-5 cm. broad (19) H. ridleyana

377

Gardens Bulletin, S.

n. Leaves drying blackish above, length 7-13 cm.

and breadth 2:8—3-5 cm. Anthers obtuse and not

free towards the apex; androecium circular in

cross-section. Fruit not seen

(11) H. penangiana

m. Veins of leaf fine or prominent above, always pro-

minent beneath, reticulations visible or not. Leaves

larger, often coriaceous

o. Leaves usually broadest at the base which is

rounded or broadly cuneate. Fruit oblong-

ellipsoid, 6-8 cm. long and 3-4 cm. in diam.,

on a long branched pendulous inflorescence, 13—

23 cm. long

(4) H. macrocoma var. canarioides:

o. Leaves not usually broadest at the base. Fruit

smaller (larger in H. wallichii) not on a long

branched pendulous inflorescence; female in-

florescence axis much shorter than 13 cm.

p. Leaves elongated, often narrow, the sides more

or less parallel and not widest at the middle

q. Twigs when dry hollow or partly hollow,

rusty-pubescent at the apex. Leaves very

coriaceous; nerves raised beneath, stout, 0:5

mm. or more thick when dry. Flowers ses-

sile or almost sessile (male pedicels up to

1 mm. long), perianth 3-valved. Fruit large,

7-8 cm. long and 5—6 cm. broad

(10) H. wallichii

q. Twigs not hollow nor rusty pubescent at the

apex. Leaves coriaceous or thinly so; nerves

less prominent beneath, thinner, less than

0-5 mm. thick beneath, when dry. Flowers

stalked and with a 2-lobed perianth. Fruit

1-3-2-5 cm. long and 1-2—2:5 cm. in diam.

r. Leaves with whitish spots above when dry

Male flowers globose, 1 mm. in diam.

(2) H. irya

r. Leaves without whitish spots when dry,

Male flowers broader than long, laterally

compressed, 3—3:5 mm. long and 4 mm.

broad (1) H. bivalvis

378

Vol. XVI. (1958).

p. Leaves broadest at the middle, the sides not

parallel but curving gradually from base to

apex

s. Veins sunk above

t. Leaves chartaceous, drying blackish green

above and medium brown beneath.

Mountain trees (12) H. subalpina

t. Leaves very coriaceous, drying pale yel-

lowish green above and pale brown be-

neath with reddish midrib and veins.

Lowland forest trees (6) H. fulva

s. Veins raised above

u. Leaves 10—14 cm. long, occasionally up to

21 cm.; 4-5-(8) cm. broad; midrib at

first rusty-pubescent, later glabrous, not

broadening above the petiole on the up-

per surface. Inflorescence axis rusty-

tomentose (17) H. polyspherula

u. Leaves larger, 15-28 cm. long and 4—10

cm. broad, midrib glabrous, broadening

above the petiole on the upper surface.

Inflorescence axis puberulous or glabrous

v. Twigs without 2 raised lines from petiole

to petiole

(18) H. subglobosa var. subglobosa

v. Twigs with 2 raised lines from petiole

to petiole

(18) H. subglobosa var. brachiata

(1) H. bivalvis (Hk. f.) Merr. in Philip. Journ. Sc. Bot. 2 (1916)

271.

Basionym: Myristica bivalvis Hk. f., Fl. Br. Ind. 5 (1886)

107; King in Ann. Roy. Bot. Gard. Calc. 3 (1891) 307 Pl. 139.

Synonyms: Horsfieldia globularia (Bl.) Warb., Monog. Myrist.

(1897) 288 T. 21 Figs 1-4; Gamble, Mat. F.M.P. 5, 23 (1912)

207; Ridley, F.M.P. (1924) 54. Myristica globularia Bl., Rum-

phia 1 (1825) 191 T. 64 Fig. 2 (non M. globularia Lamk);

Hk. f. et Th., Fl. Ind. 1 (1855) 160 excl. specimen Griff.;

A. DC., Prodr. 14, 1 (1856) 202 quoad sp. Ambon.; Miq., Fl.

Ind. Bat. 1, 2 (1858) quoad sp. Ambon.—Fig. 32. Plate VIB.

379

Gardens Bulletin, Ss.

a ae

al = OVA

oS <

OSs

Fig. 32. Horsfieldia bivalvis (Hk. f.) Merr.

A, Leafy twig with male inflorescence. B, Male flower. C, Male flower

showing staminal column. D, Staminal column in longitudinal sec-

tion. E, Fruit. A-D from tree in Botanic Gardens, Singapore. E

from Forbes 1184a.

380

Vol. XVI. (1958).

Tree 9-15 m. high with pendulous branches. Bark reddish

brown outside, soft, rough with numerous narrow, longitudinal

fissures, these being 5—10 cm. long, 2 mm. deep and 1-2 mm.

wide; inner bark reddish; wood white; sap red, watery, not copious.

Twigs glabrous except the thin, elongated terminal bud and the

extreme tips which are minutely pubescent and greenish; lower

down twigs medium brown, lenticellate, striate. Leaves thinly

coriaceous, oblong, obovate-oblong or narrowly elliptic-oblong, the

edges revolute when fresh, nearly parallel for some distance and

then narrowed to the acute base, apex acute or slightly acuminate,

dark green above and rather dull but dark brown when dry,

medium green beneath and lighter brown when dry; midrib

slightly raised above, prominent and verruculose beneath when

dry, yellowish-brown on both surfaces when fresh; nerves 13-20

pairs, fine above, prominent beneath, oblique, interarching in

loops at the margin; reticulations invisible above, occasionally a

few lax ones seen below; length 15-20 cm.; breadth 4—6 cm.;

petiole 1-1-5 cm. long, slender, channelled above. Inflorescence

a spreading panicle, glabrous or minutely rusty-puberulous, be-

coming glabrous; bracts 1-2mm. long, very early deciduous; male

inflorescence 6—16 cm. long, branched; female much shorter with

fewer, shorter branches. Flowers sweet-scented, glaucous green in

bud, changing to a waxy yellow, glabrous; perianth coriaceous,

smooth when fresh, slightly rough when dry, transversely 2-valved.

Male flowers: pedicels 2 mm. long, glabrous; perianth 3—3-5 mm.

long and 4 mm. broad, ovate, broad at the top and narrowed

towards the base; androecium an obconic, compressed, sessile,

2—lobed cup, bearing 20-30 elongate, linear, connate anthers,

each anther from its middle completely inflexed into the cup and

reaching nearly to the base of the cup. Female flowers: ovary

glabrous, sessile, stigma minutely bilobed. Fruit small, almost

globose, 1:3 cm. long and 1:2 cm. broad; pericarp thick, glabrous.

Seed 9 mm. long and 7.5 mm. broad with a thin testa, and with a

hollow cavity in the inside. Aril red, investing the seed.

SINGAPORE: Botanic Gardens’ Jungle, Murton 149 (K) holotype of

M. bivalvis Hk. f.; Ridley 393 (K, SING); Cantley, 25th Jan., 1882

(SING); by the Store, Botanic Gardens, Ridley 6736 (SING); Lawn

B, Nur 13 (SING); Furtado S.F.N. 34818 (BM, K, KEP, L, SING);

Burkill, 16th June, 1921 (SING).

DISTRIBUTION: Ambon, Celebes.

This species has, in Malaya, never been found outside the

Botanic Gardens, Singapore and there is still a large tree on Lawn

B. This tree may have been cut or pruned when young as it bears

many individual trunks close to the base of the main, very short

381

Gardens Bulletin, S.

one. All herbaria specimens collected in the Singapore Botanic

Gardens are male and the holotype described by Hooker f. was

male and was collected in the Botanic Gardens, Singapore by

Murton. My description here, except for the female flowers and

fruit, is taken from the tree on Lawn B. That of the female flowers

and fruit is from Warburg and Amboina specimens at Leiden.

Ridley and Warburg state that this species is also found in Malacca

but this is certainly a mistake, even although a label without data

has Malacca printed at the top. Further, Ridley in his Flora states,

“Bukit Naning, Malacca, Cantley.” There is a specimen of

Horsfieldia irya from Bukit Naning in the Singapore Herbarium

collected by Alvins. Its leaves resemble those of H. bivalvis. I have

seen no other collections of any Horsfieldia from Bukit Naning and

Ridley must have mistaken this sheet of H. irya for bivalvis. The

origin. of the tree in the Singapore Gardens is unknown but pro-

bably came from Bogor Botanic Gardens via Ambon as it has been

clutivated in Bogor. The Ambon plant Myristica globularia Bl. has

been identified by Warburg with H. bivalvis but is not the same as

the same as the Myristica globularia mentioned on page 160 of

Flora Indica, Hooker f. and Thomson which was collected in

Malacca by Griffith and which is H. polyspherula. In their descrip-

tion these authors state that the calyx is three to four-lobed but

bivalvis has two lobes. Myristica globularia Blume (1825) is ante-

dated by a M. globularia Lamarck (1788) — Knema globularia

(Lamk) Warb. which is not identical with Blume’s species so the

name globularia cannot be used. H. bivalvis is one of the three

species mentioned in the key in this revision which has a bi-valved

perianth. The larger flowers will distinguish it from its congeners,

H. irya and crassifolia. It is also very near H. laevigata (Bl.)

Warb. which has a tomentose ovary as distinct from the glabrous

one of bivalvis.

(2) H. irya (Gaertn.) Warb., Monog. Myrist. (1897) 317 T. 22,

Figs 1-4; Gamble, Mat. F.M.P. 5, 23 (1912) 215; Ridley,

F.M.P. 3 (1924) 58; Corner, Wayside Trees of Malaya 1

(1940) 476 Text fig. 159.

Basionym: Myristica irya Gaertn., Fruct. 1 (1788) 195 T.

41; Hk. f. et Th., Fl. Ind. (1855) 159; A. DC., Prodr. 14

(1856) 202 (excl. partim M. exaltata Wall.) ; Miq., Fl. Ind. Bat.

1, 2 (1858) 64; Thwaites, Enum. Pl. Ceylon (1864) 11; Kurz,

For. Fl. Burma 2 (1877) 282; Hk. f. et Th., Fl. Br. Ind. 5

(1886) 109 (excl. partim M. exaltata Wall.); King in Ann.

Roy. Bot. Gard. Calc. 3 (1891) 308 Pl. 141 and 141 bis;

Trimen, Fl. Ceylon 3 (1895) 435.

382

Vol. XVI. (1958).

Synonyms: M. sphaerocarpa Wall., Pl. As. Rar. (1830) 79

T. 89. M. javanica Bl., Bijdr. (1825) 576 et Rumphia 1 (1835)

190.T. 62. M. lemanniana A. DC., in Ann. Sci. Nat Sér. 4 Vol.

4 (1855) 31 T. 4 et Prodr. 14, 1 (1856) 203; Migq., Fl. Ind.

Bat. 1, 2 (1858) 66. H. lemanniana (A. DC.) Warb., Monog.

Myrist. (1897) 326; Gamble, Mat. F.M.P. 5, 23 (1912) 219;

Ridley, F.M.P. 3 (1924) 59 omnino quoad spec. typicum

tantum. M. micrantha Wall. Cat. 6807 nomen nudum.

M. vriesiana Mig., Ann. Mus. Bot. Ludg.-Bat. 2 (1865) 49.

_Horsfieldia congestiflora A. C. Smith in Journ. Arn. Arb. 22,

(1941) 64. H. amklaal Kanehira in Bot. Mag. Tokyo 47

(1933) 670 et Fl. Micr. (1933) 109.—Fig. 33. Plate IXA.

Tree 10-20 m. high with a narrow crown of slender drooping

branches at the top, sometimes slightly buttressed. Bark brownish

grey, of no great thickness, covered with characteristic, longitudin-

ally elongated fissures 3—5 cm. long and 2—3 mm. deep, not flaky,

red inside; wood white; sap copious, dark red. Twigs glabrous,

blackish green at the tips, blackish brown and rough or coarsely

striate further down, often covered with yellowish white lenticles,

prominently 2-angled with raised ridges. Leaves 2-ranked,

membranous or slightly coriaceous, rather brittle when dry,

glabrous, dull and dark green above, dark brown when dry and

often covered with whitish marks (a useful diagnostic character),

pale green beneath and medium brown when dry, oblong-lanceo-

late or narrowly oblong, base acute or rounded, apex acuminate;

nerves 10—20 pairs (average 16), at first almost horizontal, curving

gradually and interarching at the margin, rather faint above and

sunk, prominent beneath as is the midrib; reticulations mostly

invisible above, faint and slender beneath; forming a very loose

network; length 16-21 cm.; breadth 4-5—5-5 cm.; petiole 5 mm.—1

cm. long, glabrous, deeply grooved on the upper surface. [nflores-

cence in the axils of leaves or fallen leaves, + pubescent with a few

stellate hairs; bracts numerous, oblong, acute, 3 mm. long and

early deciduous, male more richly branched that the female, 10-12

cm. long with tertiary ramifications. Male flowers very numerous,

yellow, strongly scented, in glomerulose clusters on the 0-5—1 mm.

long pedicels; perianth glabrous, 2—lipped, globose, 1 mm. in

diam.; androecium broadly obovate, slightly elongated transversely

with 6-10 anthers which are free except at the base and have in-

curved apices. Female flowers fewer and larger with obovoid,

bilobed perianth; ovary glabrous, sessile, globose. Carpels in bun-

ches of 2-4, orange to reddish orange outside, pink inside, globose,

2—2:5 cm. in diam. with a circumferential ridge and a 1 cm. long

383

Gardens Bulletin, S.

Fig. 33. Horsfieldia irya (Gaertn.) Warb.

A, Leafy twig with male inflorescences. B, Male flower. C, Staminal

column. D, Fruit. E, Seed with aril. F, Seed in cross section. G,

Female inflorescences. H, Female flower. I, Ovary. J, Ovary in

longitudinal section. A—C from living material in Botanic Gardens,

Singapore. D-J from Curtis 936.

384

Vol. XVI. (1958).

stalk. Aril orange-red but more red than orange, entire or slightly

cleft at the apex, the lobes overlapping. Seed, outer coat green,

succulent, inner, hard, dull brown.

Lower S1AM: Telok Udang, Teratau, Haniff & Nur S.F.N. 7455 (K,

SING); Bukit Rajah Wang, Setul, Ridley 14957 (BM, K, SING).

KEDAH: Padang Terap, Wyatt-Smith K.F.N. 71179 (K, KEP); Tam-

pin F.R., Kota Star, Wyatt-Smith K.F.N. 64264 (KEP).

KELANTAN: Kota Bahru, Corner, 25th April, 1937 (SING); Gong

Datok, Pasir Puteh, Awang K.F.N. 68556 (KEP).

PENANG: Pulau Boetong, Curtis 936 (BM, CAL, K, SING).

PERAK: Scortechini 1738 (CAL) and s.n. (CAL, FI, G); Larut,

King 7447 (CAL, K, L, SING); Sungei Krian Estate, Bagan Serai,

Spare S.F. Nos. 33267 (SING) and 33270 (K, SING); Bruas, Din-

dings, Ridley 7206 (CAL, K, SING); Pulau Sembilan, Ridley 3043

(CAL, SING); Pulau Rumbia, Sembilan Islands, Wyatt-Smith K.F.N.

76525 (KEP).

TRENGGANU: 143 miles Kuala Trengganu-Besut Road, Sinclair &

Kiah S.F.N. 40739 (A, BK, BM, BO, DD, E, K, L, P, PNH, SING).

PAHANG: Rompin, Bidin F.D. Nos. 15611 (KEP, SING) and

15449 (KEP, SING); Soh F.D. 15408 (SING); Sedagong, Pulau Tio-

tad S.F.N. 21749 (DD, SING); Mahang, Ridley 1310 (MEL,

ING).

SELANGOR: Sungei Tinggi, Kuala Selangor, Nur S.F.N. 34145 (BM,

K, KEP, SING); Ulu Sungei Klang, 18th mile Klang-Pudu Road,

Watson F.D. 32678 (KEP).

NEGRI SEMBILAN: Sine loc. Alvins 1798 (SING). This record may be

from Malacca. More records are wanted but this common plant doubt-

less occurs here.

MALACCA: Griffith 4357 (CAL, K, M, S); Maingay 1291 (CAL, K)

and 1292 (CAL, K, L); Bukit Naning, Alvins 961 (SING).

Jouore: Sungei Pauh, Ridley 11328 (K, SING); Pulau Tinggi,

Burkill S.F.N. 897 (CAL, K, SING); Sungei Rhu Reba, Jason Bay,

Corner S.F.N. 28493 (K, SING); Lubok Pusing, Sungei Sedili, Corner

25964 (K, SING); Mawai, S. Sedili, Corner S.F.N. 25856 (K, SING);

Scudai River at 8th mile Johore-Scudai Road, Sinclair, 8th April, 1954

(E, K).

SINGAPORE: 10th mile Changi, Ridley 4814 (BM, BRI, CAL, SING);

Tyersall Avenue near Director’s House, Kiah S.F.N. 39448 (BK, BM,

BO, DD, E, K, L, P, SING); Sinclair S.F.N. 40202 (BO, DD, E, K, L,

SING); Toas, Ridley 5825 (CAL, SING); Gardens’ Jungle, Ridley

8957 (CAL, K, SING); Arboretum, Nur Gardens’ numbers 1770

(SING) and 1872 (SING); Sinclair 21st Jan., 1953 (E) and 19th

Feb., 1953 (E).

DISTRIBUTION: Ceylon, Andamans, Burma, Siam, Indo-China, Riouw,

Borneo, Sarawak, Sumatra, Java, Pulau Bawean, Celebes, Buru, Philip-

pines, Moluccas, New Guinea, Solomon Islands and Pelew (Micronesia).

A common species of lowland forest near the sea or by the edges

of streams in rather wet ground. Originally described from Ceylon

but no type specimen is specified. Warburg has divided the species

into a number of geographical forms but I think this is hardly

justifiable. It has a wide distribution, but strange to say, varies

little.

There should be no difficulty in recognizing H. irya even if

sterile but the specimen should have a good length of twig in order

385

Gardens Bulletin, S.

to see the two raised lines and the numerous white lenticels. Other

good diagnostic characters are whitish marks on the dried elongated

leaves, the small globose fruits and the numerous clusters of bivalv- |

ed flowers. H. bivalvis has similar leaves but the veins are more

erect and leave the midrib at an acute angle and there are no white

spots on the dried leaves. The New Guinea H. congestiflora and

the Micronesian H. amklaal are not different. Myristica sphaero-

carpa Wall. Cat. 6796 from Martaban, Burma (K holotype) is

also a synonym. M. exaltata Wall. (Moulmein) is partly a mixture.

The sheets in the Wallichian Herbarium, Kew consist of 6804

which is H. macrocoma var. canarioides (H. prainii) and 6804b

which is H. amydalina. The two sheets 6804 in the DC. Prodr.

herbarium (Geneva) consist of leaves of H. irya and fruit of H.

macrocoma var. canarioides (H. prainii) while the British

Museum specimen is irya only.

(3) H. crassifolia (Hk. f. et Th.) Warb., Monog. Myrist. (1897)

323; Gamble, Mat. F.M.P. 5, 23 (1912) 217; Ridley F.M.P. 3

(1924) 59. )

Basionym: Myristica crassifolia Hk. f. et Th., Fl. Ind. (1855)

160; .A DC., Prodr. 14, 1 (1856) 204; Hk. f., Fl. Br. Ind. 5

(1886) 108; King in Ann. Roy. Bot. Gard. Calc. 3 (1891)

308 Pl. 140 and Pl. 172 Fig. 4 only.

Synonyms: M. horsfieldia Wall. non B1.? Wall. Cat. 6806 pro

parte. M. paludicola King in Ann Roy. Bot. Gard. Calc. 3

(1891) 328 Pl. 169. Horsfieldia fulva var. paludicola (King)

Warb., Monog. Myrist. (1897) 299.—Fig. 34. Plate XA.

Tree 6-20 m. high with a few short stilt roots. Bark a rich

reddish-brown, rough, longitudinally fissured without loose flakes,

rough or in young trees nearly smooth; sap blood red, copious.

Twigs brown, striate, covered with minute pustular lenticels,

glabrous except the terminal bud: Leaves coriaceous and with

slightly revolute margins, dark green, glabrous and dull above,

drying olive green with blackish or brown patches, beneath covered

with yellow-brown, minute scales, the scales tending to rub off, ~

drying rich brown, elliptic or elliptic-oblong, rather wider in the

lower half, base rounded or slightly cuneate, apex usually obtuse,

less often emarginate; midrib lying in a groove and broadening

towards the base of the leaf on the upper surface, stout and pro-

minent on the lower surface; nerves 12—15 pairs, slender, oblique,

interarching faintly at the margin, level with the leaf tissue above

or slightly raised, more prominent beneath; reticulations invisible;

length rather variable, 12—28 cm.; breadth 5-10 cm.; petiole 2 cm.

long, stout, glabrous. Male inflorescence 6-13 cm. long, much

386

Vol. XVI. (1958).

‘a D

{ mm —— | mm

2 cm TURAIMG Dee 3 cm:

Fig. 34. Horsfieldia crassifolia (Hk. f. et Th.) Warb.

A, Leafy twig. B, Male inflorescence. C, Male flower. D, Staminal

column. E, Staminal column, top view. F, Female inflorescence. G,

Female flower. H, Ovary. I, Cluster of fruit. J, Fruit. K, Seed in

longitudinal section. A, I-K from Sinclair S.F.N. 40256. B from

Kiah S.F.N. 32105. C—F from Nur S.F.N. 34051. F from Koster-

mans, August 1938. G-H from Corner S.F.N. 34542.

387

Gardens Bulletin, S.

branched, rusty-tomentulose with minute stellate-hairs, the ultimate

branches slender and ending in dense clusters of flowers; bracts

ovate or oblong, 2-3 mm. long, early deciduous. Male flowers

yellow, strongly scented, 1-1-5 mm. in diam., sub-globose, slightly

flattened on top, shallowly bilobed, their pedicels 1 mm. or less

long, androecium elongated transversely, sessile, anthers 6-10,

average 7, obtuse and free at the apex. Female inflorescence

shorter, 4-8 cm. long, much stouter, the branches few and also

stouter, the tomentum similar. Female flowers fragrant, dull

orange, 2-3 mm. in diam., globose to pyriform, coriaceous,

2-lipped with 2 mm. long pedicels; ovary depressed-globose,

sessile, glabrous, ridged on one side, the ridge replaced by a furrow

on the other, the ridges and furrows persisting in mature fruit;

stigma sessile, minutely bilobed. Fruit yellow outside, pink inside,

ovoid, rough when dry, the perianth lobes reflexed and forming a

persistent collar at the base, 2—2:25 cm. long and 1-6—2 cm. broad.

Aril orange, fleshy, consisting of two separating layers and slightly

laciniate at the apex. Seed ovoid with a greenish succulent outer

coat and a hard brown inner coat.

PERAK: Pondok Tanjong, Forest Guard Salleh 9785 (KEP); Mat

Gani F.D. 9785 (SING); Parit F.R., Kinta, D.F.O. K.F.N. 54804

(KEP); Taiping, Wray 3071 (CAL, G, K, L, SING) type material of

M. paludicola; Gopeng, Kinta, King Nos. 4267 (BM, CAL, FI, G, K,

L, SING) and 4706 (BM, CAL, K, L) both type material of M. palu-

dicola; Larut, King 6688 (CAL, K) type material of M. paludicola.

TRENGGANU: Bukit Lah off Sungei Nerus near Kampong Merjor,

Sinclair & Kiah S.F.N. 40898 (A, BM, BO, E, K, L, SING).

SELANGOR: Telok F.R., Klang, Nur S.F.N. 34051 (SING).

NEGRI SEMBILAN: Tanjong Tuan F.R., Port Dickson, Sow & Lindong

K.F.N. 70486 (KEP).

MALacca: Griffith 4350 (CAL, K, P) type material of M. crassifolia;

Alvins 1288 (SING); Sungei Udang, Derry 1163 (CAL, K, SING);

Sinclair S.F.N. 40567 (E, K, SING).

JOoHORE: Sungei Kayu, Kiah S.F.N. 32105 (BM, BRI, DD, K, KEP,

SING); Nam Heng K.T., Terwya 355 (SING) Mawai, Corner, 12th

April, 1936 (SING); Sungei Sedili, Kostermans, Aug. 1938 (SING);

Mersing, Kostermanns, Aug. 1938 (SING).

SINGAPORE: Wall. Cat. 6806 (K) in part, type material of M. crassi-

folia; Anderson 9 (BM, CAL, K, MEL) type material of M. crassifolia;

Goodenough 1819 (CAL, K, SING); Bukit Mandai, Ridley 1828

(CAL, SING) and 5826 (CAL, K, SING); Goodenough 3831 (CAL,

K, SING) and 4132 (CAL, SING); Mandai Road, Corner S.F. Nos.

34542 (K, SING) and 34905 (BM, L, SING) Kiah S.F.N. 37710

(SING); Liew S.F.N. 37258 (KEP, SING); Sinclair S.F. Nos. 39533

(BO, DD, E, K, L, PNH, SAN, SING) and 40256 (BM, BO, DD, E,

K, L, P, PNH, SING); Chua Chu Kang, Ridley 10695 (CAL, K,

SING); Chan Chu Kang, Ridley 361 (CAL, SING); and 8040 (CAL,

SING); Bukit Timah, Ridley 6909 (CAL, SING); Seletar, Ridley 6126

(CAL, SING).

DISTRIBUTION: Sarawak, N. Borneo, Sumatra, Banka, Billiton.

388

ee A AF A

Vol. XVI. (1958)

A fresh water swamp forest species with bark very similar to

that of H. bivalvis. Short stilt roots are often present. The dis-

tinctive features are the coriaceous, obtuse leaves with brown

scales on the undersurface and the very small male flowers on a

short pedicel 1 mm. long or less. The female flowers are much

larger and the two-lipped perianth persists in fruit. Myristica

paludicola King and Horsfieldia fulva var. paludicola (King)

Warb. are only female plants of H. crassifolia. When King dealt

with M. crassifolia he stated that female flowers and fruit were un-

known and when he described M. fulva he stated that female flowers

were unknown. He saw female flowers and fruit of paludicola but

no male flowers. Both fulva and crassifolia have rather similar

coriaceous leaves but those of crassifolia are obtuse at the apex.

Thus, with either flowering or fruiting stages wanting, King failed

to see the connection between these species and thought that

paludicola was a separate species from crassifolia. Warburg made

the situation still more complex when he associated paludicola with

fulva and made it a variety of fulva. The male flowers of H. fulva

are much larger than those of crassifolia but are 3-lobed. The

female perianth persists in fruit in both species but it is 3-lobed in

fulva and 2-lobed in crassifolia.

(4) H. macrocoma var. canarioides (King) J. Sinclair, stat. nov.

Basionym: Myristica canarioides King in Ann. Roy, Bot.

Gard. Calc. 3 (1891) 304 Pl. 134.

Synonyms: Horsfieldia canarioides (King) Warb., Monog.

Myrist. (1897) 294 T. 21 Figs 1-2; Gamble, Mat. F.M.P. 5,

23 (1913) 208; Ridley, F.M.P. 3 (1924) 55. H. merrillii Warb.

in Perk. Frag. Fl. Philip. (1904) 49; Merr. in Philip. Journ. Sci.

Bot. 2 (1907) 274; H. oblongata Merr. in Philip. Journ. Sci. Bot.

13 (1918) 286. H. papillosa Warb., Monog. Myrist. (1897) 291

T. 21 Figs 1-3. M. papillosa (Warb.) Boerl., Handl. Fl. Ned.

Ind. 3, 1 (1900) 85 nom. alt. H. prainii (King) Warb., Monog.

Myrist. (1897) 292 T. 21 Figs 1-3. M. prainii King in Ann.

Roy. Bot. Gard. Calc. (1891) 229 Pl. 126. H. racemosa (King)

Warb., Monog. Myrist. (1897) 347; Gamble, Mat F.M.P. 5, 23

(1912) 222; Ridley, F.M.P. 3 (1924) 60. M. racemosa King

in Ann. Roy. Bot. Gard. Calc. 3 (1891) 328 Pl. 173. Myristica

sp. Hk. f., Fl. Br. Ind. 5 (1886) 113 (Maingay 1298). Embelia

ridleyi King and Gamble, Mat. F.M.P. 4, 17 (1905) 112 (type

Ridley 6324); Sinclair in Gardens’ Bull. Singapore 15 (1956)

31.—Fig. 35.

Tree 10-23 m. high. Bark about 8 mm. thick, greyish-blackish-

brown, the outer layers brittle, flaking into thin portions; wood

389

Gardens Bullennae

Fig. 35. Horsfieldia macrocoma (Mig.) Warb. var. canarioides (King) J.

Sinclair. ;

A, Leafy twig with male inflorescence. B, Male flowers. C, Male

flower. D, Staminal column. E, Fruit. A from Ridley 11270. B—D

from Ngadiman S.F.N. 36924. E from Ridley 6355.

390

Vol. XVI. (1958).

pale brown when dry. Twigs when young rather slender, smooth

and glabrous, when older stout and striate, greyish brown and

highly polished. Leaves thinly coriaceous to coriaceous, glabrous,

dark blackish brown above and shining when dry, dull and rich

medium brown beneath, broadly oblong to elliptic, broadening

slightly toward the rounded or less often slightly cuneate, some-

times unequal-sided base, apex acute; midrib flush with the upper

surface, raised beneath; nerves 13-15 pairs, nearly parallel,

oblique, fine on both surfaces, interarching near the margin;

reticulations usually indistinct but occasionally visible on the lower

surface in younger leaves, forming a faint loose network; length

12-24 cm.; breadth 3-8 cm.; petiole 1—-1:2 cm. long, grooved

above. Male inflorescence glabrous or slightly puberulous, much

branched, 6-10 cm. long and ending in sub-umbellate cymules.

Male perianth 3—4 mm. in diam., globose in bud, on slender 2—4

mm. long pedicels, lobes 3—5, spreading, acute, 1-1-5 mm. long,

papillose, minutely hairy, extending almost to the base of the flower;

androecium sub-sessile, turbinate, almost annular with 8-10

anthers incurved at the tip. Female flowers in lax, spreading,

slender 15-23 cm. long panicles, not otherwise known. Fruit

oblong-ellipsoid, 6-8 cm. long and 3-4 cm. broad, glabrous

splitting into two valves; pericarp 4-5 mm. thick; stalk 3-4 cm.

long. Aril thin, slightly lacinate at the apex. Seed 5—5-5 cm. long,

elliptic, pale brown when dry.

KEDAH: Gunong Raya F.R., Darus F.D. 12408 (SING); Ishak F.D.

7685 (SING); Gunong Jerai, compt 11, Nasruddin K.F.N. 71247

(KEP) and compt 12, Md. Salleh bin Yussof K.F.N. 73852 (KEP).

PENANG: Pulau Boetong Reserve, Curtis, March 1892 (SING); pass

to Ralan, Curtis 934 (CAL type of M. racemosa, K, SING); Waterfall

Garden above lily pond, Nauen, sine data (SING).

PERAK: Scortechini 619b (DD, K)} Ulu Bubong, King Nos. 10064

(G, K, L) type material of M. canarioides; 10194 (BM, FI, G, K)

type material of M. canarioides; 10816 (BM, K) type material of M.

eg 10845 (K, L) type material\of M. canarioides; 10562

SELANGOR: Kepong Plantation, Symington F.D. 30138 (KEP).

Matacca: Maingay 1298 (K, L) type material of M. canarioides;

Alvins 938 (SING); Lubok Kadondong, Ridley 3313 (K, SING) under

Maesa coriacea.

JoHorRE: Sungei Sedili, Ngadiman S.F.N. 36924 (SING); Mawai,

Negadiman S.F.N. 34742 (SING).

SINGAPORE: Bukit Mandai, Ridley 6324 (SING) type of Embelia

ridleyi; Ridley 8906 (SING); 11th mile Bukit Mandai, Ridley 8426

(SING); Choa Chu Kang, Ridley, 7th May, 1895 (SING); Bukit

Timah, Ridley 6355 (CAL, K, SING); Sungei Jurong, Ridley 6095

(SING); Stagmount, Ridley 11270 (K, SING); Sungei Loyang, Mat

or Ridley 6693 (SING).

DisTRIBUTION: Chitagong, Burma, Indo-China, Andamans, Malaya,

Sumatra, Java, Borneo and Philippines.

391

Gardens Bulletin, S.

H. macrocoma var. canarioides, the only monoecious Horsfieldia

species in Malaya, should be easily distinguished from the others

by the spreading perianth lobes and the elliptic, stalked fruits,

borne on a long branched inflorescence.

H. macrocoma is a polymorphic species and includes several

new synonyms. As there is some considerable range in size of

fruit, number of veins in the leaf, and in the nature of the

tomentum of the inflorescence, it seems best to split this species

into a number of varieties. I have left some of these synonyms

under the title “H. macrocoma aggregate”, as it is not possible in

the present treatise to assign every synonym to its correct variety

because of insufficient material, but some suggestions are given as

to where they might go. It is hoped that some finality may be

achieved when the requisite material from Leiden and other

herbaria is studied.

H. canarioides, originally described from Malaya, was kept

separate from H. papillosa (Java) and H. prainii (Andamans) on

the very trifling character, the length of the androecium stalk. This

varies from sessile to 0:25 mm. long in H. canarioides and up to

0-8 mm. long in the other two and flowers have to be boiled and

examined by a lens in order to see it. Plants with sessile androecia

occur in Sumatra, Borneo and Java as well as in Malaya and (one

record) Rahmat-Si Boeea 7564 from Sumatra has a stalk 0-8 mm.

long. I do not see, therefore, that the size of the androecium stalk

is a good character for separation of H. papillosa and H. prainii

and have united them under var. canarioides. A Bornean plant

which I describe as H. macrocoma var. rufirachis has a rusty

tomentose inflorescence axis, and its fruits are of the same size as

var. canarioides. The var. macrocoma from the Moluccas, differs

from var. canarioides in the smaller fruits, more veins in the leaf

and in the pubescent inflorescence axis. H. leptocarpa and Gymna-

cranthera ibutii also from the Moluccas are probably not different

from var. macrocoma. H. trifida from New Guinea is near var.

macrocoma but may have to go into a separate variety. I have

seen at Kew an undescribed species, Henry 12234 from Yunnan

which may be macrocoma or an allied species.

H. macrocoma (Miq.) Warb., Monog. Myrist. (1897) 299 T. 21. Figs

1—-6—Aggregate.

Basionym: Myristica macrocoma Mig. in Ann. Mus. Bot.

Lugd.-Bat. 1 (1864) 207; Mig. Ann. 2 (1865) 49 excl. spec.

celeb.. = H. irya.

Synonyms: Horsfieldia trifida A. C. Smith in Journ. Arn. Arb.

22 (1941) 60. H. oblongata Merr., Mgf. in Bot. Jahrb. 67

(1935) 148 quoad spec. Nov.-Guin. tantum.

392

Vol. XVI. (1958).

var. macrocoma

Probably includes the following synonyms:—

Horsfieldia leptocarpa Warb., Monog. Myrist. (1897) 346 T.

21 (excl. spec. Foster. Celeb. = H. irya). Gymnacranthera

ibutii Holth. in Blumea 5, 1 (1942) 183. Pig. 4.

DISTRIBUTION: Moluccas (Halmaheira-type locality).

var. rufirachis J. Sinclair, var. nov.

A typo inflorescentibus rufo-tomentosis, fructibus maioribus

differt.

Arbor 20 m. alta. Folia 14-32 cm. longa, 5S—11 cm. lata; nervi c.

20 pares. Flores masculi 1-5-2 mm. longi; stipes androecii 0-5 mm.

longus. Fructus 5—6 cm. longus, 2-8-3 cm. latus.

BoRNEO: One mile from Chin Lik’s camp to jetty, North Borneo

Timber Co. Concession Area, Kretam, Lahad Datu, Wood A4770

(K, L, SING holotype); Sepilok F.R., Sandakan, Wood A1983

(SING); Bettotan, Sandakan, Puasa 4644 (K, SING); Beaufort

Hill, Wood SAN 16838 (SING).

(5) H. superba (Hk. f. et Th.) Warb., Monog. Myrist. (1897)

295; Gamble, Mat. F-M.P. 5, 23 (1912) 209; Ridley, F.M.P. 3

(1924) 55; Corner, Wayside Trees of Malaya 1 (1940) 476.

Basionym: Myristica superba Hk. f. et Th., Fl. Ind. (1855)

162; A. DC., Prodr. 14, 1 (1856) 194; Miq., Fl. Ind. Bat. 1, 2

(1858) 62; Hk. f., Fl. Br. Ind. 5 (1886) 150; King in Ann.

Roy: Bot. Gard. Calc. 3 (1891) Pl. 124 bis, 125 bis.—

Fig. 36. Plate XB.

Tree 13-30 m. high Bark greyish brown, hard, longitudinally

fissured, the ridges flat; sap red, copious. Twigs light to medium

brown with numerous paler lenticels, rough, the younger parts

densely covered with rusty-stellate scurf, the older glabrous. Leaves

coriaceous, dark green and glossy above, yellowish or brownish

green beneath, both surfaces in young leaves densely rusty-

stellate-tomentose, later glabrous on the upper surface except for

the midrib which eventually becomes glabrous, the lower thinly

and harshly stellate-pubescent, elliptic to elliptic-lanceolate or

oblanceolate, apex acute, base sub-acute or rounded and slightly

cordate; midrib slightly raised above when fresh, flush with the

upper surface when dry, raised on the lower surface; nerves 15—30

pairs, average 23, impressed on the upper surface, very prominent

and like ridges on the lower, oblique, parallel, interarching at the

margin; reticulations not visible; length 25—70 cm.; breadth 10—22

cm.; petiole very stout, 2 cm. long, covered with rusty scurf.

Flowers yellow-ochre smelling of ripe pears. Male in branched, 10-

15 cm. long panicles from the axils of fallen leaves, the branches

numerous, 2—5 cm. long, all rusty-tomentose with numerous, very

early deciduous bracts; flower clusters sub-umbellate; pedicels 3—5

mm. long, glabrous; perianth 7-8 mm. long, ellipsoid or slightly obo-

void, obtuse in bud, glabrous, coriaceous, split down 1/3 way by

393

Gardens Bulletin, S.

Fig. 36. Horsfieldia superba (Hk. f. et Th.) Warb.

A, Leaf. B, Female inflorescence. C, Ovary. D, Young fruit in longi-

tudinal section. E, Male inflorescence. F, Staminal. column. A—D

from Sinclair S.F.N. 40047. E-F from Corner S.F.N. 36134.

394

Vol. XVI. (1958).

the 3, sometimes 4 teeth; androecium sessile, 4-5 mm. long, obtuse

at the apex with a very shallow depression there; anthers 16—20,

touching each other. Female flowers on short, stout, woody, 2—5

cm. long racemes, the pedicels very thick and stout, 5 mm.— | cm.

long and 3—4 mm. thick, green; perianth 9 mm.— 1:2 cm: long and

8 mm. broad, coriaceous, glabrous, 3, often 4 toothed and split

down 4 way by the acute teeth; ovary 4-5 mm. long, ovoid,

glabrous, grooved, stigma 2-lipped. Fruit ovoid-globose, at first

thinly covered with harsh, rusty scurf, becoming glabrous, the

circumferential groove of dehiscence prominent, length 7-9 cm.

and breadth 5:5—6-8 cm.; pericarp thick and fleshy, 2—2:5 cm. thick;

stalk stout, 1:8 cm. long, nearly glabrous. Aril entire, wrinkled at

the top and 2:5 mm. thick. Seed 5 cm. long, smooth.

PENANG: Phillips (K) holotype; Waterfall, Curtis 2966 (K, SING).

PERAK: Scortechini (BM, CAL) no data; near Gunong Batu Puteh,

King 8024 (CAL, K, L); D.F.O., Kinta K.F. Nos. 54655; 54657; 54659;

54663; 54675; 54700; 54713; 54721; 54725; 54816; 54818; 54820;

54824; 54837; 54846 and 54849 (all KEP).

TRENGGANU: Bukit Kajang, Kemaman, Corner, 1st and 8th Noy.,

1935 (SING); 362% miles Jerangau Road, Dungun, Sinclair & Kiah

S.F.N. 40496 (E, K, SING).

SELANGOR: Ulu Selangor, Goodenough 10526 (CAL, K, SING);

Weld Hill, Hamid, 20th May, 1918, tree No. 5 (SING); Omar F.D.

7973 (SING); F.D. C.F. 805 (KEP, SING); Sungei Lalang F.R.,

Md. Yattim K.F.N. 74246 (KEP); Ahmad 4867 (K, SING).

NEGRI SEMBILAN: Sungei Menyala F.R., Port Dickson, Motan K.F.N.

70473 (KEP).

MaLacca: Hervey, date 1891 (CAL, K, SING); Jasin-Chabau Road,

Hervey or Holmberg 2100 (CAL, K, SING); Kesang Tua, Goodenough

1279 (SING); 14th mile Sungei Udang F.R., Sinclair 40570 (BO, E,

K, L, SING).

JoHORE: 7th mile Kota Tinggi-Mawai Road, Corner S.F.N. 28763

(BM, K, SING); Bukit Kuing, Sedili Kechil, Corner 23rd June, 1934

(SING).

SINGAPORE: 114 miles Mandai Road by the shore of Seletar Reser-

voir, Sinclair S.F.N. 39538 (SING); Bukit Timah F.R., Ngadiman

S.F.N. 36141 (DD, K, KEP, SING); Reservoir Jungle, Corner S.F.N.

36134 (BM, KEP, SING); Sungei Benkang, Ridley 2101 (CAL,

SING); Tanjong Pasir Laba, Sinclair S.F.N. 40174 (E, K, SING).

CULTIVATED: Arboretum, Botanic Gardens, Singapore, Nur, 11th

June, 1924 (SING); Sinclair S.F.N. 40047 (B, BK, BM, BO, DD,

Delhi Univ., E, K, L, M, P, PNH, SAN, SING); near store, Botanic

Gardens, Singapore, Ahmad, June 1926 (SING).

DISTRIBUTION: Confined to Malaya.

This fine species has larger leaves and flowers than any of the

other Malayan species of Horsfieldia. It is a larger edition of H.

fulva, both species having twigs with pink lenticels but in fulva the

twigs are more slender. Large juvenile-leaved trees of fulva may be

mistaken for H. superba, but the leaves of fulva are glabrous. King

says the fruit is warted. This is the case only in dried specimens,

2395

Gardens Bulletin, S.

but when fresh, the fruit, like that of other Horsfieldia species, is

always smooth. The leaves resemble those of Gymnacranthera

bancana, but the undersurface in superba has a harsh covering of

stellate hairs and not the smooth feel of the former which has

stellate scales. When fresh, the upper midrib is slightly raised, but

in G. bancana it is level with the surface and sinks on drying.

(6) H. fulva (King) Warb., Monog. Myrist. (1897) 297; Gamble,

Mat. F.M.P. 5, 23 (1912) 210; Ridley, F.M.P. 3 (1924) 56.

Basionym: Myristica fulva King in Ann. Roy: Bot. Gard. Calc.

3 (1891) 297 Pl. 124.—Fig. 37.

Tree 12-15 m. high. Bark yellowish brown, thin, fissured

longitudinally but not flaking, fissures shallow; imner bark orange;

wood white; sap watery, pale pink, not copious. Twigs rough,

rusty-brown, greyish brown when older, at first pubescent, later

glabrous. Leaves very coriaceous, dark green or medium green

above, midrib pale green, lower surface paler green with a few

rusty-brown, harsh, stellate hairs present on the yellowish green

midrib, drying pale yellow-green above and light brown beneath,

elliptic-oblong, often obovate, acute at the apex and at the base;

nerves 12-14 pairs, straight, fine above, more prominent beneath,

anastomosing faintly at the margins, reticulations not visible, length

variable, 13-24 cm.; breadth 4-11 cm.; petiole 1-2 cm. long,

slightly pubescent. Male inflorescence a much branched, rusty-

stellate-pubescent panicle up to 10 cm. long, the branches slender.

Male flowers orange, 3—S mm. long, on slender pedicels 2 mm.

long; perianth obovoid, glabrous, divided for 4 of its length into 3

broadly triangular teeth; androecium 2-3 mm.- long, sessile,

cylindric, slightly apiculate and with 10 connate anthers. Female

inflorescence much stouter, about 2 cm. long, mostly unbranched,

bearing 6—8 flowers. Female flowers more elongated than the male,

less obovoid, widest at the middle, 3-lobed, 5—6 mm. long; pedicels

stouter, 3 mm. long; ovary sessile, glabrous, 2-5-3 mm. long, with

sessile, bilobed stigma. Fruit 1-2 on each female inflorescence,

glabrous, yellow, ovoid, slightly pointed, 2-5-3 cm. long and 2:5

cm. broad, dehiscing into 2 divaricating valves, the perianth

persisting as a collar slightly below the fruit; stalks 5 mm. long.

Seed broadly ovoid; testa thin, pale, covered by the reddish-orange,

entire aril.

PERAK: Scortechini 184a (BM, CAL, G, K, L) type material; Pulau

Lalang, Sembilan Islands, Wyatt-Smith, K.F.N. 76531 (KEP); Chior

F.R., Kinta, Kochummen K.F.N. 80627 (KEP, SING).

SELANGOR: Near Forest Research Institute, Kepong (3 trees prob-

ably planted), Ja’amat & Tachun F.D. 56304 (KEP); Walton F.D.

44933 (KEP); Symington F.D. 51845 and 51846 (KEP); Awang

K.F.N. 52079 (KEP); Wyatt-Smith, 15th Sept., 1950 (KEP); Sinclair

396

Vol. XVI. (1958).

Fig. 37. Horsfieldia fulva (King) Warb.

A, Leafy twig. B, Male inflorescence. C, Male flower. D, Staminal

column. E, Fruit. A from Sinclair S.F.N. 40170. B from King’s

Plate 124. C-D from Wyatt-Smith 15-9-50, Forest Plantation, Ke-

pong. E from Kochummen K.F.N. 80627.

397

Gardens Bulletin, S.

S.F.N. 40170 (BK, BM, BO, DD, E, K, L, P, SAN, SING); Ulu Lan-

gat, Symington F.D. 51757 (KEP); Gunong Moyang, Ulu Selangor,

Symington F.D. 56704 (KEP). |

MALACCA: Maingay 1304 (CAL, K) type material.

DISTRIBUTION: Confined to Malaya.

This species is nearest to H. superba. See notes under the latter.

H. fulva var. paludicola is H. crassifolia. See notes under crassi-

folia.

(7) H. flocculosa (King) Warb., Monog. Myrist. (1897) 297;

Gamble, Mat. F.M.P. 5, 23 (1912) 210; Ridley, F.M.P. 3

(1924) 55.

Basionym: Myristica flocculosa King in Ann. Bot. Gard. Calc.

3 (1891) 302 Pl. 131.—Fig. 38.

Tree 12-20 m. high with spreading branches. Tomentum of

twigs, leaves petioles, inflorescence axis and its bracts very dense,

soft, yellowish or light brown, floccose on young leaves, the hairs

1 mm. long or less, stellate or branched with barbules. Twigs stout,

densely tomentose, the bark rugose and cracking in the older

portions. Leaves coriaceous, oblong-lanceolate or oblanceolate,

both surfaces at first covered with tomentum, the upper, including

the midrib, soon glabrous, slightly rough with numerous papillae,

the lower soft with persistent tomentum, apex acute, base rounded or

slightly cuneate, margins slightly revolute when dry; nerves 15—20

pairs, impressed on the upper surface, bold and prominent on the

lower, straight at first and then curving and interarching close to

the margins; reticulations most often invisible, occasionally a few

showing here and there on the lower surface; length 30-45 cm.;

breadth 10-18 cm.; petiole 1-2 cm. long, very stout, tomentose.

Male inflorescence a much branched, densely tomentose panicle,

with deciduous bracts (their length 1-5—1-7 cm. and their breadth

4 mm.— 1 cm.), the main axis 14—20 cm. long, the main branches

5 mm.— 1-5 cm. long and the ultimate 3-4 mm. long, bearing

numerous, glabrous, obovoid flowers on 3-4 mm. long, slender

pedicels. Male perianth 3 mm. long, the lobes broadly ovate, split

down half-way, thin, membranous, reticulate and covered with

minute, hyaline, round spots, the circumferences of which are opa-

que and brown; androecium sessile, obovoid, truncate at the apex, 2

mm. high and 1-1-5 mm. broad with about 10 anthers, the apices of

which are incurved at the top. Female inflorescence 2:5 cm. long,

not or slightly branched, bearing immature fruit; female flowers and

mature fruit unknown. Young fruit (will probably not grow much

larger) sub-globose to slightly ellipsoid, glabrous, the ridge of

398

Vol. XVI. (1958).

Fig. 38. Horsfieldia flocculosa (King) Warb.

A, Leafy twig with young male inflorescences. B, Male inflorescence.

C, Male flower. D, Staminai column. E, Staminal column from

above. F, Immature fruit. A from Kiah S.F.N. 32314. B-E from

Abdul Rahman C.F. 368. F from Burkill & Haniff S.F.N. 16394.

399

Gardens Bulletin, S.

circumferential dehiscence prominent, length 3 cm.; breadth 2:5

cm.; pericarp 5 mm. thick and stalk 5 mm. long. Seed covered by

the aril.

SELANGOR: Ulu Kerling, King 8618 (BM, CAL holotype, FI, G, K,

KEP, L, P); Ulu Gombak, Burkill & Haniff S.F.N. 16394 (K, SING);

Weld Hill, Abdul Rahman C.F. 368 (SING); 21st mile Ginting Sim-

pah, Strugnell F.D. 12729 (KEP, SING).

JoHORE: Sungei Kayu, Kiah S.F.N. 32314 (SING).

DISTRIBUTION: Confined to the Malay Peninsula.

This species is in a group with grandis, tomentosa and motleyi.

It is nearest to grandis and the following table will show the diffe-

ences between

Tomentum

Twigs

Leaves

Flowers

the two.

H. grandis

Darker, longer, hairs up to

2 mm. long, less dense,

harsh.

Bark not tending to crack.

Thinly coriaceous, almost

bullate, harsh and pu-

bescent above including

the pubescent midrib.

Tomentum of lower sur-

face less dense than

flocculosa, harsh; reticu-

lations very distinct on

both surfaces.

Smaller, 1-2 mm. long,

globose; perianth cover-

ed with minute black

dots; staminal column 1

mm. high and 4-1-5 mm.

broad with 13-15 an-

thers; male pedicels 1

mm. long.

H. flocculosa

Lighter, more _ yellowish,

shorter, hairs 1 mm. long

or less, denser, soft.

Bark tending to crack.

Coriaceous, not bullate, less

harsh and quite glabrous

above. Tomentum of

lower surface, denser,

soft; reticulations mostly

invisible, sometimes a

few, scattered, faint ones.

on the lower surface.

Larger, 3 mm. long obo-

void; perianth covered

with circles which are

hyaline in the centre and

brown round the circum-

ference; staminal column

2 mm. high and 1-5 mm.

broad with about 10 an-

thers; male pedicels 3—4

mm. long.

(8) H. grandis (Hk. f.) Warb., Monog. Myrist. (1897) 301;

Gamble, Mat. F.M.P. 5, 23 (1912) 211; Ridley, F.M.P. 3

(1924) 56.

Basionym: Myristica grandis Hk. f. in Trans. Linn. Soc. 23

(1860) 157.

Synonym: M. rubiginosa King in Ann. Bot. Gard. Calc. 3

(1891) 302 Pl. 130—Fig. 39.

400

Vol. XVI. (1958).

A Shae < = e \ F534 ae \ i,

ES i h ee oS, 7% i J ,

SR = ~*~ = J -

; S : Y

Seca - Ss sf x

2 wer, oo a> x

a . Gas Fx me 3 oD

= : = ~

—=S > SP = x

RTs FSS eS as : ;

22 24 = ee P

+ < 7 % me ~

= > 2 = — = ——

Bs ~ gn z SE, SJ = : -

; = eS) : >

nah st ES - : Ss

x. FS ~ — > SF

s Sy >

: SSS

2 =F arn .

2 & - .

e = . = :

: 3

> 2s

~ ”

1 mm

Imm D

Fig. 39. Horsfieldia grandis (Hk. f.) Warb.

A, Leafy twig with very young fruit. B, Male inflorescences. C. Male

flower. D, Perianth segment, inside view. E, Staminal column. F,

Female inflorescence. G. Female flower. H, Ovary. A from Neadi-

man S.F.N. 36831. B—E from Holttum S.F.N. 9404. F-H from

Ramhli SAR 158.

401

Gardens Bulletin, S.

Tree 5-10 m. high. Bark yellowish-brown, rough with closely

placed longitudinal striations 1 mm. deep or less, scarcely fissured.

Twigs, young parts densely tomentose with erect, 2 mm. long, rusty

hairs, older parts becoming glabrous, brown, striate. Leaves thinly

coriaceous, dark green above, slightly bullate and rough with

scabrid hairs which tend to get rubbed off leaving their harsh bases,

midrib permanently tomentose, lower surface densely rusty-

tomentose with the same kind of hairs as on the twigs, petioles and

upper surface of the leaves, these being stellate or branched with

short barbules, the whole resembling a tangled mass of fungal

threads or the filaments of the alga Rhizocolonium, elliptic-oblong,

apex rounded and then rather abruptly acute, base rounded and

slightly subcordate; nerves 16-19 pairs, oblique and then curving

and anastomosing at the extreme edge, impressed above, raised

beneath; reticulations very distinct, impressed above, raised be-

neath, a scalariform set at right angles to and between the nerves,

a second lax network mingling with the scalariform ones; length

25—40 cm.; breadth 10-20 cm.; petioles 7 mm.—1 cm. long, densely

rusty-tomentose. Male inflorescence a much branched, rusty-temen-

tose panicle, the main axis 12—25 cm. long, the main branches 2-3

cm. long and the ultimate branches 7 mm.—1 cm. long, bearing nu-

merous, glabrous, yellow, globose flowers on 1 mm. long, slender

pedicels. Male perianth 1-2 mm. in diam., split down half way into

3, sometimes 4 broadly ovate lobes, thin, black-dotted when dry;

androecium sessile, depressed in the centre, 1-1-5 mm. in diam. and

1 mm. high; anthers 13-15. Female inflorescence 1-5—2:5 cm long,

unbranched or with a few stout branches 7 mm. long; perianth

yellow, coriaceous, globose, glabrous, 2 mm. in diam., sessile;

ovary globose, glabrous with 2-lipped stigma. Fruit very young on

the above described inflorescence, sub-globose, glabrous, 5 mm. in

diam., the circumferential groove of dehiscence prominent and the

perianth persistent; stalks 3 mm. long.

JoHORE: Mount Austin Woods, Ridley, date 1906 (SING); Kluang,

Holttum S.F.N. 9304 (K, SING); Sungei Sedili, Ngadiman S.F.N.

36831 (SING).

SINGAPORE: King 1232 (CAL, K, L) type of M. rubiginosa; Mount

Faber, Ridley 4827 (SING); Seletar, Ridley, date 1894 (SING); Bo-

tanic Gardens’ Jungle, Ridley 4133 (CAL, K, SING); Liane Road,

Botanic Gardens’ Jungle, Burkill, 20th March, 1921 (SING) the tree

appears to be extinct now; Mac Ritchie Reservoir, Sinclair 9363 (A,

B, BM, BRI, E, K, L, M, NY, P, SING).

DISTRIBUTION: Borneo and Sumatra.

Type: Low s.n., Borneo (K) holotype of M. grandis.

402

Vol. XVI. (1958).

(9) H. tomentosa Warb., Monog. Myrist. (1897) 302; Gamble,

Mat. F.M.P. 5, 23 (1912) 212; Ridley, F.M.P. 3 (1924) 56.

Basionym: Myristica tomentosa Hk. f. et Th., Ind. I (1855)

161 (non BI., nec Thb. nec Grah.) et Fl. Br. Ind. 5 (1886) 105;

A.DC., Prodr. 14, 1 (1856) 204; Mig., Fl. Ind. Bat. 1, 2 (1858)

68; King, Ann. Roy. Bot. Gard. Calc. 3 (1891) 301 Pl. 129.

Myristica horsfieldia Wall. Cat. 6806 pro parte.—Fig. 40.

Tree 8—18 m. high with spreading branches. Twigs, the young

parts and the terminal bud densely and softly rusty-tomentose, the

hairs stellate and mixed with plumose or irregularly branched hairs,

the older parts glabrous, medium brown and distinctly striate.

Leaves membranous, elliptic, elliptic-obovate, or oblong-obovate,

narrowed to the acute or rounded base, rather abruptly acute at the

apex, dark green, shining and glabrous above except the midrib,

dark brown when dry, lower surface rusty brown, thinly stellate-

tomentose at first, the hairs tending to rub off later; midrib sunk

above, raised beneath; nerves 12—15 pairs, slender, sunk above,

prominent beneath, oblique, interarching 3-4 mm. from the margin;

reticulations not visible; length 12—25 cm.; breadth 5-12 cm.;

petiole 1-5—1-8 cm. long. Flowers waxy yellow in axillary branched

panicles, the male panicles 6-10 cm. long, the female 6 cm. long,

less branched, the branches shorter, both densely and softly

tomentose, the hairs light brown, erect, stellate or plumose. Male

flowers on slender, 2-3 mm. long pedicels; perianth thin, sub-

globose, minutely covered with raised dots when dry, 3-lipped, 2

mm. long, the lobes acute, reaching down less than 1/3 of the whole

flower; androecium sessile, depressed-globose, 2 mm. broad and

1 mm. high; anthers 10—15, average 10. Female flowers on

pedicels 2 mm. long; perianth as in the male but more coriaceous,

the lips sub-acute to blunt; ovary globose, sessile, tomentose; stigma

sessile with 2 obtuse lobes. Fruit 2—2:5 cm. long and 1:5 cm.

broad, ovoid, ridged, sub-glabrous, yellow; stalk 6 mm.—1l cm.

long, rather slender, often with a small collar where it joins the

fruit. Aril red, fleshy, entire, covering the seed.

KepaH: Yan, Ridley 5489 (CAL, SING); Koh Mai F.R., Kiah

S.F.N. 35134 (SING); Sungei Patani, Meh F.D. 10191 (EB).

PENANG: Gaudichaud, March 1837 (G, Prodr.); Wall. Cat. 9025

(BM, CAL, G, K holotype, SING); King 1551 (CAL); Phillips (K);

- Pulau Boetong, Curtis i748 (BM, CAL, K, SING); Waterfall, Curtis

1197' (BM, K, SING); Ridley 10787 (CAL, K, SING) and 10240 (K,

SING) and 7205 (CAL, K, SING): Haniff, 4th April, 1917 (CAL);

hill above swimming bath, Fox 13 (SING); Balek Pulau, Forest Guard

s.n. (K, SING).

Perak: King 7998 (CAL, DD, UPS); Chanderiang, King 5671 (BM,

CAL, FI, G, K, L); Ulu Bubong, King Nos. 10386 (CAL, E, SING)

and 10557 (BM, CAL, DD, FI, G, K, L, MEL): near Ulu Kerling,

403

Gardens Bulletin, S.

i Era 6 hee

Fig. 40. Horsfieldia tomentosa Warb.

A, Leafy twig. B, Male inflorescences. C, Male flower. D, Staminal

column. E, Female inflorescence. F, Female flower. G, Ovary. H,

Fruit. A-D from Burn-Murdoch 48. E-G from Nur S.F.N. 21751.

H from Ridley 3171. .

404

Vol. XVI. (1958).

King 8642 (CAL, K, SING); Larut, King 4165 (CAL, DD, K, MEL);

Gopeng, King 6102 (CAL, FI, G, K, L, P); Batu Togoh, Taiping,

Henderson F.M.S. Mus. Herb. 11604 (SING).

PAHANG: Cameron Highlands, Batten-Pooll, Nov. 1939 to Jan., 1940

(SING); S. Sat, Ulu Tembeling, Henderson S.F.N. 22061 (CAL, NY,

SING); Tembeling Henderson S.F.N. 21804 (NY, SING); Bukit Batu

Berendum, Pulau Tioman, Nur, S.F.N. 21751 (BM, DD, NY, SING);

3 miles south of Kuala Lipis, Burkill & Haniff 17167 (SING).

SELANGOR: Ulu Gombak, Ja’amat F.D. 25160 (SING); Weld Hill,

Burn-Murdoch 44 (CAL, SING) and 48 (SING); near Ulu Selangor,

King 8552 (CAL, E, FI, K).

MALAcca: Alvins 672 (SING) and 680 (SING); Sungei Jerneh, Derry

967 (CAL, K, SING); woods between Cleabana and Jassin, Ridley

3171 (CAL, SING); Gaudichaud 114 (P).

SINGAPORE: Bukit Timah, Cantley 30 (K) has never been seen in

Singapore since.

DISTRIBUTION: Siam, Sumatra (Rantau Parapat, Bila) Rahmat Si

Toroes 2242 (NY, SING) and (Silo Meradja, Asahan) Rahmat Si

Toroes 801 (S, SING). Borneo, Loa Haur, west of Samarinda, Koster-

mans 6859 (PNH).

This species is in a group. with H. grandis and H. flocculosa,

having a similar, punctate perianth. It is nearest to H. motleyi

from Borneo which has shorter flowering pedicels and larger

reticulate leaves, while in the present species reticulations are

almost invisible. When sterile, the tomentose bud of H. tomentosa

is a good distinguishing mark. The inflorescence axis, like that of

polyspherula, is tomentose but tomentosa has erect hairs of a

lighter brown colour while polyspherula has darker, shorter and

more scurf-like hairs.

(10) H. wallichii (Hk. f. et Th.) Warb., Monog. Myrist. (1897)

305; Gamble, Mat F.M.P. 5, 23 (1912) 213; Ridley, F.M.P. 3

(1924) 57.

Basionym: Myristica wallichii Hk. f. et Th., Fl. Ind. 1 (1855)

161; A. DC., Prodr. 14, 1 (1856) 203; Mig, Fl. Ind. Bat. 1,

2 (1858) 67; Hk. f., Fl. Br. Ind. 5 (1886) 105; King in Ann.

Roy. Bot. Gard. Calc. 3 (1891) 303 Pl. 132 and 133 (excl.

synon. crassifolia Hk. f. et Th.).

Synonym: M. horsfieldia Wall. (non. Bl.) Cat. No. 6806 pro

parte—Fig. 41. Plate XIA.

Tree 13-30 m. high with straight trunk and crown of branches

at the top. Bark greyish brown, hard and of no great thickness,

longitudinally striate with narrow striations 2—4 cm. apart, not

flaking; sap blood-red, not copious; wood cream-coloured. Twigs

dark grey, minutely pubescent, becoming glabrous, hollow in parts,

the younger tips smooth, the older portions striate or rough. Leaves

405

Gardens Bulletin, S.

TURAL Pte

18s

Fig. 41. Horsfieldia wallichii (Hk. f. et Th.) Warb.

A, Leafy twig with male inflorescences. B, Male flower. C, Staminal

column. D, Female inflorescence. E, Female flower. F, Ovary. G,

Cluster of fruit. H, Seed in cross section. I, Embryo, top view. J,

Embryo side view. A-—C from Sinclair S.F.N. 39487. D-F from

King’s Plate 133. G-J from Sinclair S.F.N. 39486.

406

Vol. XVI. (1958).

coriaceous, dark green, shining above, paler beneath, at first rusty-

stellate-pubescent on the midrib and veins beneath, soon glabrous,

lanceolate, oblong-lanceolate or oblong, apex acute, base cuneate

or rounded; midrib flat above at the base, prominent beneath;

nerves 16—22 pairs, straight, oblique, parallel, mterarching m a

distinct line near the margin, a second loop in the interarching bays

often visible; reticulations not visible when fresh, faint above and

below when dry, scalariform and rather widely spaced; length 15—

35 cm.; breadth 5—10 cm.; petiole 2 cm. long, stout, not grooved.

Male inflorescence rusty-stellate-tomentose, up to 33 cm. long,

thrice branched, branches of the first order 6-11 cm. long, of the

second 2—3 cm. and of the third 1-1-5 cm., bearing the flowers in

sub-umbellate clusters. Male flowers scentless, nearly sessile on

pedicels up to 1 mm. long, pear-shaped and glaucous green in bud,

yellow, when mature, fleshy, nearly glabrous, 3-4 mm. long. 3—

sometimes 4—toothed, the teeth acute, scarcely reaching half way

down; anthers 15-16, white, touching each other in a 3, sometimes

4-lobed sessile cup, their apices incurved, obtuse; some aborted

male flowers present, smaller than the normal and flowering earlier,

the anthers only partially developed, not fertile. Female inflore-

scence shorter, stouter, and with less branches and fewer flowers.

Female flowers larger, glabrescent, the teeth spreading: ovary

globose, sessile, glabrous; stigma sessile, linear. Fruit yellowish

green, later yellowish-ochre and sometimes with a pink tinge,

ellipsoid, pointed at each end, smooth, glabrous, 7—8 cm. long and

5—6 cm. broad; the circumferential ridge prominent; pericarp 1-1-5

cm. thick; stalk 5 mm. — 1 cm. long, thick. Ari/ a rich orange, cons-

isting of 2 layers and wrinkled and the apex. Seed 4 cm. long with

a pale yellow outer, fleshy layer and an inner dark brown one and

sinking in water when fresh.

KeEDaH: Teloi F.R., Batu Besar, Md. Salleh bin Yusoff KF .N. 60474

(KEP).

PENANG: Waterfall Gardens, Hanif? S.F.N. 3660 (SING).

PROVINCE WELLESLEY: Kubang Ulu Reserve, Cwrtis 2423 (BM,

SING).

PERAK: Scortechini 246a (CAL, G, L): Gopeng, King 4827 (BM,

CAL, FI, K): Kinta, D.F.O., K.F. Nos. 54814 (KEP) and 54821

(KEP).

TRENGGANU: 26th mile Kuala Trengganu-Besut Road. Sinclair &

_Kiah S.F.N. 40484 (E, K, P, SING).

SELANGOR: 20th mile Ginting Simpah, Strugnell F_.D. 12664 (SING):

16th mile Ulu Gombak, Strugnell 12638 (SING) mixed, fruits of both

H. wallichii and M. maxima mounted on the same sheet; Bukit Cheraka

F.R., Ulu Selangor, Wyatt-Smith K.F.N. 70357 (KEP): Sungei Lalang

F.R., Md. Yattim K.F.N. 66860 (KEP); Sungei Buloh F.R., Ja’amat

F.D. 15265 (KEP); Rantau Panjang F.R., Strugnell F.D. 12477

(KEP).

407

Gardens Bulletin, S.

MALACCA: Griffith (K) type material; Maingay 1284 (CAL, K, L); -

Alvins 657 (SING).

JOHORE: Rengam F.R., Kluang, Cousens K.F. Nos. 69777 (KEP)

and 69791 (KEP); 6th mile Tg. Laboh, Batu Pahat, Sulaiman bin

Manja K.F.N. 70180 (KEP).

SINGAPORE: Wall. Cat. 6806 (K) in part, type material; Maingay

1283 (CAL, K, L); Murton 1481 (K); Cantley s.n. (SING); Mac

Ritchie Reservoir, Corner §.F.N. 33556 (BM, BRI, K, KEP, SING)

and Corner, 15th March, 1937 (SING); Sinclair S.F.N. 40216 (BO,

E, K, L, SING); Reservoir Jungle, Merah, 20th May, 1937 (SING);

Bajau, Ridley, date 1892 (SING); Chan Chu Kang, Ridley 5061 (CAL,

K, SING); behind carpenter’s shed, Botanic Gardens, Ridley 4422

(CAL, SING).

CULTIVATED: Arboretum, Botanic Gardens, Sinclair §.F. Nos. 39486

(BO, DD, E, K, L, P, SING) and 29487 (BK, BO, DD) Ea

PNH, SAN, SING); Ngadiman S.F.N. 34520 (K, SING); Corner S.F.

Nos. 33140 (SING) and 34439 (K, KEP, SING).

DISTRIBUTION: Borneo and Sumatra.

This species resembles M. maxima in its leaves and venation. In

maxima the leaves are always glabrous beneath but young leaves

of H. wallichii are slightly pubescent on the lower midrib and lower

nerves. The younger parts of the twigs are hollow in H. wallichii

but solid in M. maxima.

(11) H. penangiana J. Sinclair, sp. nov.—Fig. 42.

Haec species ad Horsfieldias cum perianthio 3-lobato, androecio

in sectione transversa rotundo, antheris apice non liberis pertinet,

sed inter eas foliis minoribus, nervis subtus non elevatis facile dis-

tinguitur. In aspectu H. ridleyanae haec species valde similis, sed

illa inter Horsfieldias cum androecio in sectione transversa tri-

angulari, antheris apice liberis ponenda; etiam foliis illius in sicco

viridibus vel luteo-viridibus, subtus nervis leviter elevatis.

Arbor 8-16 m. alta. Ramuli graciles, griseo-brunnei, glabri,

striati. Folia chartacea, glabra, supra nigro-brunnea subtus saturate

{erruginea, elliptica vel elliptico-lanceolata, basi apice acuta, 7-13

cm. longa, 2:5—3:5 cm. lata; nervi 8-10 pares utrinque graciles

non prominentes, supra depressi, subtus non elevatis, sensim

adscendentes; reticulationes invisibiles; petioli graciles, c. 1 mm.

longi. Inflorescentia mascula ad 6 cm. longa, primo ferrugineo-

pubescens, deinde glabra breviter pluri-ramosa; bracteae deciduae

minutae, acutae, ferrugineo-pubescentes, 1 mm. longae. Flores

masculi glabri, globosi 3-dentati, 1-1-5 mm. longi; pedicelli apice

incrassati; androecium sub-globosum vel cylindricum, 0-8 mm.

longum; antherae 7-10 apice obtusae, connatae. Flores feminei et

fructus non visi.

Tree 8-16 m. high. Twigs slender, greyish brown, glabrous,

striate. Leaves chartaceous, drying blackish brown above and dark

408

Vol. XVI. (1958).

4 cm aa bon

Fig. 42. Horsfieldia penangiana J. Sinclair.

A, Leafy twig with male inflorescences. B, Male flower. C, Staminal

column. D, Staminal column, top view. A—D from Curtis 2406.

409

Gardens Bulletin, S.

rusty-brown beneath, elliptic or elliptic-lanceolate, apex and base

acute; nerves 8-10 pairs, fine and faint on both surfaces, sunk

above, ascending gradually, reticulations invisible; length 7-13

cm.; breadth 2-5-3-5 cm.; petiole slender, about 1 cm. long. Male

inflorescence up to 6 cm. long with several short, rusty-pubescent

branches, becoming glabrous; bracts deciduous, minute, acute,

rusty-pubescent, 1 mm. long. Male flowers glabrous, globose, 3-

toothed, 1-1-5 mm. long on a stalk 1 mm. long which is thickened

towards the base of the flower; androecium sub-globose to cylindric,

0-8 mm. long; anthers 7-10, apices slightly obtuse and not free.

Female flowers and fruit not seen. ;

PENANG: Moniot Road, Curtis 2458 (CAL, SING); between coolie

lines and Experimental Nursery, Curtis 2406 io CAL, K, P, SING,

holotype).

DISTRIBUTION: Confined to Malaya.

These two numbers of Curtis were quoted by King under

Myristica griffithii, which is a synonym of Gymnacranthera

eugeniifolia var. griffithii. They clearly belong to a Horsfieldia on

account of the striate twigs and the structure of the flower and I

have named them H. penangiana.

H. penangiana resembles Gymnacranthera eugeniifolia very

much in the leaves which have faint, non-raised nerves but the

similarity stops there. Hence the reason why previous authors

have not noticed that it is a Horsfieldia. H. penangiana is very

like H. ridleyana but the leaves of the latter dry olive-green and

the nerves, although faint on the lower surface, are raised. In

penangiana the androecium is circular in cross section while the

anthers are slightly obtuse and united at the tips. In ridleyana

the androecium is entirely different, being triangular in cross

section or trigonous with erect anthers which are free and acute at

the tips.

(12) H. subalpina J. Sinclair, sp. nov.

Synonym: H. sucosa King quoad Wray 467.

Species affinis H. glabrae et H. macrothyrsae. Ab hac fructibus

maximis ab illa floribus minoribus, pedicellis masculis longioribus

et tenuioribus, ab utrisque nervis supra depressis, differt.

Arbor alta. Ramuli apice laevi, in gemma elongata puberula vel

pubescenti terminati, in partibus senioribus valde striati lenti-

cellati. Folia chartacea, glabra, oblongo-lanceolata, supra viridius-

culi-nigra, subtus modice brunnea, ad 21 cm. longa et 6 cm. lata,

basi cuneata, apice acuta; nervi c. 18 pares supra tenues, depressi,

410

Vol. XVI. (1958).

subtus prominentes, obliqui, prope marginem anastomosantes;

reticulationes supra fere indistinctae paucae, subtus invisibiles;

petioli 1:8 cm. longi, glabri, supra alte canaliculati. [nflorescentia

mascula ad 9 cm. longa, ramosa, puberula mox glabra. Flores

masculi 2 mm. longi cum pedicellis tenuibus 20 mm. longis;

perianthium glabrum, tenue, apice obtusum 3-dentatum aliquando

2-dentatum; androecium oblongum, 1:5 mm. longum, 11-13

antheris connatis praeditum. Flores feminei non visi.

PERAK: Gunong Batu Puteh, Wray 467 (BM, CAL, K holotype, L).

DISTRIBUTION: Borneo.

A mountain species of altitude 1,130 m, or more and of which

more material is badly wanted. The sunk veins on the upper surface

of the leaf may be a good diagnostic character and it is desirable

to know if they are always sunk. Kostermans 7414 (L,SING)

from the Peak of Balikpapan, East Borneo, has leaves and twigs

similar to the Malayan plant. The orange-brown fruit is large, 8 cm.

x 6:5—7 cm. The red aril covers the seed and is slightly fimbriate at

the apex. The seed is 4 cm. long and 2:6 cm. broad.

H. subalpina belongs to a group of closely allied species, namely

H. macrothyrsa, glabra, valida and costulata. Unfortunately the

flowers of H. valida (Sumatra) are unknown, but the fruit is large

about the size of a goose’s egg (Teijsmann) and the leaves agree

with H. macrothyrsa (also Sumatra) both having raised veins on

the upper surface. H. costulata (Celebes) is also very similar and

has raised veins on the upper surface of the leaf but the flowers

are smaller than in macrothyrsa. The flowers of glabra (Java and

Sumatra) are smaller than that of macrothyrsa and the veins of

the leaf are neither raised or sunk above. Its fruit is smaller than

that of subalpina (Balikpapan) and there is no trace of any hairs

on the inflorescence axis.

(13) H. glabra (Bl.) Warb., Monog. Myrist. (1897) 313 T. 21

Figs 1-2.

Basionym: Myristica glabra Bl., Bijdr. (1825) 576 et

Rumphia 1 (1835) 191 T. 64 Fig. 1 (non Hk. f. et Th., nec

King?); A. DC., Prodr. 14, 1 (1856) 202 quoad sp. javan.;

Mig., Fl. Ind. Bat. 1, 2 (1858) 65 quoad sp. javan.; Koorders

et Valeton, Mededeel. uit’s Land’s Plantentiun 17 (1896) 181.

Synonyms: M. globularia Mig. (non Bl. nec Lam.), Pl.

Junghuhn. (1852) 171. M. laevigata Miq. (non Bl. ) Fi. Ind.

Bat. 1, 2 (1858) 65 pro parte quoad specimina javanica.—Fig.

43.

411

Gardens Bulletin, S. —

Fig. 43. Horsfieldia glabra (B1.) Warb.

A, Leafy twig with male inflorescence. B, Male flower. C, Staminal

column. D, Fruit. A-C from Symington F.D. 36245. D from

Holttum S.F.N. 8679.

412

Vol. XVI. (1958).. ;

Tree 6-12 m. high with Garcinia-like branching. Twigs medium

brown, very rough, striate, glabrous and covered with numerous

raised lenticels. Leaves thinly coriaceous, drying dark brown

above and pale brown beneath, oblong-obovate or oblanceolate-

elliptic, glabrous, base acute and sometimes slightly decurrent on

to the petiole, apex obtuse or rounded; nerves 12-15 pairs, fine

above and level with the surface of the leaf or very slightly raised,

fairly prominent beneath, oblique, curving slightly and interar-

ching faintly near the margin; length 9-18 cm.; breadth 4—6 cm.;

petiole 1 cm. long, drying blackish, deeply grooved. Male inflore-

scence 2—6:5 cm. long with 3—6 short lateral racemose branches,

each, 1-1-5 cm. long. Male flowers 1-5-2 mm. long, oblong,

obtuse in bud, nearly glabrous; pedicels 1—1:5 mm. long; androe-

cium sessile, oblong, obtuse at the apex; anthers 6—7, the apices

blunt and not free. Female inflorescence shorter than the male, 4

cm. long, with a few short branches. Female flowers ovoid-

globose, with short, thick pedicels; ovary glabrous; stigma sessile.

Fruits 1-2, (those seen immature) globose, glabrous, succulent,

2 cm. in diam.; stalks 5 mm. long.

PAHANG: Pine Tree Hill Path, Fraser’s Hill, Purseglove 4212 (BO,

_E, K, SING); Fraser’s Hill, Burkill & Holttum S.F.N. 8679 (K, L,

SING); Sungei Pahang, Cameron Highlands, Symington F.D. 36245

(SING).

DISTRIBUTION: Java and Borneo.

TyPE MATERIAL: Java, s.l., Blume 2160 ‘L); 2160b (L); 2206 (L).

In the group with H. bracteosa, amygdalina and kingii. It is a

new record for Malaya. It agrees with the Javan material although

the male inflorescence does not always reach 12 cm. long as stated

by Warburg. H. glabra ascends from sea-level to 1300 m. in Java

and in Malaya is found at 1300 m. or over. The leaves recall those

of H. subglobosa and H. bracteosa. The staminal column is like

that of bracteosa being oblong and blunt at the apex with the

anthers not free. The description of the female flowers is taken

from Javan material.

(14) H. punctatifolia J. Sinclair, sp. nov.—Fig. 44. Plate XIB.

Haec species inter Horsfieldias cum signis sequentibus ponenda:

foliis glabris, perianthio 3-lobato, androecio cylindrico disciform1

vel oblongo et in sectione transversa rotundo, fructibus maximis.

In hac grege H. sucosae et H. bracteosae, proxima, sed fructibus

maioribus, foliis subtus nigro punctatis, ramulis ferrugineis haec

species differt. Ab H. bracteosae perianthio fructus deciduo

recedit.

413

Gardens Bulletin, S.

laQam | ew

Fig. 44. Horsfieldia punctatifolia J. Sinclair.

A, Leafy twig with fruit. B, Undersurface of leaf showing dots. C,

Fruit cut open to show aril. D, Aril and seed in longitudinal sec-

tion. E, Female inflorescences. F, Female flower. G, Ovary. H, Male ’

inflorescence, very young. I, Male flower, immature. J, Staminal

column. K, Staminal column from above. A-—D from Sinclair

S.F.N. 40211. E-G from Strugnell 12699. H-K from Alvins 8538.

414

Vol. XVI. (1958).

Arbor 13-30 m. alta. Cortex rubro-brunneus, fissuris propinquis

verticalibus leviter orantus non caducus; latex roseus aquosus

modice copiosus. Ramuli ferruginei, glabri, pustulati, striati. Folia

coriacea, glabra, oblongo-lanceolata, 10-21 cm. longa; 3-5-6 cm.

lata, utrinque acuta, supra in vivo atro-viridia cum nervis costaque

pallidioribus subtus pallidiora minute punctata, cum nervis costa-

que flavido-brunneis, in sicco supra griseo-brunnea subtus modice

brunnea; nervi c. 11 pares, supra tenues obliqui plani, subtus pro-

minentes, ad marginem anasomosantes; reticulationes invisibles;

petioli 1:5 cm. longi, glabri. Inflorescentia mascula ad 8 cm. longa

leviter ramosa, ramis brevibus 1 cm. longis. Flores masculi flavidi

globosi pubescentes 1:5 mm. in diam.; pedicelli 1 mm. longi;

perianthium 3-lobatum, lobis late-ovatis pluri-nervatis; androe-

cium sessile, antherae 7-10 cum apicibus acutiusculis. Inflore-

scentia feminea c. 3 cm. longa pauciflora, non ramosa. Flores

Jeminei coriacei 1—3 fasciculati, 3 mm. longi, 3-dentai; pedicelli

crassi 2 mm. longi; ovarium 1 mm. longum glabrum cum stigmate

minute 2-labiato. Fructus ellipsoideus utrinque acutis glaber laevis

6-9 cm. longus, 6 cm. in diam-, cum linea suturali prominenti;

pericarpium 2 cm. crassum flavido-roseum. Arillus aureo-ruber

apice paulo lobatus, semen omnino tegens. Semen 3:5 cm. longum,

2°5 cm. latum, laeve, modice brunneum.

Tree 13-30 m. high. Bark reddish brown, rough, longitudinally

fissured but not flaking, the fissures close and not deep; sap watery,

red, fairly copious. Twigs rusty-brown, glabrous, pustular, striate.

Leaves coriaceous, glabrous, dark green and glossy above with

paler green veins and midrib, paler green beneath and punctate

with minute black dots, and yellow brown midrib, drying greyish-

brown above and a rusty medium brown beneath, oblong-lanceo-

late, acute at apex and base; nerves about 11 pairs, fine and level

with the surface above, prominent beneath, oblique, interarching

at the margin; reticulations invisible; length 10-21 cm.; breadth

3-5-6 cm.; petiole 1:5 cm. long, glabrous. Male inflorescence

about 8 cm. long, slightly branched, the branches about | cm.

long. Flowers yellow, globose, pubescent, 1:5 mm. in diam.

on stalks 1 mm. long; perianth 3-lobed, the lobes broadly

ovate, several veined; androecium sessile, disc-shaped; anthers

7-10, slightly acute at their apices. Female inflorescence about 3

cm. long, unbranched, the flowers in groups of 1—3. Female

flowers coriaceous, 3 mm. long on a stout stalk 2 mm. long;

perianth with 3 short teeth; ovary 1 mm. long, glabrous; stigma

minutely 2-lipped. Fruit ellipsoid, pointed at both ends, yellowish

ted, glabrous, smooth, the dehiscence suture prominent; length

415

Gardens Bulletin, S.

6-9 cm.; breadth 6 cm.; pericarp 2 cm. thick. Aril orange-red,

covering the seed, slightly lobed at the apex. Seed 3-5 cm. long

and 2-5 cm. broad, smooth, medium brown.

PERAK: Larut, King 4078 (CAL, K, L, SING); Parit F.R., Kinta,

Calder K.F.N. 54695 (KEP) and Zabid K.F.N. 54691 (KEP).

SELANGOR: Weld Hill, Motan K.F.N. 51948 (KEP); Sinclair S.F.N.

40082 tree No. 14 (BO, E, K, L, SING); Wyatt-Smith K.F.N. 66456

tree No. 14 (KEP); 12th mile Ulu Gombak, Strugnell F.D. 12699

(SING); Klang, Kehding 98 (FI).

ons SEMBILAN: Senaling Inas, Kuala Pilah, Raju K.F.N. 62883

Matacca: Bukit Naning, Alvins 898 or 853 (SING).

SINGAPORE: Liane Road, Botanic Gardens’ Jungle, Sinclair S.F.N.

40211 (BK, BO, DD, E, K, L, P, SING holotype); Sinclair 8th May,

1953, from the same tree (BO, E, K, L, P, SING) and Sinclair 9365

(A, B, BM, BRI, E, FI, K, L, M, P, PNH, SAN, SING); without loc. or

collector, Schlesisch botanischer Tauschverein 862 (BRSL); Botanic

Gardens’ Jungle, Murton 76 (K).

DISTRIBUTION: Sumatra, Tjerenti, bb25216 (L). Borneo, Kostermans

6713 (L).

A species with large glabrous fruit and black-punctate marks on

the underside of the leaf. It has been either left unnamed in

herbaria or confused with H. bracteosa or with sucosa (King

4078). The twigs are reddish brown and not pale straw-coloured

as in these two species. The perianth does not persist in fruit as

in bracteosa.

(15) H. sucosa (King) Warb., Monog. Myrist. (1897) 322;

Gamble, Mat. F.M.P. 5, 23 (1912) 217; Ridley, F.M.P. 3

(1924) 58.

Basionym: Myristica sucosa King in Ann. Roy. Bot. Gard.

Calc. 3 (1891) 301 Pl. 172 Fig. 1 excl. Figs 2—9.—Fig. 45.

Plate XIIA.

Tree 15-30 m. high. Bark thin, reddish brown outside, red

inside, flaking, longitudinally fissured, the fissures shallow; sap

dark red, not copious. Twigs pale straw-coloured, longitudinally

striate, glabrous except at the apex where they end in a large,

acute, rusty-tomentose, terminal bud. Leaves crowded towards the

terminal bud, cariaceous, glabrous, dark green and glossy above

with a paler green midrib, greyish brown when dry, paler green

beneath with a yellow-green midrib, medium brown when dry,

oblanceolate or obovate-lanceolate, acute at the apex, much

narrowed at the base from the middle downwards, margins slightly

revolute; main nerves 12-16 pairs, fine and faint above, pro-

minent beneath, curving evenly; midrib flush with the upper surface

or slightly raised towards the apex of the leaf, prominent on the

416

Vol. XVI. (1958).

Fig. 45. Horsfieldia sucosa (King) Warb.

A, Leafy twig with male inflorescences. B, Male flower. C, Staminal

column. D, Staminal column, top view. E, Fruit. A-D from Holt-

tum S.F.N. 37781. E from Nilsum S.F.N. 34442.

417

Gardens Bulletin, S.

lower surface; reticulations invisible or very faint; length 12-25

cm.; breadth 3-5—4:5 cm.; petiole 2-2-5 cm. long, bordered by

the decurrent margins of the blade. Male inflorescence a much

branched panicle from the branchlets below the leaves, scurfy,

ending in sub-umbellate cymes. Male flowers yellow, globose,

glabrous, 2-3 mm. in diameter on slender, 2-3 mm. long

pedicels; perianth 3-lobed, occasionally 2-lobed, the teeth triangu-

lar, acute, several veined on the inside (when dry); androecium sub-

globose or slightly flattened, very shortly stalked; anthers 6-8,

incurved at the apices. Female inflorescence stout, rigid, 3—4 cm.

long. Female perianth ovoid-globose, glabrous, fleshy and with 3

triangular teeth; ovary sessile, ovoid-globose, glabrous; stigma

sessile. Fruit solitary or 2—3 in a raceme, ovoid or pyriform,

slightly narrowed towards the apex, glabrous, very succulent,

pinkish yellow, 6—7:5 cm. long and 4—5 cm. broad; the pericarp

2 cm. thick but shrivelling much in dried specimens so that the ap-

pearance of the fruit is altered and much smaller; stalk woody,

about 4 cm. long. Aril fleshy, yellow, enveloping the seed, slightly -

laciniate and conduplicate at the apex, otherwise entire. Seed 3—4

cm. long.

KEDAH: Perangin F.R., Awang F.D. 42382 (KEP); Bigia Enggong

F.R., Sik, Ali K.F.N. 73767 (KEP).

PENANG: Telok Bahang, Yahaya S.F.N. 35768 (SING).

PERAK: Gopeng, King 4647 (CAL, G, K, L) type material; Ulu

Bubong, King 10475 (CAL, K) type material; Tualang F.R., Kinta,

K.F.N. 63252 (KEP); Pondok Tanjong, Cubitt F.D. 9680 (KEP,

SING); Kinta, D.F.O. K.F. Nos. 54701 (KEP); 54716 (KEP); 54726

(KEP); 54728 (KEP) and 54829 (KEP); Parit F.R., D.F.O. K.F.

Nos. 54667 (KEP) and 54803 (KEP).

TRENGGANU: Jerangau S.L., Dungun, Kochummen K.F.N. 80805

(SING).

PAHANG: Kuantan Water Works, Soh F.D. 15728 (KEP, SING);

Kemasul F.R., Mat Yassim F.D. 14080 (KEP, SING).

SELANGOR: Serdang, Milsum S.F.N. 34442 (SING); Weld Hill F.R.,

Ahmad C.F. 2839 (K, KEP). .

NEGRI SEMBILAN: Sungei Menyala F.R., Port Dickson, Sinclair

S.F.N. 40155 (BM, BO, E, K, L, SING).

MaLacca: Maingay 1300 (CAL, K) type material; Maingay 1306/2

(K); Alvins 1349 (SING); Tubong, Goodenough 1994 (SING).

SINGAPORE: Botanic Gardens, Potting Yard, Holttum S.F.N. 37781, —

(KEP, SING); Burkill S.F.N. 1278 (KEP, SING); Botanic Gardens’

Jungle, Nur, 18th June, 1929 (K, NY, SING); Ridley 6559 (CAL, K,

SING); Aroid Rockery, Botanic Gardens, Sinclair, 24th Feb., 1954

(DD, E, K, L, P); Changi, Ridley, date 1892 (SING); Cluny Road,

Ridley 11354 (SING) and 11355 (CAL, K).

DISTRIBUTION: Confined to Malaya.

Unfortunately the type material quoted by King is a mixture.

The excluded specimens are:— 4078 = H. punctatifolia; 467 ==

H. subalpina. The remainder 4647 and 10475 are H. sucosa.

418

Vol. XVI. (1958).

King’s plate 172 is likewise a mixture. Fig. 1 is correct but the

rest are doubtful. Fig. 3 appears to be H. crassifolia (female). H.

sucosa is a fine, distinct species. It is nearest to H. bracteosa and

belongs to the group of H. punctatifolia, kingii and amygdalina.

It is distinguished from bracteosa by the following:— Leaves

crowded round the larger and thicker terminal bud, more coriace-

ous and not membranous, narrowed to the base and distinctly de-

current on to the longer petiole, 3-5-4-5 cm. broad at the middle as

against 5—8 cm. broad in bracteosa, drying brown and not blackish

or blackish green and the perianth not persistent in fruit. For other

differences see notes under bracteosa and punctatifolia.

(16) H. bracteosa Hend. in Gard. Bull. Str. Settl. 7, 2 (1933)

£20 Pi: 30.

Synonym: H. amygdalina (Wall.) Warb., Monog. Myrist.

(1897) 310 quoad specimina malayana tantum; Gamble P. 214

et Ridley P. 57 quoad specimina malayana tantum.—Fig. 46.

Tree 5—15 m. high. Bark light brown, slightly rough but not flak-

ing, soft; sap red, copious. Twigs stout, pale straw-coloured, rough

with longitudinal striations, lenticels numerous. Leaves membran-

ous to sub-coriaceous, green or yellowish green above and slightly

glossy and with whitish-green midrib when fresh, blackish or

blackish-green when dry, slightly paler beneath, blackish-brown

when dry, margins slightly revolute; elliptic-oblong or slightly ob-

lanceolate, apex acute, base acute and only slightly decurrent on to

the petiole, midrib flush with the upper surface, broadening to-

wards the base of the leaf, prominent and slightly rough beneath;

nerves 12-16 pairs, faint and fine above, prominent beneath, obli-

que, curving slightly, interarching near the margin; reticulations

not visible when fresh, faint and few when dry, forming a loose net-

work; length 16-28 cm.; breadth 5—8 cm.; petiole stout, 1-1-5 cm.

long, black when dry, often swollen at the base. Male inflorescence

a much branched panicle up to 20 cm. long from the axils of fallen

leaves, the rachis flattened, minutely rufous-puberulous, becoming

glabrous, terminating in umbellate cymules of 3—6 flowers; bracts

caducous, variable in shape, ovate and blunt or lanceolate and sub-

acute, up to 3 mm. long and 1-8 mm. broad, rufous-pubescent on

the edges and black-dotted. Male flowers ochraceous-yellow,

strongly scented, glabrous, globose, 1:5—2 mm. in diam., with three

acute teeth and on slender, 2 mm. long pedicels; androecium shortly

stalked, depressed-globose with a small cavity in its centre; anthers

- 7—-8-(10), obtuse at the apex. Female inflorescence a short 2-3

cm. long axis, unbranched or with 1-2 branches. Female flowers

on very stout, 1-1-5 mm. long pedicels; perianth 2-25 mm. long,

419

Gardens Bulletin, S.

Fig. 46. Horsfieldia bracteosa Hend.

A, Leafy twig with male inflorescences. B, Male flower. C, Staminal

column. D, Female inflorescences. E, Female flower. F, Ovary. G,

Fruit. H, Seed in longitudinal section. A-C from Sinclair S.F.N.

40629. D-F from Ridley 9461. G-H from Sinclair S.F.N. 39937.

420

Vol. XVI. (1958).

very coriaceous, glabrous, shortly 3-toothed at the apex; ovary sub-

globose, and slightly angled at the apex, glabrous, 1 mm. in diam.

with a minutely bilobed stigma. Fruit yellowish, turning pink at the

apex, succulent and of a pleasant acid taste, glabrous, pointed at

the apex and broader at the base, the perianth persisting, 4—5 cm.

long and 3-5 cm. broad; pericarp 1 cm. thick; stalk 3—4 cm. long.

Aril orange red, shortly laciniate at the apex. Seed 2:5 cm. long;

testa shining, pale grey mottled with light brown.

KELANTAN: Bukit Batu Papan, Sungei Lebir, Henderson S.F.N.

29516 (K, KEP, SING).

PROVINCE WELLESLEY: Krian, Ridley 9461 (CAL, K, SING); Tassek

Glugor, Curtis 3735 (CAL, K, SING).

TRENGGANU: Bukit Kajang, Corner, 7th Nov., 1935 (SING); 36th

mile Kuala Trengganu-Besut Road, Sinclair & Kiah S.F.N. 39937 (BM,

BO, E, K, KEP, L, P, SING) tree was cut down in 1953 by villagers,

unfortunately.

PAHANG: Tembeling, Henderson S.F.N. 24521 (DD, K, SING holo-

type).

SELANGOR: Bukit Cheraka F.R., Compt 1, Ulu Selangor, Wyatt-

Smith K.F.N. 71371 (KEP); Sungei Pelek, Denny, 27th June, 1949 &

Sept. 1949 (SING); Sungei Buloh Reserve, Abu C.F. 3317 (K, SING).

JOHORE: Kota Tinggi-Mawai Road, 54 miles, Corner S.F.N. 29310

(BM, K, SING) and Corner, 1st Sept., 1935 (SING); Mawai, Ngadi-

man, S.F.N. 36789 (SING); Sungei Sedili below Mawai, Corner, 16th

July, 1939 (SING); Kota Tinggi-Mawai Road, 2nd mile in orchard of

J.A. le Doux, Sinclair S.F. Nos. 40629 (BM, BO, E, K, L, SING) and

40621 (E, K, SING).

SINGAPORE: Mandai Road, Kiah, 26th July, 1940 (SING).

DISTRIBUTION: Sumatra, bb1709 (L) and bb31936 (L). Borneo,

Winkler 2695 (SING). The fruits of Winkler and other recent Bornean

material are smaller and they are probably a variety of H. bracteosa.

Mr. Henderson in Gard. Bull. Str. Settl. (1933) 120 states that

this species can be distinguished from the other Malayan ones by

the presence of the inflorescence bracts. His type specimen, how-

ever, is in flower bud and at this stage as in other Horsfieldia spe-

cies, the bracts are always present but will fall when the flower

opens. Corner 29310 is an exact match and in it the bracts have

fallen. Henderson also states, ‘anthers 15-16’. I think, however, the

anthers should be half that number and that he has counted one

pollen sac as two owing to the line of dehiscence being in the

middle of each sac.

H. bracteosa is near amygdalina, a species from Burma, Cachar,

Assam and Indo-China, but not found in Malaya. Many specimens

of bracteosa in the past have been named amygdalina or confused

with sucosa. In amygdalina the leaves are smaller, the petiole thin-

ner, not swollen and not drying black. The flowers are just a trifle

larger and the staminal column not so shallow. The perianth is not

persistent in fruit and in this respect it differs from bracteosa. The

421

Gardens Bulletin, S.

Indian kingii has flowers twice as large as those of the other two and

the perianth is persistent. The inflorescence axis is rusty-tomentose,

not glabrous and the leaves are more coriaceous. They are larger

than those of amygdalina.

I reduce Horsfieldia thorelii Lecomte, Nat. Systemat. (1909)

99 and in Flor. Gén. de L’Indo-Chine 5 (1914) 100 to H. amyg-

dalina. Lecomte says that the perianth is bivalved but as in H.

amygdalina, there are flowers on the type sheet in Paris with 2 and

3 lobes. |

(17) H. polyspherula (Hk. f. emend. King) J. Sinclair, comb. nov.

Basionym: M. polyspherula Hk. f., Fl. Br. Ind. 5 (1886) 108

pro parte, alterae partes — H. subglobosa var. brachiata et al.

{vide observ.); (King in Ann. Roy. Bot. Gard. Calc. 3 (1891)

312 Pl. 146 emend.).

Synonyms: M. globularia sensu Hk. f. et Th. (non Bl.) Fl. Ind.

(1855) 160; A. DC., Prodr. 14, 1 (1856) 202 quoad sp. malacc.

‘HW. lemanniana (A. DC.) Warb. sensu Warb., Monog. Myrist.

(1897) 326 excl. typ.; Gamble, Mat. F.M.P: 5, 23 (1912 e223,

Ridley, F.M.P. 3 (1924) 59 (omnino typo excl.).—Fig. 47.

Plate XIIB.

Tree 5—20 m. high. Bark reddish brown, rough and with shallow

elongated depressions but not flaking; inner bark dark red, sap

red, watery, not copious. Twigs rusty-pubescent when young, older

glabrous, brown, slightly rough. Leaves chartaceous, fragile and

brittle when dry, oblong, elliptic to elliptic-lanceolate, dark green

and shining above, olive green when dry, paler green beneath, dark

brown when dry, glabrous except the lower, rusty-pubescent mid-

rib which ultimately becomes glabrous, apex acute or acuminate,

base narrowed and acute, rarely rounded; midrib raised above,

lying in a furrow, thin, also raised beneath; nerves 10-15 pairs,

thin but standing up in relief on both surfaces, oblique, curving

gently and interarching 2-3 mm. from the margin; reticulations in-

visible above, a few scalariform ones faintly visible beneath; length

10-21 cm.; breadth 4-8 cm.; petiole 1-2 cm. long, channelled,

rusty-pubescent. Male inflorescence a much-branched, spreading,

rusty-tomentose, 6-12 cm. long panicle from the axis of the leaves

or fallen leaves, the branches ending in sub-umbellate cymules;

bracts lanceolate, densely rusty-tomentose, soon deciduous. Male

flowers on 1 mm. long pedicels, globose, yellow, faintly scented;

perianth 1 mm. long, glabrous, 3-toothed; androecium sessile, tri-

angular; anthers 9-(12) acute and free at the apices. Female in-

florescence much shorter and condensed with fewer branches,

rusty-pubescent, 2:5 cm. or more long. Female flowers larger, 2

422

Vol. XVI. (1958).

Fig. 47. Horsfieldia polyspherula (Hk. f. emend. King) J. Sinclair.

A, Leafy twig with male inflorescences. B, Male flower. C, Staminal

column. D, Female inflorescence. E, Female flower. F, Ovary.

G, Fruit. A-—C from Sinclair S.F.N. 40711. D-F from Derry

1136. G from Corner S.F.N. 29366.

423

Gardens Bulletin, S.

mm. long, rusty pubescent; ovary 1 mm. long, sessile, ovoid, pube-

rulous; stigma sessile, 2-lobed. Fruit ovoid, slightly narrowed at

each end, 2—5 in a cyme, greenish-yellow, glabrous, suture promi-

nent, 2-4 cm. long and 1-5—2:5 cm. broad; pericarp 5 mm.

thick; stalk glabrous, 2-5 mm. long. Aril orange, entire or very

slightly lobed at the top. Seed ovoid smooth, shining.

KEDAH: K. Muda Sungkop F.R., Abdulla K.F.N. 59620 (KEP);

Gunong Jerai Reserve, Mat Ariff F.D. 7502 (K, KEP, SING); Perangin

F.R., Awang F.D. 31350 (K, KEP).

PERAK: Scortechini s.n. (DD, E) and 2iia (CAL, FI, G, L) the

figure 2 is not clear and may be 9; Ulu Bubong, King Nos. 10513 (BM,

CAL, K, MEL, SING, UPS) 10256 (CAL, DD, K) and 10431 (BM,

CAL, G, K, L); Larut, King Nos. 3309 (CAL, FI, G, LE); Sa¢6Gnr

DD, FI, K); 6566 (K) and 7526 (CAL, G, K, L); Taiping, Wray 2088

(CAL, SING); Keroh F.R., Symington 31061 (KEP); Trolak F.R.,

Browne 43462 (KEP).

PAHANG: Sawak F.R., Ulu Rompin, Soh F.D. 15484 (KEP).

SELANGOR: Public Gardens, Kuala Lumpur, Ahmat C.F. 2487 (K,

KEP, SING); Akil & Ja’amat F.D. 12088 (KEP, SING); Pawanchee

F.D. 12942 (KEP); Rawang, Ridley 7629 (CAL, K, SING).

Ma.acca: Maingay 1286 (BM, FI, G, K) one sheet in Kew is H.

subglobosa; Griffith 4354 -°(C, CAL, FI, G & Boiss, K holotype of H.

polyspherula, M, P, S); Alvins 1198 also numbered 385 (SING); Mer-

limau, Derry 1216 (CAL, K, SING); Bukit Panchor, Derry 1136 (CAL,

SING).

JoHoRE: Sungei Kayu, Kiah S.F.N. 32109 (BM, K, KEP, SING)

and §S.F.N. 31976 (BM, BRI, DD, K, KEP, SING); S. Kayu Ara,

Corner S.F.N. 29366 (K, KEP, SING); near Kota Tinggi, Ridley 4162

(K, SING); Pulau Tinggi, Sinclair S.F.N. 40288 (E, K, L, SING).

SINGAPORE: Bukit Timah F.R., Ngadiman S.F. Nos. 34630 (CAL,

K, L, SING) and 36122 (BM, DD, K, KEP, SING); Jurong, Ridley

2042 (CAL, SING) and Corner, 5th March, 1933 (K, SING); Mandai

Road, Burkill S.F.N. 6514 (SING): Corner S.F.N. 37147 (SING);

Kiah, 24th July, 1940 (SING); Sinclair S.F.N. 39601 (B, BK, BO, DD,

Delhi Univ., E, K, L, M, P, PNH, SAN, SING); Chan Chu Kang,

Ridley 2108 (CAL, SING); Botanic Gardens, Ridley 8930 (CAL, DD,

SING); Botanic Gardens Jungle, Sinclair S.F. Nos. 40681 (BM, BO,

DD, E, K, KEP, L, P, PNH, SING) and 40711 (BO, E, K, L, SING).

DISTRIBUTION: Sumatra, Borneo and Banka.

The name H. lemanniana cannot be used for this species. I have

seen a male flowering specimen, the holotype of M. lemanniana

A. DC. in an attic in Herb. Boissier in Geneva and it is H. irya.

Warburg included M: polyspherula Hk. f. emend. King in leman-

niana and subsequent authors followed him without re-examining

the type.

From the description of M. globularia [Fl. Ind. (1855) 160], it

appears that Hooker f. and Thomson had an unnumbered specimen

collected by Griffith which must be identified as H. polyspherula.

Apparently Griffith’s specimens were subsequently numbered. In

establishing M. polyspherula, Hooker f. quoted it as Griffith 4352

along with other collectors’ specimens. King, who recognized the

424

Vol. XVI. (1958).

mixture, gave an accurate description and plate of the species, but

unfortunately quoted Griffith 4351, not only under his new species

M. brachiata but also under M. polyspherula Hk. f. emend. King.

It is presumed that Griffith 4252 quoted by Hooker f. and Griffith

4351 quoted by King are misprints for Griffith 4354, for both these

two former numbers and their several duplicates could not have

been the basis of so accurate a description and plate. Other speci-

mens cited by King are correctly named. In view of this I propose

that Griffith 4354 should be considered as the holotype of H. polys-

pherula. 1 now give the identifications of numbers quoted or mis-

quoted under M. polyspherula Hk. f. on page 108 of F. B. I.

Wallich 6806—H. polyspherula, H. crassifolia and H. walli-

chii.

Griffith 4351—-H. subglobosa var. brachiata.

Griffith 4352—M. fragrans.

Maingay 1286—(misprinted as 2286)—H. polyspherula

while one of the Kew sheets is H. subglobosa.

The best character for identification is the rusty-tomentose in-

florescence. The leaves are brittle when dry and are often much

broken in appearance in herbaria. If young, the lower midrib is

rusty-pubescent and this character is sometimes helpful, although

the pubescence does not persist for long. The veins and midrib are

fine but stand up well in relief on the upper surface. The midrib

does not broaden above the petiole on the upper surface as in H.

subglobosa, the nearest ally, and the petioles are more slender.

The leaves of subglobosa are very similar but larger and thicker.

(18) H. subglobosa (Miq.) Warb., Monog. Myrist. (1897) 328;

Gamble, Mat. F.M.P. 5, 23 (1912) 220; Ridley, F.M.P. 3

(1924) 60.

Basionym: Myristica subglobosa Miq., Fl. Ind. Bat. Suppl.

(1861) 383 et in Ann. Mus. Lugd.-Bat. 2 (1865) 49.

Synonyms: M. globularia Bl. var. subglobosa Miq., in Ann.

Mus. Bot. Lugd.-Bat. 1 (1864) 206. M. glabra Migq., (non BI.) Fl.

Ind. Bat. 1, 2 (1858) 65 pro parte quoad spec. sumatranas.

M. glabra Bl. var. sumatrana Migq. in Ann. Mus. Bot. Lugd.-Bat.

2 (1865) 49. M. collettiana King in Ann. Roy. Bot. Gard. Calc.

3 (1891) 313 Pl. 147. Horsfieldia majuscula (King) Warb.,

Monog. Myrist. (1897) 315; Gamble, Mat. F.M.P. 5, 23 (1912)

215; Ridley, F.M.P. 3 (1924) 57. M. majuscula King in Ann.

Roy. Bot. Gard. Calc. 3 (1891) 310 Pl. 143. M. horsfieldia

425

Gardens Bulletin, x:

Wall. (non BI.) in Cat. 6806 pro parte. M. polyspherula Hk. f.,

Fl. Br. Ind. 5 (1886) 108 pro parte, vide observ. sub. H.

polyspherula.

var. subglobosa—Figs. 48-50 & 51, A-D.

Tree 8—20 m. high with straight trunk and Garcinia-like branch-

ing. Bark medium brown, slightly rough and with short, narrow,

vertical lines or shallow fissures, not flaking; sap pale pink, watery,

not very abundant. Twigs brownish or greyish-brown, green at the

tips, glabrous, striate and with a few pustular lenticels. Leaves

variable in size, thinly coriaceous to coriaceous, glabrous, dark

green and glossy above, paler green beneath, drying olive-green

above with some darker patches and medium brown beneath,

oblong-elliptic or obovate-elliptic, apex acute, base acute; midrib

and the 13-20 pairs of oblique nerves raised above and below; in-

terarching faintly at the margin; transverse reticulations faintly visi-

ble beneath only; length 15-28 cm.; breadth 4—10 cm.; petiole

stout, glabrous, channelled above, 1-5—2 cm. long with the leaf de-

current on to it. Male inflorescence \axly branched, up to 10 cm.

long, the main axis stout, angular, sparsely pubescent, becoming

glabrous, ending in few-flowered cymes. Male flowers globose,

waxy-yellow, fragrant, 3-lipped, glabrous, 1-2 mm. in diam. on

1-2 mm. long pedicels; androecium nearly sessile, trigonous; an-

thers 8—13, attached to the androecium at their bases but free at

their edges and at their acute and erect apices. Female inflorescence

shorter with fewer branches. Female flowers waxy-yellow, fragrant,

decurved, fleshy, glabrous, 3 or occasionally 4-toothed, ovoid and

slightly flattened at the top, 2-3 mm. long on 2—3 mm. long pedi-

cels; ovary pale green, sub-globose, glabrous, 1:5 mm. in diam.,

ridged on one side and grooved on the other, the groove opposite

and continuous with the ridge, stigma minutely bilobed, sessile.

Fruit yellow, later tinged reddish, ovoid or sub-globose, pointed

or rounded at the apex, glabrous, variable in size, 3-4 cm. long and

2:3-3:3 cm. broad. Aril orange-pink, fleshy, covering the ovoid

seed.

PERAK: Scortechini 837 (CAL, K, L); King 6765 (CAL, K, SING)

type material of M. majuscula; Gopeng, King 6004 (BM, CAL, G, L,

SING) type material of M. majuscula; Ulu Bubong, King 10413 (CAL,

FI, K) type material of M. majuscula; Larut, King Nos. 6566 (BM,

CAL, G, K); 3620 (CAL, E, K) type material of M. collettiana; 5059

(CAL, G, K, L) type material of M. majuscula; 6672 (CAL, SING)

type material of M. collettiana; 6737 (BM, CAL, K, L) type material

of M. collettiana; 3899 (CAL, FI, G, K, P) type ‘material of M. collet-

tiana; B.P.D., King 7965 (BM, K) type material of M. majuscula;

Barnard C.F, 30 (KEP); Waterfall Hill, Taiping, Wray 2705 (CAL,

SING) type material of M. majuscula; Wray 2218 (CAL) type mate-

rial of M. majuscula; Wray 2064 (CAL, K); Taiping Hill, Haniff &

426

Vol. XVI. (1958).

Fig. 48. Horsfieldia subglobosa (Miqg.) Warb. var. subglobosa.

A, Leafy twig with fruit. B, Seed with aril. C, Male inflorescence. D,

Male flower. E, Staminal column. F, Staminal column from above.

A-B from Sinclair S.F.N. 39989. C-F from Ridley 12155.

427

Gardens Bulletin, S.

Fig. 49. Horsfieldia subglobosa (Mig.) Warb. var. subglobosa.

A, Female inflorescences. B, Leafy twig with young fruit. C, Female

flower. D, Female flower showing ovary. E, Young fruit in longi-

tudinal section. A-E from Sinclair S.F.N. 39697.

428

a a

Vol. XVI. (1958).

Fig. 50. Horsfieldia subglobosa (Migq.) Warb. var. subglobosa.

A, Leafy twig. B, Female inflorescence. C, Female flower showing

ovary. D, Fruit. E, Fruit in longitudinal section. F, Seed with aril.

A, D, E and F from Denny, 25-3-1954. B and C Denny 20-5-1954.

429

Gardens Bulletin, S.

TRAM)

Det

Fig. 51. Horsfieldia subglobosa (Mig.) Warb. var. subglobosa and var.

brachiata (King) J. Sinclair.

A-B, var. subglobosa. Leafy twig and fruit. C, Fruit of the same in

longitudinal section. D, Seed and aril of the same. E, Twig of var.

brachiata showing the two ridges. A-D from Sinclair S.F.N. 40048.

E from Corner S.F.N. 28962.

430

Vol. XVI. (1958).

Nur 2477 (SING); Kurau, Wray 122 (CAL, K, SING); Ridley

11919 (K, SING); Batu Gajah, Wells F.D. 3038 (K, KEP, SING).

TRENGGANU: Ulu Ayam, Corner S.F.N. 30556 (SING).

PAHANG: 3rd mile Pulau Manis Road, Kuantan, Soh F.D. 15091

(KEP, SING); Pulau Tioman, Juara Bay, Burkill, June 1915 (SING).

SELANGOR: Sungei Tinggi, Kuala Selangor, Nur S.F.N. 34117 (KEP,

L, SING); S. Buloh, Sow & Tachun F.D. 16450 (KEP, SING); Sungei

Pelek, Denny, Dec. 1947 (SING); Denny, 25th March, 1954 (BK,

BM, E, K, L, PNH, SING); Denny, 20th May, 1954 (SING); 9th

mile Pahang Road, Klang Gates, Watson C.F. 537 (KEP, SING).

Matacca: Derry 134 (CAL, SING); Bukit Sidenan, Penghulu Bebas

134 & 136 (SING); Brisu, Derry 649 (CAL, SING).

JOHORE: Sungei Kayu, Kiah S.F.N. 32064 (SING); 27th mile Kota

Tinggi-Jemaluang Road, Henderson S.F.N. 35779 (SING); Road to

Gunong Pulai, Ridley 12155 (K, SING); Pulau Tinggi, Lindong K.F.N.

70915 (KEP); Gunong Panti, Bain F.D. 5959 (K, KEP).

SINGAPORE: Hullett 590 (K, SING) type material of M. majuscula;

Ridley or Mat 5971 (CAL, SING); Bukit Timah, Ridley Nos. 6451

(CAL, SING); 6449 (CAL, SING); 4438 (CAL, SING); 10839

(SING); Sinclair S.F.N. 39989 (BO, E, K, L, P, SING); Mandai

Road, Sinclair S.F.N. Nos. 39599 (E, K, L, P, SING) and 40043 (BM,

BO, E, K, L, P, SING); Cluny Road, Ridley 4439 (CAL, K, SING)

Kew sheet is numbered 4459; Bajau, Ridley 6448 (CAL, K, SING); off

9th mile Upper Thomson Road, Sinclair S.F.N. 39697 (BO, DD, E,

K, LP; SING).

DISTRIBUTION: Sumatra, Borneo, Billiton, Philippines, Celebes.

var. brachiata (King) J. Sinclair, stat. nov.

Basionym: Myristica brachiata King in Ann. Roy. Bot. Gard.

Calc. 3 (1891) 311 Pl. 144.

Synonym: Horsfieldia brachiata (King) Warb., Monog, My-

rist. (1897) 325; Gamble, Mat. F.M.P. 5, 23 (1912) 218;

Ridley, F.M.P. 3 (1924) 59.—Fig. 51, E.

Twigs with two raised ridges equidistant from each other and

extending from leaf base to leaf base.

KEDAH: Baling, Md. Salleh K.F.N. 66360 (KEP); Gunong Jerai,

Yakim F.D. 7723 (K, KEP, SING).

KELANTAN: Kelumpur, Haniff & Nur S.F.N. 10387 (K, SING); Ulu

Kusial, Abdulla bin Ismail K.F.N. 68058 (KEP).

PERAK: Scortechini 1649 (CAL, K, L) type material of M. brachiata;

Gopeng, King 4704 (CAL, K, L) type material of M. brachiata; Larut,

King 6771 (CAL, FI, G, K); S. Krian Estate, Spare S.F.N. 34569 (K,

KEP, SING); Kulim hot springs near Grik, Henderson S.F.N. 23869

(K, SING); Grik, Corner S.F.N. 31617 (SING) ;{Sungei Tukang Sidin,

T. Anson, Haniff S.F.N. 14334 (K, KEP, SING)]

S° + — Trenccanu: Ulu Brang, Moysey & Kiah S|F.N. 33753 (BM, DD,

y*® ~—sCK, KEP, SING); 34th mile Kuala Trengganu-Besut Road, Sinclair &

) Kiah S.F.N. 40772 (A, BO, E, K, L, SING).

PAHANG: Raub, Kalong F.D. 20324 (KEP, SING); Pekin, Ja’amat

F.D. 16507 (KEP, SING).

Matacca: Griffith 4351 (CAL, K, M, P, S) type material of M.

brachiata.

431

Gardens Bulletin, S.

JOHORE: Mawai, Ngadiman S.F. Nos. 34708 (BM, DD, KEP, SING)

and 36791 (BRI, DD, K, KEP, SING); Corner, 7th Oct., 1934

(SING); 8th mile Kota Tinggi-Mawai Road, Corner S.F. Nos. 28707

(K, SING) and 32281 (KEP, SING); Sungei Berassau, Mawai-Jema-

luang Road, Corner S.F.N. 28963 (BM, K, PNH, SING); S. Dohol,

Mawai-Jemaluang Road, Corner S.F.N. 29466 (K, SING); S. Sedili,

Corner S.F.N. 25991 (K, KEP, NY, SING).

DISTRIBUTION: Siam, Billiton, Sumatra, Borneo and Philippines.

H. subglobosa is a variable species with a wide distribution and

the Malayan material agrees quite well with that of Borneo and

Sumatra. The fruit varies in size and shape there being forms which

are quite round and others with a pointed apex. I do not see the

value of the characters given to separate H. subglobosa, majuscula

and brachiata. The leaves and fruit vary but the flowers in all three

are remarkably uniform and distinct in the loosely attached anthers

which form a triangular androecium. H. brachiata can be separated

from subglobosa only by the presence of the two raised ridges on

the twigs, but the flowers and other features are essentially the

same. After much consideration I have given it varietal status. It

is found near streams and in wet places. Warburg was mistaken

when he stated in his key that H. majuscula had a bilobed perianth.

In his description he is correct when he says it is three-valved.

Gamble and Ridley follow him and make the same mistake.

(19) H. ridleyana (King) Warb., Monog. Myrist. (1897) 331;

Gamble, Mat. F.M.P. 5, 23 (1912) 221; Ridley, F.M.P. 3

(1924) 60.

Basionym: Myristica ridleyana King in Ann. Roy. Bot. Gard.

Calc. 3 (1891) 311 Pl. 145.—Fig. 52.

Tree 12—18 m. high. Twigs slender, dark brown, striate, glabrous

except the terminal bud. Leaves chartaceous, rather brittle when

dry and then becoming olive green above and medium brown be-

neath, lanceolate, base cuneate, apex acuminate; midrib slightly

sunk or flush with the upper surface, raised beneath; nerves 7-10

pairs, sunk above, raised beneath, very fine and faint on both sur-

faces, spreading, interarching at the margin; reticulations invisible;

length 5-5—10:5 cm.; breadth 2—3 cm.; petiole 1 cm. long, slender,

channelled above. Male inflorescence 3—5 cm. long, nearly glabrous,

branched, the branches ending in sub-umbellate cymes. Male

flowers globose on 1 mm. long pedicels; perianth glabrous, 1 mm. in

diam., 3-toothed, the teeth acute; androecium sessile, trigonous and

triangular in cross section; anthers 7-10, erect, attached to the an-

droecium at their bases but free higher up at their margins and their

acute apices. Female inflorescence a short raceme, 1:5—2 cm. long.

Female flowers globose, coriaceous, 1:5—-2 mm. in diam.; ovary

432

Vol. XVI. (1958).

Fig. 52. Horsfieldia ridleyana (King) Warb.

A, Leafy twig with male inflorescences. B, Male flower. C, Staminal

column. D, Staminal column from above. E, Fruit. A-D from

Ridley 16177. E from Mahamud F.D. 20411.

433

Gardens Bulletin, S.

sessile, glabrous, grooved; stigma oblique, grooved. Fruit (imma-

ture) solitary or 2—4 in a short raceme, ellipsoid, 1-2 cm. long,

1-1-2 cm. broad; stalk 5 mm. long, stout; pericarp thick, gla-

brous. Avil thin and covering the seed. Seed oval, smooth.

PERAK: King 10917 (BM, K, L) type material; Haram, Scortechini

862 (BM, CAL, K, SING) type material; Ulu Bubong, King 10917

(CAL, K) type material. ,

PAHANG: Wray’s Camp, Tahan, Ridley 16177 (K, SING); Bukit

Kajang F.R., Raub, Mahamud F.D. 20411 (KEP, SING).

SELANGOR: Kanching, Symington F.D. 30107 (KEP).

MALACCA: Sungei Ujong, Cantley 1798 (K) type material.

DISTRIBUTION: Sumatra, bb26117 (L, SING).

This species seems to be rare. It belongs to the group having a

trigonous androecium and erect anthers which are free along the

margins and apices. It has smaller leaves than any of the other

Malayan species except H. penangiana and the reticulations are

entirely invisible. For differences between it and H. penangiana see

notes under that species. It recalls H. polyspherula but that species

has larger leaves, the nerves and midrib raised above, the midrib

beneath pubescent when young and the inflorescence axis rusty-

pubescent.

4. GYMNACRANTHERA (A. DC.) Warb. [in Ber. Pharmac.

Ges. 2 (1892) 227 et Ber. Deutschen Bot. Ges. 13 (1895) 82

& 94 et] Monog. Myrist. (1897) 131: Gamble, Mat. F.M.P. 5,

23 (1912) 222; Ridley, F.M.P. 3 (1924) 61. [Myristica section

Gymunacranthera A. DC. in Ann. Sc. Nat. 4, 4 (1855) 31 et

Prodr. 14, 1 (1856) 200; Migq., Fl. Ind. Bat. 1, 2 (1858) 63;

King in Ann. Roy. Bot. Gard. Calc. 3 (1891) 304].

Trees or shrubs. Dioecious. Twigs greyish or reddish brown,

pustular with lenticels, not striate, often angled, smooth and shin-

ing, especially in the upper and younger parts. Leaves glabrous or

in one species covered with rusty-stellate scales beneath, reddish-

white or glaucous beneath; reticulations few, obscure or invisible,

never forming a dense, close network; nerves distinctly interarching

at the margin in a single or double loop; mesophyll cells sclerenchy-

matous. Inflorescence axillary, paniculate, the male branched, the

female contracted, bracts caducous, bracteoles absent. Flowers ur-

ceolate, 3-toothed, the teeth acute. Filaments united in an oblong,

thick, connate column, anthers elongate, 2-celled, 6-12, covering

the whole column and adnate to it, except at the apex where they

are free and acute, the acute apices often inflexed. Stigma sessile,

minutely bilobed. Fruit never very large, 2—3 cm. long and 1-2 cm.

434

Vol. XVI. (1958).

broad; aril laciniate almost to the base; testa woody; albumen rumi-

nate containing a fixed oil but no starch; cotyledons divaricate,

connate at the base.

TYPE OF GENUS: G. paniculata (A. DC.) Warb. Philippines.

DISTRIBUTION: S. India, Siam, Malay Peninsula and Islands including

the Philippines and New Guinea. Six species and four varieties.

Gymnacranthera is distinguished from the other Malayan Myris-

ticaceae by its staminal column and its non-striate twigs. From

Horsfieldia it is distinguished also by the leaves which are whitish

beneath and the aril which is laciniate to the base; from Knema by

the reticulations of the leaves never forming a dense network, the

branched inflorescences without bracteoles, the absence of a style,

the aril laciniate to the base, the absence of starch in the albumen

and the cotyledons connate at the base. It differs from Myristica

both in the absence of a bracteole and starch in the albumen.

There are six species and four varieties in the genus. Of the

Malayan ones G. bancana is the most distinct. The remaining ones

are so close to each other that they are not always easily distin-

guished except geographically. Some botanists may refer to regard

them as sub-species and others as varieties. They differ among

themselves in such minor characters as the thickness of the leaves,

the number of nerves, whether these are raised or level with the

leaf surface and in the shape of the fruit. The flower is remarkably

uniform and is unreliable in the separation of species. G. ibutii

Holth. in Blumea 5 (1942) 183 from Karakelong in the Talaud

Islands is Horsfieldia macrocoma or a variety of it.

KEY

a. Leaves large, 20-40 cm. long and 7:5—16-5 cm. broad, vivid

rusty-pubescent beneath especially when young (later greyish),

nerves 19-23 pairs. Inflorescence axis rusty, woolly-tomentose

with hairs 1 mm. long, flowers, both sexes rusty-tomentose,

6 mm. long. Fruit rusty-tomentose with an acute, sometimes

slightly uncinate apex (1) G. bancana

a. Leaves smaller, not over 28 cm. long and not over 10 cm.

broad, nerves not more than 20 pairs. Inflorescence axis not

tomentose but often puberulous to adpressed-pubescent,

flowers scarcely tomentose mostly pubescent, less than 6 mm.

long. Fruit not tomentose, apex not sharply acute

b. Leaves 12-27 cm. long and 3-3—10 cm. broad, nerves 13-20

pairs. Fruit ellipsoid or oblong

435

Gardens Bulletin, S.

c. Leaves coriaceous, slightly glaucous or not beneath, the

sides more or less parallel, nerves 15—20 pairs, not form-

ing a distinct double loop at margin. Female inflorescence

1:5 cm. long, very densely crowded with the flowers

touching each other (2) G. contracta

c. Leaves chartaceous or thinly coriaceous, very glaucous be-

neath, the sides curving and less often parallel, nerves

13-17 pairs, more distinct, forming a distinct double

loop at the margin. Female inflorescence 2-5-3 cm. long,

the flowers not densely crowded, scarcely touching each

other (3) G. forbesii

b. Leaves 6-15 cm. long and 1-5—5-5 cm. broad, nerves 6—9

pairs. Fruit ovoid-elliptic or sub-globose

d. Nerves not raised above, very fine and faint on the lower

surface and not raised, cannot be felt by rubbing with

the finger. Male flowers 3-4 mm. long. Fruit ovoid-

elliptic, 2 cm. long and 1-:3-1:5 cm. broad

(4) G. eugeniifolia

d. Nerves raised above and below, fine but those below can

be felt by rubbing with the finger. Male flowers 4—5

mm. long. Fruit oblong or sub-globose, up to 2-8 cm.

long and 2—2:5 cm. broad

(4) G. eugeniifolia var. griffithii

(1) G. bancana (Mig.) J. Sinclair, comb. nov.

Basionym: Myristica bancana Mig., Fl. Ind. Bat. Suppl. 1

(1861) 383; Warb., Monog. Myrist. (1897) 518.

Synonyms: Gymnacranthera murtonii (Hk. f.) Warb., Monog.

Myrist. (1897) 357 T. 20, Figs 1-3; Gamble, Mat. F.M.P. 5,

23 (1912) 223; Ridley, F.M.P. 3 (1924) 61. Myristica murtonii

Hk. f., Fl. Br. Ind. 5 (1886) 105; King in Ann. Roy. Bot. Gard.

Calc. 3 (1891) 297 Pl. 124 ter. M. ferruginea King in Ann.

Roy. Bot. Gard. Calc. 3 (1891) 298 Pl. 125. M. amplifolia

Warb., Monog. Myrist. (1897) 517.

var. banacana—Fig. 53. Plate XIIIA.

Tree 16-25 m. high. Bark reddish-brown, slightly rough, not fis-

sured, inner layers dark red, sap dark red, not copious; sap-wood

white, heart-wood not seen. Young twigs rusty-tomentose, later

glabrous, pustular greyish-brown. Leaves coriaceous, very dark

green, glabrous and shining above, conspicuous at a distance by

the lower surface which is a vivid rusty-brown and covered with

soft stellate scales when young, the older leaves duller and tending

436

Vol. XVI. (1958).

s fall

duepim: pew

Fig. 53. Gymnacranthera bancana (Miq.) J. Sinclair var. bancana.

A, Leaf twig with male inflorescences. B, Male flower. C, Male flower

from above. D, Staminal column. E, Fruit. A—D from Sinclair

S.F.N. 40045. E from Goodenough, date 1893, Seletar.

437

Gardens Bulletin, S.

to become glabrous and glaucous-green beneath, elliptic-ovate to

broadly oblong with rounded and often emarginate base, apex

acute; midrib above flat and flush with the upper surface of the

leaf when fresh, slightly sunk when dry, raised beneath; nerves

19-23 pairs, slightly impressed above or level with the surface,

curving gradually and interarching at the margin; reticulations

obscure above, faint, transverse nervules present between the

nerves beneath; length 20-40 cm.; breadth 7:5—16:5 cm.; petiole

stout, rusty-pubescent, becoming glabrous, 2 cm. long. Male inflo-

rescence 6-10 cm. long, with several 2—3 cm. long branches, the

axis and deciduous bracts rusty-lanose with branched, rhizoid-like

wool. Male flowers with a spicy odour when crushed, golden yellow

with a brownish tinge on 3 mm. long, tomentose pedicels of the

same colour. Perianth urceolate, 6 mm. long, tomentose outside

and slightly less so inside, 3, occasionally 4-toothed, split 1/3 way

down, teeth acute; androecium sessile, 3 mm. long with 7—10 an-

thers (average number 9), their apices free and also their edges

slightly free from the column. Female flowers (those examined

not open and somewhat immature). on a short, 4—5 cm. long inflo-

rescence, 4 mm. long, probably reaching 6 mm. when mature,

rusty-tomentose, coriaceous, nearly sessile on stout, 1 mm. long

pedicels (the pedicel too will probably lengthen a bit when the

flower opens; it is longer in fruit); ovary sessile, sub-globose, densely

rusty-tomentose with an oblique, 2-lobed, sessile stigma. Fruit

rusty-tomentose, ovoid, the apex pointed and slightly uncinate;

length 2-8-3 cm.; breadth 1-5—1-6 cm.; stalk thick 5—7 mm. long;

aril deeply laciniate, fresh colour not observed. Testa yellowish

brown, shining.

JOHORE: 134 miles Mawai-Jemaluang Road, Corner S.F.N. 29945

(A, BM, CAL, K, SING); Kulai Young Estate, Corner S.F.N. 33597

(DD, K, KEP, SING); Sungei Kayu, Kiah S.F.N. 32193 (A, BM, DD,

K, KEP, SING); Compt. 19, Panti F.R., Sulaiman bin Manja K.F.N.

70190 (KEP).

SINGAPORE: Serangoon, Murton 13 (K) type of M. murtonii; Wall.

Cat. 6803 (BM, G, K) type material of M. ferruginea; Seletar, Ridley

Nos. 1835 (BM, CAL, MEL, SING) and 4815 (CAL, MEL, SING)

both type material of M. ferruginea; Bajau, Ridley 3364 (CAL, K,

SING) type material of M. ferruginea. Calcutta specimen as to fruit

only, leaf = H. subglobosa; Sungei Morai, Goodenough 3890 (CAL,

K, SING); 1232 miles Mandai Road, Sinclair S.F. Nos. 44045 (A, B,

BK, BM, BO, Univ. of Delhi, DD, E, FI, K, L, M, NY, P, PNH, SAN,

SING) and 39502 (DD, E, K, L, P, SING).

DISTRIBUTION: Sarawak (as var. borneensis), Banka and Sumatra.

A very handsome tree especially when in flower. The flowers

resemble those of Horsfieldia superba on account of their acute,

erect lobes and their large size. The anthers in both cases are

adnate to a rather elongated column but the edges are free in G.

438

Vol. XVI. (1958).

bancana and not in the other. G. bancana differs from the other

Malayan species of the genus by the larger flowers and fruit and

larger leaves having more nerves and the rusty tomentum beneath,

especially of the younger leaves.

I have seen the type of Myristica bancana Miq. (Teijsmann

3279) in the Utrecht herbarium. The fruit is identical with that of

G. murtonii but the leaves are less coriaceous and are from a young

twig. The twig is that of a Gymnacranthera and not a Myristica. I

have therefore used the older name bancana. I have also seen

Beccari Nos. 1211 (FI, G, K, M, P, S) and 3977 (FI, G, K)

which are named G. murtonii var. borneensis Warb. This variety

differs from the type in the larger flowers and in the cuneate leaf

base. M. amplifolia Warb. (L holotype), sterile collected from a

sapling shows juvenile leaves. It also is not different from G.

bancana.

It is convenient here to make the formal transfer for this variety

from G. murtonii to G. bancana.

Gymnacranthera bancana (Mig.) J. Sinclair var. borneensis (Warb.)

J. Sinclair, comb. nov.

Basionym: G. murtonii (Hk. f.) Warb. var. borneensis Warb.,

Monog. Myrist. (1897) 359.

Synonym: M. murtonii Hk. f. var. borneensis (Warb.) Boerl.,

Handl. Fl. Ned. Ind. 3, 1 (1900) 88 nom. alt.

(2) G. contracta Warb., Monog. Myrist. (1897) 360 T. 20 Figs

1-4.

Synonym: Myristica contracta (Warb.) Boerl., Handl. FI.

Ned. Ind. 3, 1 (1900) 88 nom. alt.—Fig. 54.

Tree up to 20 m. high. Bark reddish brown, rough; sap red.

Twigs greyish brown, stout, angled, glabrous. Leaves coriaceous,

oblong-lanceolate, glabrous, drying dark chocolate-brown, apex

acute, base acute or rounded; midrib sunk above, raised, glossy

and smooth beneath; nerves 15-20 pairs, oblique, faint above,

sometimes prominent and sometimes faint beneath; reticulations

not visible; length 16—27 cm.; breadth 4-10 cm.; petiole 1—2 cm.

long, deeply grooved above. Male inflorescence 4-8 cm. long,

branched at the base, rusty-pubescent, and with deciduous, rotund

bracts 5 mm. long and 6 mm. broad. Male flowers yellow, 3-3-5

mm. long on slender pedicels 2—5 mm. long; perianth split into 3-4

lobes extending down to half its length, rusty pubescent or tomen-

tulose outside and inside; anthers about 8. Female inflorescence

simple, contracted, 1-5-2 cm. long, very densely covered with

flowers. Female flowers yellow, 2‘5—3 mm. long, with pedicels 1-2

mm. long, pubescent outside and inside, the lobes extending half-

way down the perianth; ovary tomentose, globose, 1-5-2 mm. in

439

Gardens Bulletin, S.

VAM

SNS ot SOS

. = aS

JN. —

ry — :

ta ue

Ky te

ext ie> (=

Ns apNS

\2 ao &. e~.

YES TE.

ot / Vy BD

5 cm

Fig. 54. Gymnacranthera contracta Warb.

A, Leafy twig with female inflorescences. B, Female flower. C, Ovary.

D, Male inflorescences. E, Male flower. F, Staminal column. G,

Fruit. A-—C from Haviland & Hose 3308. D-F from Haviland &

Hose 1650. G from Kostermans 6086.

440

Vol. XVI. (1958).

diam. with a glabrous, bi-lobed stigma. Fruit orange, at first with

rusty scurf, later glabrous, solitary or in pairs, 2 cm. long and 1

cm. broad, oblong or ellipsoid, on a stalk 7 mm. long; aril laciniate,

fresh colour not seen.

Maracca: Bukit Maning, Alvins 854 (SING) also numbered 897.

DISTRIBUTION: Kuching, Sarawak, Beccari 321 (A, FI. G & Boiss,

K, M, NY, P, S) and Beccari 2999 (FI, G, K, P) both type material;

Beccari 419 (FI, G, K) type material; Hose 3308 (CAL, K, SAR).

Indonesian Borneo, Billiton.

Very rare and not collected again in Malaya. Sinclair S.F.N.

39581 (A, B, BKF, BM, BO, Univ. of Delhi, DD, E, K, L, M,

NY, P; PNH, SAN, SING) sterile, from Bukit Timah, Singapore

is somewhat intermediate between forbesii and contracta. The

leaves are not quite so large as in typical contracta, but are more

like those of it than of forbesii, so the specimen is best placed in

contracta.

(3) G. forbesii (King) Warb., Monog. Myrist. (1897) 363 T. 20

Figs 1-2; Gamble, Mat. F.M.P. 5, 23 (1912) 224; Ridley,

F.M.P. 3 (1924) 61.

Basionym: Myristica forbesii King in Ann. Roy. Bot. Gard.

Calc. 3 (1891) 306 Pl. 137.—Fig. 55. Plate XIIIB.

Tree 16-25 m. high. Bark greyish-brown, slightly rough but not

fissured or flaky, inner bark pinkish, sap red, scant, wood pale.

Twigs glabrous except the apical bud, angled, greyish, smooth,

shining in the apical parts, lower down terete, greyish brown and

pustular. Leaves chartaceous or thinly coriaceous, dark green and

dull above, dark brown when dry, glaucous beneath with brown

veins, the glaucous colour often persisting when dry, elliptic-ob-

long or less often oblong-lanceolate, apex acute to acuminate, base

rounded or acute; midrib and the 13-17 pairs of oblique nerves

sunk above, rather prominent beneath, the latter ascending rather

steeply, interarching in 2 loops near the edge; reticulations invisible

above, invisible or few and faint beneath; length 12-22 cm.;

breadth 3-3-9 cm.; petiole 1-1-5 cm. long, glabrous, channelled

above. Male inflorescence a spreading panicle, 5—6 cm. long,

branched, rusty-pubescent, bearing flowers in umbellate clusters.

Male flowers yellow with a faint scent of lemons, on slender 3 mm.

long pedicels; perianth 3-4 mm. long, ovoid in bud, urceolate when

open, puberulous outside and inside, several-veined from apex to

base (the hairs along the veins, seen best on the inside), teeth 3,

sometimes 4 or 5; reflexed; androecium elongate, 2-2-5 mm. long,

including the 0-5 mm. long stalk; anthers 7-9, slightly twisted, their

apices and edges free but their backs united to the column. Female

44]

Gardens Bulletin, S.

3 mm

Fig. 55. Gymnacranthera forbesii (King) Warb.

A, Leafy twig with fruit. B, Male inflorescences. C, Male flower. D,.

Staminal column. E, Female inflorescences. F, Female flower. G,,

Ovary. A from Ridley 9464. B—D from Hamid F.D. 10864. E-G:

from Ahmad C.F. 4581.

442

Vol. XVI. (1958).

inflorescence much shorter, 2-5-3 cm. long with 1—2 branches at

the base only. Female flowers more coriaceous, 3-4 mm. long on a

stouter stalk 2 mm. long; ovary sub-globose, tomentose; stigma

2-lipped, oblique, sessile. Fruit bright pink, slightly scurfy, ellip-

soid, less often slightly obovoid, 2—2:5 cm. long and 1:3 cm. in

diam. on a 7 mm. long stalk. Seed smooth, oblong or ellipsoid; aril

pale, divided to the base into many narrow segments.

PENANG: Government Hill, Maingay 1293 (CAL, K) type material;

Telok Bahang, Forest Guard 12741 (SING).

PROVINCE WELLESLEY: Krian,) Ridley 9464 (CAL, K, SING).

PERAK: Without locality, King 8159 (CAL, DD, K, P); Larut, King

Nos. 8783 (CAL, K, SING, UPS); 6591 (BM, CAL, E, FI, G, K, L,

P,: SING, UPS); 7645 (E, K, L); 6784. (BM, CAL, E, FI, G, K, L);

6973 (CAL, FI, G, K, L, UC); 7419 (CAL, E, K, L, SING) and 7732

(CAL, FI, K, L, P); near Ulu Kerling, King Nos. 8722 (CAL, DD, K,

MEL) and 8756 (CAL, DD, FI, G, K); Ulu Bubong King 19080

(CAL); Sungei Gebal State land, Pandak F.D. 29861 (SING); Tapah,

Wray 1429 (BM, CAL, FI, G, L).

PAHANG: Kemansul Reserve, Temerloh, Hamid F.D. 10864 (K,

SING); Belingo, Temerloh, Awang Lela C.F. 2683 (K, SING); Ulu

Chimeras, Kuala Lipis, Burkill & Haniff S.F.N. 15665 (A, CAL,

SING); Endau, Rompin, Mahamud F.D. 17127 (SING); Rompin,

Bidin F.D. 15671 (SING).

SELANGOR: Weld Hill, Hamid C.F., tree No. 1, 8th June, 1918

(SING); Public Gardens, Kuala Lumpur, Hashim C.F. 2455 (SING);

Ahmad C.F. 4581 (K, SING) and Burkill S.F.N. 6334 (SING); Klang

Gates, Hashim 25 (SING); Damansara Road, Ahmad C.F. 5423

(SING); Kanching F.R., Arnot F.D. 14801 (K, KEP, SING); 12th

mile Ulu Gombak, Strugnell F.D. 11151(SING); Sungei Lalang, Ka-

jang, Symington F.D. 24081 (SING).

NEGRI SEMBILAN: Sungei Menyala F.R., Port Dickson, Wyatt-Smith

K.F.N. 64086 (KEP); Tampin Hill, Goodenough 1845 (CAL, DD,

K, SING).

MALACCA: Maingay 1295 (K) type material; Yvan, s.n. (G); Kesang

Tua, Goodenough 1317 (CAL, SING); Ayer Panas, Holmberg 818

(SING); Sungei Udang, Goodenough 1355 (CAL).

JOHORE: Sedili Bot. Reserve, Corner S.F.N. 36961 (K, KEP, SING);

Sungei Sedili, Ngadiman S.F.N. 36866 (A, BM, BRI, DD, K, KEP,

SING) and S.F.N. 36918 (A, BM, BRI, K, KEP, SING); 8th mile

Kota Tinggi-Mawai Road,. Corner S.F.N. 28712 (A, K, SING); Sungei

Berassau, Mawai-Jemaluang Road, Corner S.F.N. 28970 (BM, K,

SING); Batu Pahat, Yeob F.D. 5920 (K, SING).

SINGAPORE: Seletar, Ridley 6270 (K, SING); and 6157 (SING)

both type material; Dalvey Road, Ridley, no data (SING); 114 miles

Mandai Road, Sinclair S.F.N. 39539 (E, SING); east end of Mac Rit-

chie Reservoir, Sinclair S.F.N. 37936 (DD, E, L, SING).

DISTRIBUTION: Siam, Sumatra, Banka and Riouw (P. Bintang).

The leaves are thinner in texture and the veins and their arching

more distinct than those of G. contracta. The twigs of both are

‘similar.

443

Gardens Bulletin, S.

(4) G. eugeniifolia (A. DC.) J. Sinclair, comb. nov.

Basionym: Myristica eugeniifolia A. DC., Ann. Soc. Nat. 4, 4

(1855) 29 et Prodr. 14, 1 (1856) 190; Miq:, Fl Ind? are

(1858) 58; Hk. f., Fl. Br. Ind. 5 (1886) 113.

Synonyms: Gymnacranthera farquhariana (Hk. f. et Th. sensu

King) Warb., Monog. Myrist. (1897) 365 T. 22 Figs 1-2;

Gamble, Mat. F.M.P. 5, 23 (1912) 225; Ridley, Fyre

(1924) 62 [omnino pro parte quoad specimina malayana tan-

tum]. Myristica farquhariana Hk. f. et Th. sensu Fl. Br. Ind. 5

(1886) 108 pro parte; King in Ann. Roy. Bot. Gard. Calc. 3

(1891) 305 pro parte [non Hk. f. et Th., Fl. Ind. (1855) 162

sensu stricto — A. DC., Prodr. 14, 1 (1856) 200 et Migq., FI.

Ind. Bat. 1, 2 (1858) 63 = species ex India tantum — Synonym

G. canarica (King) Warb.]—Fig. 56. Plate XIV.

var. eugeniifolia

Tree 10-25 m. or more high with spreading branches. Bark red-

dish-brown, rough but not fissured or flaky; sap red, abundant, wood

pale. Twigs slender at the extremities, soon thickening, rusty-pubes-

cent, short, soon joining the older branches, terete or becoming

terete, less often angled for a very short way only at the apex; older

parts glabrous, reddish-brown and slightly lenticellate. Leaves coria-

ceous, dark green and dull above, glaucous beneath, drying yellow-

ish to brownish-green above and greyish-brown below with reddish-

brown midrib, lanceolate to oblong-lanceolate, narrowed gradually

from the acuminate apex to the cuneate base, margins revolute;

nerves 8—10 pairs, very faint above, visible beneath but never pro-

minent (cannot or rarely be felt when the finger is rubbed on the

lower surface), oblique, indistinctly arching at the margins; reticu-

lations invisible; length 6-15 cm.; breadth 1:5—5-5 cm.; petiole

1 cm. long, slender. Male inflorescence 3—5 cm. long, branched,

rusty-pubescent as are the 2-3 mm. long, slender pedicels. Male

flowers yellow, 3-4 mm. long, in umbellate cymules; perianth

ovoid, inflated, puberulous outside, the inside with several vertical

lines, teeth 3—4, acute; androecium cylindric, nearly sessile, 2 mm.

long with 7—8 anthers, their apices and edges free from the column.

Female inflorescence 1—2 cm. long. Female flowers more coriace-

ous, 3 mm. long, rusty-pubescent outside and inside, deeply

cleft with 3, less often 4, reflexed teeth, the pedicels 2 mm.

long, stouter than in the male, reflexed; ovary ovoid-globose, pubes-

cent with a sessile, oblique, 2-lobed stigma. Fruit in woody racemes

of 2—5, ovoid-elliptic, 2 cm. long and 1-3—1-5 cm. broad, pinkish-

orange and with a longitudinal circumferential groove; stalk 5

+44

Vol. XVI. (1958).

> C

E 4 mm

B 4 mm

Fig. 56. Gymnacranthera eugeniifolia (A. DC.) J. Sinclair var. eugeniifolia.

A, Leafy twig with male inflorescences. B, Male flower. C, Staminal

column. D, Female inflorescences. E, Female flower. F, Ovary. G,

Twig with fruit. A-C from Curtis 3768. D—-F from Alvins, sine

data, Malacca. G from Haniff S.F.N. 9056.

445

Gardens Bulletin, S.

mm.—1 cm. long. Seed ellipsoid, mottled, smooth, covered by the

red aril which is laciniate to the base.

KEDAH: Bukit Malut, Darus F.D. 7682 (K, SING); Bigia Enggang

F.R., Sik, Ali bin Din K.F.N. 73766 (KEP).

PENANG: Curtis 2563 (SING); back of West Hill, Curtis 804 (CAL,

K, SING); Highland Hill, Haniff & Nur 3020 (CAL, K, SING); road

to the Spout, Government Hill, Curtis 487 (K, SING); Mount Olivia,

Haniff S.F.N. 9056 (BRI, SING, UC); Pantai Achie F.R., Telok Ba-

hang, Ishmail bin Ali K.F.N. 66379 (KEP); without locality, Gaudi-

chaud 116 (FI, G & Prodr. holotype of M. eugeniifolia, P).

PERAK: Scortechini, Sept. 1887 (CAL); Larut, King Nos. 3833

(CAL, DD, US); 5408: (CAL, DD, K, MEL, SING, UPS); 6620

(CAL, DD, E, FI, L, P); 6631 (CAL; FL’°G, L); 6652. (04

SING, UPS); 6932 (CAL) and 7481 (CAL, DD, E, FI, SING, UPS);

Gopeng, King Nos. 4640 (CAL, FI, G, K, L, MEL); 5801 (CAL, DD,

E, FI, G, K, L, MEL); 6141 (BM, CAL, DD, FI, G, K, MEL, SING);

Waterfall Hill, Wray 2084 (CAL, DD, SING); Tapah, Wray 1436

(CAL, SING).

SELANGOR: Batu Tiga, Curtis 3768 (CAL, K, SING).

NEGRI SEMBILAN: Gunong Angsi Res., Kuala Pilah, Syed Ali F.D.

23691 (SING).

MALACCA: Maingay 1290 (K); Maingay 1306 (A, BM, K, L); Alvins

1275 (SING); Alvins, 12th April, 1886 (SING); Hervey, Aug. 1886

(CAL). .

JOHORE: Kulai, Corner S.F.N. 29955 (A, K, SING); Sungei Sedili,

Negadiman S.F.N. 36909 (DD, K, KEP, SING); Bukit Patani, Batu

Pahat, Ridley 11029 (CAL, K, SING).

SINGAPORE: Maingay Nos. 1302 (CAL, K, L) and 1303 (A, CAL,

K, L); Woodlands, Ridley 11646 (SING); Cantley, no data (A, SING);

Bukit Timah, Ngadiman S.F. Nos. 34628 tree 43 (A, DD, K, SING);

36491 tree 400 (SING) and 36446 tree 357 (SING); near potting

shed, Botanic Gardens, Ridley 2102 (CAL, K, MEL, SING); Sungei

Bajau, Ridley 3366 (K, SING).

DISTRIBUTION: Sumatra and Lingga (Pulau Sinkep).

This species is close to G. forbesii especially in the flower struc-

ture, but the flowers of G. forbesii are slightly larger and the female

ones are in denser clusters. The most obvious difference is in the

leaves. In the present species, they are more coriaceous and nar-

rower; the veins are not prominent on the lower surface and lack

the double loop of interarching as in those of forbesii. The fruit

too is smaller and not so elongated and the pedicel just a trifle

thicker. The twigs are a richer brown and the youngest are shorter,

more slender and lack the angled, shining, grey appearance of

jorbesii. The bark appears to be of a darker brown colour and

probably has more abundant sap of a blood red colour while the

bark of forbesii in the few samples seen is grey-brown and the

paler sap does not appear to be copious. The question of sap and

bark, however, requires checking.

The name Myristica farquhariana was applied by A. DC.

to the Indian plant (which King later called M. canarica and

446

Vol. XVI. (1958).

Warburg subsequently Gymnacranthera canarica) since the mate-

rial quoted by Hk. f. et Th. in Flora Indica was a mixtum

compositum and consisted of Indian, Malayan and Philippine spe-

cimens. The name G. farquhariana will stand for the Indian plant,

eugeniifolia is the next oldest name available for the Malayan while

A. DC. chose the name paniculata for the Philippine. All three

species are very close as has already been stated.

(4) G. eugeniifolia var. griffithii (Warb.) J. Sinclair, comb. nov.

Basionym: G. farquhariana var. griffithii (Hk. f.) Warb.,

Monog. Myrist. (1897) 368; Gamble, Mat. F.M.P. 5, 23 (1912)

226 pro parte; Ridley, F.M.P. 3 (1924) 62.

Synonyms: Myristica griffithii Hk. f., Fl. Br. Ind. 5 (1886)

109; King in Ann. Roy. Bot. Gard. Calc. 3 (1891) 304 PI. 145

excl. Curtis 2406 and 2458. M. farquhariana Wall. Cat. 6795,

sphalmate 6798, nom. nudum; Hk. f. et Th., Fl. Ind. (1855)

162; Hk. f. Fl. Br. Ind. 5 (1886) 108; King in Ann. Roy. Bot.

Gard. Calc. 3 (1891) 305 Pl. 135 [omnino pro parte]. G. farqu-

hariana (Hk. f. et Th.) Warb. sensu Warb., Monog. Myrist.

(1897) 365; Gamble, Mat. F.M.P. 5, 23 (1912) 226; Ridley,

F.M.P. 3 (1924) 62. M. farquhariana var. major King in Ann.

Roy. Bot. Gard. Calc. 3 (1891) 306 Pl. 136. Fig. 4. G. farqu-

hariana var. major (King) Gamble, Mat. F.M.P. 5, 23 (1912)

226; Ridley, F.M.P. 3 (1924) 62. G. apiculata Warb., Monog.

Myrist. (1897) 359 T. 20 Fig. 1-2. M. apiculata (Warb.)

Boerl., Handl. Fl. Ned. Ind. 3, 1 (1900) 88 nom alt.—Fig. 57.

A few stilt roots up to 1 m. long present, not numerous. Sap

rather scant, red. Leaves sub-coriaceous, glossy-green above and

glaucous beneath, drying olive-green and retaining the gloss,

medium brown beneath often with a glaucous tinge, oblong-elliptic,

often obovate-elliptic, broadening just below the acute or slightly

apiculate apex, cuneate or occasionally slightly rounded at the base;

midrib flat above, lying in a groove; nerves 6—9 pairs, sometimes a

secondary one present between two main ones, fine above but dis-

tinctly raised (a good diagnostic character), also raised beneath,

those next the base oblique, those above the middle of the leaf more

curved. Male flowers rather lax; pedicels slender, 5—6 mm. long;

perianth 4-5 mm. long. Female flowers not seen. Fruit on a short,

1—2 cm. long axis, pinkish or apricot-yellow, sub-globose or ob-

long, up to 2-8 cm. long and 2-2-5 cm. in diam., on stout staiks

1 cm. long and 3-4 mm. thick.

Perak: Larut, King Nos. 6548 (CAL, DD, E, FI, K, L) type mate-

rial of var. major; 6622 (BM, CAL, DD, G, K, L, SING) type material

of var. major; 6736 (CAL, MEL, UPS) type material of var. major;

B.P.D., King 7928 (CAL, FI, G. K, L) type material of var. major;

447

Gardens Bulletin, S.

Dies

JURA IM) DE 4 om

Fig. 57. Gymnacranthera eugeniifolia (A. DC.) J. Sinclair var. griffithii

(Warb.) J. Sinclair.

A, Leafy twig with fruit. B, Seed with aril. C, Male inflorescence. D,

Male flower. E, Staminal column. A and B from Sinclair S.F.N.

40716. C-E from Ridley 3365.

448

Vol. XVI. (1958).

Taiping, Wray Nos. 2399 (CAL, K, SING) type material of var.

major and 2695 (CAL, FI, K, SING) type material of var. major.

SELANGOR: Telok F.R., Wyatt-Smith K.F.N. 6608 (KEP); Sungei

Tinggi, Kuala Selangor, Nur S.F.N. 34093 (A, K, KEP, L, SING).

MALACCA: Griffith 4355 (A, CAL, FI, K, M) type material of var.

major; Griffith 4356 (A, CAL, K, P) type material of M. griffithii.

JOHORE: Tanjong Kupang, Ridley 3365 (K, SING); Sungei Kayu, S.

Sedili, Corner S.F.N. 29499 (K, SING); S. Kayu, Kiah S.F.N. 32157

(A, BM, BRI, DD, K, KEP, SING); ridge of Gunong Panti, Corner

S.F.N. 32218 (SING); Pengkalan Raja, Pontian, Ngadiman, 2nd July,

1939 (SING); Tempayan River, Ridley 13263 (K, SING).

SINGAPORE: Wall. Cat. 6795 (BM, K, M) syntype of M. farquhariana

Hk. f. et Th.; Wall. ex Herb. Brown (CAL); Chan Chu Kang, Ridley

Nos. 1834 (CAL, SING) and 3961 (CAL, K, SING); Mandai Road,

Kiah S.F.N. 37718 (A, BRI, DD, K, KEP, SING) and Kiah, 26th

July, 1940 (SING); Nee Soon, Seletar Forest, Sinclair S.F.N. 40716

(A, BM, BO, E, K, L, SING).

DISTRIBUTION: Sarawak, N. Borneo and Banka.

A tree of the fresh water swamp or peat forest. It is in many

ways intermediate between G. forbesii and G. eugeniifolia. A bark

specimen, Nur S.F.N. 34093 was found to be intermediate in

colour between the two. The twigs are intermediate in size and

appearance, being slightly angled but not so sharply angled as in

forbesti. They are less slender than those of eugeniifolia but not so

stout as in the other. The leaves have raised veins on the lower sur-

face and thus are easily distinguished from eugeniifolia. The veins

on the upper surface are also raised but are more prominent than

those of forbesii. The double loops of interarching, so characteristic

of forbesii are sometimes present but are very faint. The texture ap-

proaches that of eugeniifolia but is usually just a trifle less coriace-

ous. The flowers too, are slightly larger than in the latter. The fruit,

in shape, is nearer to that of the latter but is larger and almost

globose. The fruiting pedicels are stouter than in the other two.

There are, in addition, specimens which are intermediate be-

tween the variety griffithii and both forbesii and eugeniifolia. Here

it is difficult to decide which of the three taxa should claim them.

One lot resembles eugeniifolia except that the veins on the lower

surface are raised a trifle but not so much as in the variety. The

veins on the upper surface are more distinct than in eugeniifolia.

Less common are forms nearer to forbesii but with some charac-

ters of eugeniifolia. The following are intermediate forms:—

KELANTAN: Bertam, Osman bin Salleh K.F.N. 65176 (KEP) G.

eugeniifolia approaching var. griffithii.

SELANGOR: Bukit Lagong F.R., Wyatt-Smith K.F.N. 60617 (KEP)

nearest to forbesii but approaching eugeniifolia; Kuala Lumpur, Motan

K.F.N. 51949 (KEP) intermediate between eugeniifolia and var.

griffithii.

NEGRI SEMBILAN: Sungei Menyala, Wood K.F.N. 71886 (KEP)

intermediate between eugeniifolia and var. griffithii.

449

Gardens Bulletin, S.

Both the Malayan syntypes of M. farquhariana, Wall. Cat. 6795

and Griffith 4355 are identical, having raised veins on the lower

surface of the leaf. King confused M. farquhariana with what I

have here called G. eugeniifolia (A. DC.) Sinclair and also with

some other species, e.g. Maingay 1293 = G. forbesii. What he

should have considered as the type from of his interpretation of

M. farquhariana represented by Griffith 4355 (an original syntype

of the species) he described as var. major. This view is supported

by a large number of quoted specimens. On account of this the

ternary epithet major becomes superfluous under the Code and so

unavailable here. The next available varietal epithet is griffithii

for this variety of G. eugeniifolia with the veins raised on the

lower surface of the leaf. I cannot see any differences between M.

griffithii, M. farquhariana var. major King and G. apiculata. All

these I reduce to G. eugeniifolia var. griffithii. The Penang speci-

mens, Curtis 2406 and 2458 quoted by King and Gamble as

Myristica griffithii or G. farquhariana var. griffithii must be ex-

cluded. They are Horsfieldia penangiana J. Sinclair, sp. nov.

450

LIST OF COLLECTORS’ NUMBERS

These specimens have all been seen by me. G = Gymnacran-

thera, H — Horsfieldia, K = Knema, M = Myristica.

ABDUL MAJID—K.F.N. 68860 M. gigantea.

ABDUL RAHMAN—C.F. 368 H. flocculosa; C.F. 1802 M. maxima;

2802 K. furfuracea.

ABDULLA—K.F.N. 51737 K. globularia; K.F.N. 59620 H. polys-

pherula.

ABDULLA & ARIFFIN—F.D. 12209 K. malayana.

ABDULLA BIN ISMAIL—K.F.N. 68058 H. subglobosa var. bra-

chiata.

ABDULLA BIN MD. SHARIFF—K.F.N. 69957 M. lowiana.

ABIDI—K.F.N. 54653 M. gigantea; F.D. 54697 M. gigantea.

ABuU—C.F. 3317 H. bracteosa.

AHMAD—C.F. 3024 K. laurina; C.F. 4581 G. forbesii; C.F. 4867

H. superba; C.F. 5081 M. malaccensis; C.F. 5423 G. forbesii;

June 1926 H. superba.

AHMAD BIN IBRAHIM—K.F.N. 53361 M. maingayi; 55352 M.

iners.

AHMAT—C.F. 2487 H. polyspherula.

AKIL & JA’*AMAT—F.D. 12088 H. polyspherula.

ALI—K.F.N. 73767 H. sucosa.

ALI BIN DIN—K.F.N. 73766 G. eugeniifolia.

ALvINS—33 K. globularia; 34 K. intermedia; 262 M. elliptica;

385 H. polyspherula; 525 K. intermedia; 530 M. elliptica;

575 M. crassa; 653 K. intermedia; 657 H. wallichii;

672 & 680 H. tomentosa; 703 K. curtisii; 783 K. laurina;

853 H. punctatifolia; 854 G. contracta; 896 K. malayana;

897 G. contracta; 898 H. punctatifolia; 901 K. laurina; 938

H. macrocoma var. canarioides; 961 H. irya; 1198 H. polys-

pherula; 1275 G. eugeniifolia; 1288 H. crassifolia; 1349 H.

sucosa; 1798 H. irya; 2071 & 2074 M. malaccensis; 14th

December, 1885 M. crassa; 16th January, 1886 K. globularia;

17th April, 1886 G. eugeniifolia.

AMAT—K.F.N. 65903 M. gigantea.

ANDERSON—9 H. crassifolia; 10 K. hookeriana; 11 M. fragrans.

ANNANDALE—28th January, 1916 K. globularia.

ARNOT—F.D. 14801 G, forbesii.

451

Gardens Bulletin, S.

AwaNnc—F.D. 31350 H. polyspherula; 42382 H. sucosa; K.F.N.

52079 H. fulva; K.F.N. 68556 H. irya.

AWANG LILA or LELA—C.F. 2683 G. forbesii.

AWANG PIAH—F.D. 40372 M. maxima.

Aziz—F.D, 46238 M. lowiana.

BAIN—F.D. 3959 H. subglobosa.

BARNARD—C.F. 30 H. subglobosa.

BATTEN-PooLL—November 1939 to January 1940 H. tomentosa.

BEsT—S.F.N. 8292 M. elliptica; S.F.N. 21260 K. hookeriana.

- Biin—F.D. 15449 H. irya; F.D. 15611 H. irya; F.D. 15671 G.

forbesii; K.F.N. 53659 M. gigantea.

BOswWELL—F.D. 12595 K. furfuracea.

BROWNE—43462 H. polyspherula.

BURGESS—K.F.N. 69783 M. maingayli.

BURKILL—AIl numbers preceded by S.F.N.—877 K. sipbulsmeet

897 H. irya; 944 K. globularia; 1014 M. iners; 1111 K. cur-

tisii; 1278 H. sucosa; 1536 M. elliptica; 2032 K. hookeriana;

4582 & 6261 K. intermedia; 6334 G. forbesii; 6514 H. polys-

pherula; 6850 K. plumulosa; date 1913 K. hookeriana; June

1915 K. globularia; June 1915 H. subglobosa; 30th March,

1920 H. grandis; 16th June, 1921 H. bivalvis; 10th Septem-

ber, 1921 M. guatteriifolia.

BURKILL & HANIFF—AlIl numbers preceded by S.F.N.—15665 G.

forbesii; 16394 H. flocculosa; 16448 M. elliptica; 17053 K.

glaucescens f. rubens; 17167 H. tomentosa.

BURKILL & HOLTTUM—S.F.N. 8679 H. glabra.

BURN-MurpDocH—10 K. glaucescens var. patentinervia; 44 & 48

H. tomentosa; 101 K. furfuracea; 161 K. kunstleri; 179 K.

malayana; 265 M. cinnamomea; 279 K. plumulosa (or

Hashim); 373 M. elliptica; 380 K. communis; 14247 K. plu-

mulosa; date 1909 K. malayana.

CALDER—K.F.N. 54695 H. punctatifolia.

CANTLEY—13 K. hookeriana; 20 K. malayana; 21 K. latericia; 29

K. intermedia; 30 H. tomentosa; 31 K. globularia; 35 M.

crassa; 69 K. latericia; 83 K. intermedia; 195 K. laurina; 449

K. intermedia; 1798 H. ridleyana; 2261 K. laurina; 2851 K.

malayana; 2904 K. hookeriana; 2919 K. malayana; 3083 K.

laurina; 25th January, 1882 H. bivalvis; date 1885 K.

latericia. ‘4

452

Vol. XVI. (1958).

CARRIER—F.D. 27317 K. rigidifolia.

CHIK—K.F.N. 65681 M. maingayi.

CoRNER—AIl numbers preceded by S.F.N.—21333 K. intermedia;

25856 H. irya; 25916 K. furfuracea; 25954 M. elliptica;

25964 H. irya; 25991 H. subglobosa var. brachiata; 26050

K. intermedia; 26055 M. elliptica; 26155 K. curtisii; 28097

M. iners; 28131 M. lowiana; 28133 M. guatteriifolia; 28176

K. glaucescens; 28186 M. iners; 28322 M. elliptica; 28440

K. malayana; 28453 K. intermedia; 28493 H. irya; 28504

K. malayana; 28505 K. conferta; 28512 M. guatteriifolia;

28584 M. elliptica; 28615 K. laurina; 28649 K. glaucescens

var. patentinervia; 28671 M. cinnamomea; 28703 H. superba;

28707 H. subglobosa var. brachiata; 28711 K. latericia;

28712 G. forbesii; 28962 K. glaucescens; 28963 H. subglo-

bosa var. brachiata; 28970 G. forbesii; 28975 K. latericia;

29015 K. glaucescens var. cordata; 29252 & 29268 K. inter-

media; 29275 K. glaucescens; 29277 & 29290 K. latericia;

29310 H. bracteosa; 29366 H. polyspherula; 29367 K. plu-

mulosa; 29402 M. elliptica; 29403 K. glaucescens f. rubens;

29419 K. glaucescens var. cordata; 29420 K. latericia; 29436

K. glaucescens var. cordata; 29466 H. subglobosa var.

brachiata; 29477 K. glaucescens var. cordata; 29499 G.

eugeniifolia var. griffithii; 29824 & 29838 K. globularia; 29841

M. guatteriifolia; 29855 & 29859 K. globularia; 29944 K.

plumulosa; 29945 G. bancana; 29955 G. eugeniifolia; 30043

& 30088 K. stenophylla; 30181 M. cinnamomea; 30243 M.

elliptica; 30284 K. mandaharan; 30339 K. glaucescens var.

patentinervia; 30388 M. cinnamomea; 30482 K. glaucescens

var. patentinervia; 30486 K. mandaharan; 30489 M. elliptica;

30556 H. subglobosa; 30754 K. globularia; 30972 and

31473 K. plumulosa; 31617 H. subglobosa var. brachiata;

32209 K. malayana; 32218 G. eugeniifolia var. griffithii; 32281

H. subglobosa var. brachiata; 33140 H. wallichii; 33145 K.

conferta; 33226 K. rigidifolia; 33555 M. maingayi; 33556

H. wallichii; 33597 G. bancana; 33598 K. latericia; 37078

K. laurina; 34439 H. wallichii; 34542 H. crassifolia; 34544

K. intermedia; 34905 H. crassifolia; 34912 M. crassa; 36134

H. superba; 36282 M. iners; 36951 K. laurina; 36961

G. forbesii; 37078 K. laurina; 37134 M. iners; 37147

H. polyspherula; 37275 M. elliptica; 20th July, 1932 M.

maingayi; 28th August, 1932 K. mandaharan & K. fur-

furacea; 19th October, 1932 M. lowiana; 5th March, 1933

H. polyspherula; 15th April, 1934 K. curtisii; 20th June,

453

Gardens Bulletin, S.

1934 M. guatteriifolia; 23rd June, 1934 H. superba; 7th Oct- |

ober, 1934 H. subglobosa var. brachiata; November 1934 M. i

guatteriifolia; 10th March, 1935 K. laurina; 12th March, 1935 |

K. hookeriana; 14th April, 1935 K. malayana; 10th May,

1935 K. oblongifolia var. monticola; K. latericia & H.

bracteosa; Ist November, 1935 K. glaucescens var. patenti-

nervia & H. superba; 7th November, 1935 H. bracteosa;

K. mandaharan & K. glaucescens var. patentinervia; 8th

November, 1935 H. superba; 10th November, 1935 K. man-

daharan; 11th November, 1935 M. maingayi; 12th Novem-

ber, 1935 K. mandaharan; 14th November, 1935 K. laurina;

12th April, 1936 H. crassifolia; 24th January, 1937 K. glau-

cescens var. patentinervia; 30th January, 1937 M. guatterii-

folia; 15th March, 1937 H. wallichii; 25th April, 1937 H.

irya; 12th August, 1937 K. oblongifolia var. monticola; 2nd

March, 1938 M. iners; 18th June, 1939 M. iners; 16th July,

1939 K. glaucescens var. patentinervia; M. cinnamomea &

H. bracteosa; 24th July, 1940 K. glaucescens var. patenti-

nervia; 7th August, 1940 K. glaucescens var. patentinervia; :

15th November, 1941 K. furfuracea; 22nd November, 1941 |

K. globularia.

CORNER & HENDERSON—S.F.N. 36603 M. lowiana.

‘CORPORAL—442 K. conferta.

CoUSENS—K.F.N. 65588 & K.F.N. 69775 K. glaucescens f.

rubens; K.F.N. 69777 H. wallichii; K.F.N. 69783 M. main-

gayl; K.F.N. 69791 H. wallichii.

‘CUBITT—F.D. 9680 H. sucosa.

CUMING—2315 K. globularia; 2418 M. fragrans.

CuRTIs—202 K. hookeriana; 487 G.-eugeniifolia; 700 K. globu-

laria; 804 G. eugeniifolia; 934 H. macrocoma var. canari-

oides; 935 K. globularia; 936 H. irya; 1024 K. curtisii; 1044

K. intermedia & K. laurina; 1122 M. elliptica; 1191 K.

laurina; 1197 H. tomentosa; 1320 K. curtisii; 1459 K. fur-

furacea; 1497 M. maxima; 1559 K. globularia; 1748 H.

tomentosa; 2050 K. oblongifolia var. monticola; 2051 K.

kunstleri; 2406 H. penangiana; 2423 H. wallichii; 2455 M.

iners; 2456 K. furfuracea; 2457 K. laurina; 2458 H. penan-

giana; 2479 K. hookeriana; 2563 G. eugeniifolia; 2728 K.

furfuracea; 2769 K. furfuracea; 2770 K. laurina; 2827 K.

furfuracea; 2843 K. laurina; 2925 K. globularia; 2966 H.

superba; 3402 K. intermedia; 3679 K. stenophylla; 3735 H.

454

Vol. XVI. (1958).

bracteosa; February 1890 K. malayana; June 1890 K. globu-

- Jaria; March 1892 H. macrocoma var. canarioides; June 1890

K. globularia. | |

Darus—F.D. 7682 G. eugeniifolia; F.D. 12408 H. macrocoma

var. canarioides.

Daup—F.D. 8153 K. giobularia.

DENNy—December 1947 H. subglobosa; 27th June and September

1949 H. bracteosa; 20th March and 20th May, 1954 H.

subglobosa.

DERRY—134 H. subglobosa; 139 K. laurina; 152 K. furfuracea;

485 K. laurina; 648 K. kunstleri; 649 H. subglobosa; 909 M.

elliptica & K. glaucescens mounted on same sheet; 967 H.

tomentosa; 979 K. laurina; 1033 M. iners; 1038 K. glauces-

cens var. patentinervia; 1136 H. polyspherula; 1149 K. inter-

media; 1163 H. crassifolia; 1204 K. curtisii; 1216 H. polys-

pherula; date 1892 M. fragrans.

DoLMAN—F.D. 21488 K. laurina.

FLEMMICH—24917 M. gigantea.

Fox—13 H. tomentosa; 11275 M. elliptica; 12649 M. cinnamo-

mea; December 1894 K. globularia.

FoxworTHY—C.F. 2363. K. glaucescens var. patentinervia; F.D.

3215 M. cinnamomea.

FRANCK—473 K. globularia.

FurTADOo—Gardens’ No. 1194 K. imtermedia; S.F.N. 21153;

21154; 21177 K. latericia; S.F.N. 34818 H. bivalvis; S.F.N.

34860 M. fragrans; 6th January, 1928 M. guatteriifolia; 11th

January, 1928 K. furfuracea; 22nd September, 1929 K.

hookeriana.

GAUDICHAUD—114 H. tomentosa; 116 G. eugeniifolia.

GOODENOUGH—1279 H. superba; 1317 & 1355 G. forbesu; 1471

& 1632 M. elliptica; 1800 (or Ridley) K. laurina; 1819 H.

crassifolia; 1845 G. forbesii; 1854 K. curtisii; 1994 H. sucosa;

2001 K. furfuracea; 3376 K. curtisii; 3831 H. crassifolia;

3890 G. bancana; 4132 H. crassifolia; 5572 M. cinnamomea;

7122 K. globularia; 27th November, 1890 K. latericia.

GOODENOUGH & RIDLEY-—1621 M. lowiana.

GRIFFITH—4342 K. hookeriana; 4343 K. malayana; 4344 K.

globularia; 4345 K. conferta; 4346 K. furfuracea; 4349 K.

glaucescens; 4350 H. crassifolia; 4351 H. subglobosa var.

455

Gardens Bulletin, S.

brachiata; 4352 M. fragrans; 4353 M. philippensis; 4554 H.

polyspherula; 4355 & 4356 G. eugeniifolia var. griffithii; 4357

H. irya; 4358 H. amygdalina; 4359 K. intermedia.

Hamip—The following numbers preceded by C.F.—561 K. glau-

cescens var. patentinervia; 934 K. furfuracea; 981 K. kunstleri;

2958 K. globularia; 6445 M. fragrans. The following numbers

preceded by F.D.—10583 K. glaucescens f. rubens; 10584

K. laurina; 10693 M. elliptica; 10864 G. forbesii; 10888 M.

elliptica; 24951 H. flocculosa; 46212 M. lowiana; 20th May,

1918 H. superba; 8th June, 1918 G. forbesii.

HAMID & JA’AMAT—F.D. 10913 M. elliptica.

HANIFF—AIl numbers preceded by S.F.N.—328 & 339 K. inter-

media; 376 K. globularia; 3660 H. wallichii; 3855 H. sub-

globosa var. brachiata; 9056 G. eugeniifolia; 9063 K.

laurina; 9072 M. iners; 10401 K. globularia; 13127 K.

kunstleri; 13147 & 14260 K. furfuracea; 14310 K. glauces-

cens; 14334 H. subglobosa var. brachiata; 14917 M. fragrans;

15497 K. globularia; 15510 K, furfuracea; 21024 K. oblongi-

folia var. monticola; April 1901 M. elliptica; 4th April, 1917

H. tomentosa.

HANIFF & NUR—AIl numbers preceded by S.F.N.—2477 H. sub-

globosa; 2736 K. globularia; 3020 G. eugeniifolia; 7455 H.

irya; 7500 & 7569 K. globularia; 10256 K. stenophylla; 10387

H. subglobosa var. brachiata.

HAsHIM—25 G. forbesii; 279 K. plumulosa; C.F. 2455 G. forbesii.

HassAN—7571 M. iners.

HENDERSON—The following numbers preceded by F.M.S. Mus.

Herb.—1342 H. bivalvis; 10714 K. malayana; 10767 K.

laurina; 11261 K. rigidifolia; 11545 K. oblongifolia var. mon-

ticola; 11604 H. tomentosa. The following numbers preceded

by S.F.N.—18220 K. curtisii; 18244 K. globularia; 19684 K.

kunstleri; 21675 K. curtisii; 21684 K. hookeriana; 21804 H.

tomentosa; 21810 K. laurina; 22061 H. tomentosa; 22978 K.

globularia; 23328 K. rigidifolia; 23795 K. laurina; 23869 H.

subglobosa var. brachiata; 24521 H. bracteosa; 24817 K.

glaucescens; 29516 H. bracteosa; 29519 K. laurina; 29531 K.

glaucescens; 35779 H. subglobosa; 37918 M. maingayi; 10th

April, 1930 K. rigidifolia.

HeRvEY—August 1886 K. globularia & G. eugeniifolia; date

1891 H. superba.

HoOLMBERG—773 K. furfuracea; 818 G. forbesii; 2100 H.

superba.

456

Vol. XVI. (1958).

‘HOLTTUM—AIl numbers preceded by S.F.N.—9304 H. grandis;

9425 K. glaucescens f. rubens; 9867 K. laurina; 10606 K.

latericia; 15156 K. globularia; 36392 M. elliptica; 37781 H.

sucosa.

HuLLeETT—314 K. hookeriana; 590 H. subglobosa; 5739 and 7th

May, 1893 K. intermedia.

HUME—AIl numbers preceded by F.M.S. Mus. Herb.—7952 M.

iners; 9241 & 9261 K. oblongifolia; 9929 K. laurina.

IBRAHIM BIN HUSSIN—K.F.N. 72752 M. lowiana.

IsSHAK—F.D. 7685 H. macrocoma var. canarioides.

ISMAIL BIN AMIN—K.F.N. 66379 G. eugeniifolia.

JA’AMAT—AIl numbers preceded by F.D.—14913 K. stenophylla;

14938 K. malayana; 15265 H. wallichii; 15330 K. malayana;

16507 H. subglobosa, var. brachiata; 25160 H. tomentosa;

28191 K. oblongifolia var. monticola, 43436 K. communis;

43479 K. plumulosa; 46501 K. laurina.

JA’7AMAT & TACHUN—F.D. 56304 H. fulva.

JANTAN—F.D. 26347 M. lowiana. }

JIDIN—F.D. 18409 & 18410 M. gigantea.

JINAL—F.D. 8831 K. furfuracea; F.D. 20456 K. kunstleri.

KaLonc—F.D. 20313 M. cinnamomea; 20324 H. subglobosa var.

brachiata; 22434 K. rigidifolia.

KEHDING—98 H. punctatifolia; 150 K. stenophylla; 192 K. plu-

mulosa.

KiaH—AIl numbers preceded by S.F.N.—24381 M. elliptica;

31951 K. glaucescens; 31976 H. polyspherula; 32004 M.

elliptica; 32034 K. conferta; 32064 H. subglobosa; 32105 H.

crassifolia; 32109 H. polyspherula; 32130 K. plumulosa;

32157 G. eugeniifolia var. griffithii; 32193 H. bancana; 32314

H. flocculosa; 32334 & 32379 K. laurina; 32384 K. glauces-

cens; 35070 K. laurina; 35097 K. malayana; 35135 H. tomen-

tosa; 35156 K. malayana; 35176 K. scortechinii; 35330 K.

furfuracea; 35398 K. laurina; 35975; 37123 & 37148 K.

curtisii; 37710 H. crassifolia; 37718 G. eugeniifolia var.

griffithii; 39448 H. irya, 26th July, 1940 G. eugeniifolia var.

egriffithii & H. bracteosa.

KiAH & STRUGNELL—S.F.N. 24034 K. kunstleri.

KiAI—F.D. 8263 M. malaccensis; F.D. 8265 M. cinnamomea;

F.D. 8288 K. plumulosa.

457

Gardens Bulletin. S.

Kinc—( Usually as King’s collector, Kunstler) —835 K. oblongi-

folia; 1057 M. cinnamomea; 1233 H. grandis; 1372 K. lau-

rina; 1551 H. tomentosa; 1677 M. fragrans; 2614 & 2743

K. kunstleri; 2758 M. crassa; 3309 H. polyspherula; 3350

K. plumulosa; 3372; 3393 & 3510 K. kunstleri; 3534 M. cin-

namomea; 3576 K. kunstleri; 3582 K. oblongifolia var. mon-

ticola; 3620 H. subglobosa; 3732 M. elliptica; 3783 G.

forbesii; 3810 K. oblongifolia var. monticola; 3833 G. eugenii-

folia; 3899 H. subglobosa; 4078 H. punctatifolia; 4150 K.

kunstleri; 4165 H. tomentosa; 4216 K. kunstleri; 4267 H.

crassifolia; 4276 M. elliptica; 4307 K. laurina; 4352 K.

malayana; 4414 K. kunstleri; 4426 M. elliptica; 4475 M.

crassa; 4534 K. oblongifolia; 4605 K. kunstleri; 4640 G.

eugeniifolia; 4647 H. sucosa; 4703 M. elliptica; 4704 H. sub-

globosa var. brachiata; 4706 H. crassifolia; 4827 H. wallichii;

4949 K. kunstleri; 5059 H. subglobosa; 5065 M. crassa;-5092

K. laurina; 5198 & 5288 M. elliptica; 5299 K. glaucescens;

5317 K. plumulosa; 5353 M. cinnamomea; 5408 G. eugenii-

folia; 5419 K. intermedia; 5458 M. cinnamomea; 5513 M.

maxima; 5536 H. polyspherula; 5537 M. crassa; 5600 K.

furfuracea; 5614 K. plumulosa; 5617 K. scortechinii; 5671

H. tomentosa; 5706 K. malayana; 5720 K. furfuracea; 5726

K. malayana; 5754 K. hookeriana; 5801 G. eugeniifolia;

5819 K. furfuracea; 5866 M. gigantea; 5867 K. kunstleri;

5939 K. scortechinii; 5983 K. oblongifolia; 6001 M. crassa; .

6004 H. subglobosa; 6007 K. hookeriana; 6025 K. furfuracea; |

6043 K. scortechinii; 6050 M. gigantea; 6059 K. furfuracea;

6061 M. crassa; 6102 H. tomentosa; 6128 K. malayana; 6141 q

G. eugeniifolia; 6146 K. intermedia; 6211 K. conferta; 6330

K. oblongifolia var. monticola; 6371 K. intermedia; 6440 K.

kunstleri; 6514 K. glaucescens; 6515 K. mandaharan; 6521 :

K. glaucescens; 6548 G. eugeniifolia var. griffithii; 6566 H.

subglobosa; 6569 K. plumulosa; 6591 G. forbesii; 6620 G.

eugeniifolia; 6622 G. eugeniifolia var. griffithii; 6631 & 6652

G. eugeniifolia; 6656 K. hookeriana; 6672 H. subglobosa; __

6688 H. crassifolia; 6694 K. scortechinii; 6696 M. cinnamo-

mea; 6704 K. intermedia; 6736 G. eugeniifolia var. griffithii;

6737 & 6765 H. subglobosa; 6771 H. subglobosa var. bra-

chiata; 6784 G. forbesii; 6867 K. plumulosa; 6932 G. eugenii-

folia; 6960 M. maxima; 6973 G. forbesii; 7180 K. kunstleri

7258 M. crassa; 7290 K. plumulosa; 7419 G. forbesii; 7447

H. irya; 7452 K. laurina; 7474 M. cinnamomea; 7475 K. glau-

cescens; 7481 G. eugeniifolia; 7526 H. polyspherula; 7551 K,

furfuracea; 7576 K. intermedia; 7599 K. malayana; 7645 G.

458

Vol. XVI. (1958).

forbesii; 7686 K. laurina; 7690 K. retusa; 7732 G. forbesii;

7756 M. crassa; 7926 K. scortechinii; 7928 G. eugeniifolia

var. griffithii; 7965 H. subglobosa; 7998 H. tomentosa; 8024

H. superba; 8159 G. forbesii; 8277 K. glaucescens; 8322 K.

oblongifolia var. monticola; 8443 K. plumulosa; 8552 H.

tomentosa; 8559 M. elliptica; 8618 H. flocculosa; 8642 H.

tomentosa; 8645 K. laurina; 8722 & 8756 G. forbesii; 8443

K. plumulosa; 10022 K. kunstleri; 10038 M. crassa; 10064

H. macrocoma var. canarioides; 10080 G. forbesii; 10194 H.

macromoca var. canarioides; 10256 H. polyspherula; 10286

K. communis; 10295 K. conferta; 10321 M. elliptica; 10349

K. furfuracea; 10386 H. tomentosa; 10413 H. subglobosa;

10431 H. polyspherula; 10444 K. glaucescens; 10475 H.

sucosa; 10513 H. polyspherula; 10549 K. kunstleri; 10557 H.

tomentosa; 10562 H. macrocoma var. canarioides; 10594 K.

malayana; 10635 K. scortechinii; 10691 K. glaucescens;

10749 M. malaccensis; 10816 H. macrocoma var. canarioides;

10826 & 10845 K. kunstleri; 10917 H. ridleyana; 10953 H.

oblongifolia var. monticola; 10826 K. kunstleri; 8th Septem-

ber, 1879 K. laurina. |

KLoss—6699 K. globularia.

KOCHUMMEN—K.F.N. 80627 H. fulva; K.F.N. 83809 K. steno-

phylla.

KOsTERMANS—August 1938 H. crassifolia.

_ Lake & KELSALL—4011 K. glaucescens; 4012 K. latericia; 6th

November, 1892 K. malayana.

LAMBAK—C.F, 4169 K. globularia.

LATTIF (A.) BIN DOMAL—K.F.N. 70063 M. maingayi.

LELAH BIN KHAMIS—K.F.N. 65569 M. lowiana.

LiEw—All numbers preceded by S.F.N.—36441 K. intermedia;

37254 & 37256 K. conferta; 37258 H. crassifolia.

LINDONG—55792 K. intermedia; 70915 H. subglobosa.

Lopp—315 K. intermedia.

LupIN—C.F. 1873 K. giaucescens var. patentinervia.

MAHAMED; MoHAMUD; MoHAMAD; MoHuamMup—(Variants of

spelling; there is probably more than one collector involved)

—C.F. 881 & C.F. 3724 M. cinnamomea; F.D. 14960 K.

glaucescens var. patentinervia; F.D. 14999 K. globularia; F.D.

17127 G. forbesii; F.D. 17570 K. hookeriana; F.D. 204i1

H. ridleyana; F.D. 22519 K. rigidifolia.

459

Gardens Bulletin, S.

MAINGAY—1279 K. hookeriana; 1280 K. glaucescens var. patenti-

nervia & K. malayana; 1280/2 K. glaucescens var. patenti-

nervia; 1281 K. intermedia; 1282 K. globularia; 1283 & 1284

H. wallichii; 1285 M. fragrans; 1286 H. polyspherula & H.

subglobosa; 1287 K. furfuracea; 1288 K. intermedia (Kew)

& K. laurina (Arnold Arboretum); 1289 M. maingayi; 1290

& 1291 G. eugeniifolia; 1292 H. irya; 1293 G. forbesii; 1294

K. laurina; 1295 G. forbesii; 1296 M. elliptica; 1297 K. con-

ferta; 1298 H. macrocoma var. canarioides; 1299 K. mala-

yana; 1300 H. sucosa; 1301 K. curtisii; 1302 & 1303 G.

eugeniifolia; 1304 H. fulva; 1305 M. malaccensis; 1306 G.

eugeniifolia; 1306/2 H. sucosa.

MANaAP BIN A. RAHMAN—K.F.N. 51208 M. maingayi.

MatT—3538 K. glaucescens; 5972 K. intermedia; F.D. 17946 K.

laurina; date 1892 M. elliptica.

MAT ArIFF—F.D. 7502 H. polyspherula.

MAT GANI—F.D. 9777 M. crassa; F.D. 9785 H. crassifolia.

Mat HassaN—K.F.N. 69416 K. hookeriana; K.F.N. 69417 M.

elliptica.

MaT Nonc—C.F. 4010 M. cinnamomea.

MaT YassIiM—F.D. 14080 H. sucosa.

MaT ZIN—F.D. 27513 K. furfuracea.

MeEH—F.D. 10176 K. globularia; F.D. 10191 H. tomentosa; F.D.

42268 K. malayana.

MERAH—20th May, 1937 H. wallichii.

MILSUM—S.F.N. 34442 H. sucosa.

Moup. JAYA—K.F.N. 53688 M. gigantea.

MOHD. BIN Poz—K.F.N. 62873 M. maingayi.

Monp. RANI—K.F.N. 65949 K. hookeriana.

Moup. SALLEH—K.F.N. 66360 H. subglobosa var. brachiata.

Moupb. SALLEH BIN YUSSOF—K.F.N. 60474 H. wallichii; K.F.N.

73852 H. macrocoma var. canarioides.

Moup. YAKIM—C.F. 515 M. elliptica.

Moupb. YATIM—K.F.N. 66860 H. wallichii; K.F.N. 74246 H.

superba; K.F.N. 74452 M. maingayi.

MotaN—AIl numbers preceded by K.F.N.—51948 H. punctati-

folia; 53930 & 70412 K. plumulosa; 70460 K. furfuracea;

70473 H. superba. |

MoysEy & KiAH—S.F.N. 33753 H. subglobosa var. brachiata;

S.F.N. 33839 K. laurina.

460

Vol. XVI. (1958).

MurTON—13 G. bancana; 76 H. punctatifolia; 149 H. bivalvis;

150 K. latericia; 1481 H. wallichii.

NASRUDDIN—K.F.N. 71247 H. macrocoma var. canarioides.

NAUEN—S.F.N. 35848 K. hookeriana; S.F.N. 35852 & S.F.N.

38091 M. elliptica.

NGADIMAN—ALIl numbers preceded by S.F.N.—34520 H. wallichii;

34628 G. eugeniifolia; 34630 H. polyspherula; 34708 H. sub-

globosa var. brachiata; 34714 K. conferta; 34729 M. ellipiica;

34740 M. maxima; 34742 H. macrocoma var. canarioides;

34952 M. elliptica; 34957 K. latericia; 34966 K. laurina;

35901 K. glaucescens; 35923 M. elliptica; 36122 H. polys-

pherula; 36141 H. superba; 36146 K. laurina; 36446 &

36491 G. eugeniifolia; 36622 M. lowiana; 36789 H. brac-

teosa; 36791 H. subglobosa var. brachiata; 36831 H. grandis;

36863 M. iners; 36866 G. forbesii; 36880 M. lowiana; 36900

M. maingayi; 36909 G. eugeniifolia; 36918 G. forbesii;

36919 M. crassa; 36924 H. macrocoma var. canarioides;

36927 K. laurina; 36935 K. glaucescens var. patentinervia;

2nd July, 1939 G. eugeniifolia var. griffithii.

NcAaH—F.D. 41778; 41779 & 41784 M. gigantea.

Nur—These numbers preceded by S.F.N.—10532 K. rigidifolia;

11126 K. oblongifolia var. monticola; 11249 K. oblongifolia

var. monticola; 11313 K. rigidifolia; 11668 K. laurina; 21740

K. globularia; 21749 H. irya; 21751 H. tomentosa; 26166 K.

hookeriana; 34001 M. lowiana; 34043 & 34046 K. intermedia;

34051 H. crassifolia; 34093 G. eugeniifolia var. griffithii;

34117 H. subglobosa; 34124 M. lowiana; 34125 K. intermedia;

34145 H. irya; 34307 M. maxima—The following are Singa-

pore Gardens’ Numbers—13 H. bivalvis; 402 H. hookeriana;

1424 K. furfuracea; 1459 K. latericia; 1466 K. furfuracea;

1770 & 1872 H. irya; 11th June, 1924 H. superba; 16th; 17th

June, 1924 & 4th July, 1928 M. fragrans; 18th June, 1929 H.

sucosa; 20th April, 1937 K. oblongifolia var. monticola.

Nur & FoxwortHy—S.F.N. 12121 K. intermedia.

Nur & KiAHn—S.F.N. 7757 K. latericia; S.F.N. 7806 K. steno-

phylla.

OMAR—C.F. 805 & F.D. 7973 H. superba; 8529 K. laurina.

OsMAN—F.D. 28358 K. oblongifolia var. monticola; F.D. 29253

K. stenophylla.

PADUKA BIN ABDULLA—K.F.N. 65054 M. gigantea.

PANDAK—F.D. 29861 G. forbesii.

PAWANCHEE—F.D. 12942 H. polyspherula.

461

Gardens Bulletin, S.

PENGUHULU BEBAS—134 & 136 H. subglobosa.

PHILLIPS—Date 1824 K. furfuracea & H. tomentosa.

PURSEGLOVE—4212 H. glabra; 4213 K. rigidifolia; 4246; 4275 &

4276 K. oblongifolia var. monticola.

RAFFLES—1382 K. hookeriana.

RAaJU—K.F.N. 62883 H. punctatifolia.

REJAB GURUN—K.F.N. 73503 K. scortechinii.

RIDLEY—44 K. glaucescens var. patentinervia; 274 & 331 M.

elliptica; 361 H. crassifolia; 393 H. bivalvis; 442 K. conferta;

1310 H. irya; 1541 K. globularia; 1621 (or Goodenough) M.

lowiana; 1646 K. intermedia; 1800 (or Goodenough) K.

laurina; 1820 K. intermedia; 1828 H. crassifolia; 1833 K.

malayana; 1834 G. eugeniifolia var. griffithii; 1835 G. bancana;

1854 K. curtisii; 2037 K. latericia; 2039 K. intermedia; 2040

M. guatteriifolia; 2042 H. polyspherula; 2043 K. latericia;

2044 K. laurina; 2101 H. superba; 2107 K. intermedia; 2108

H. polyspherula; 2109 K. hookeriana; 2261 K. furfuracea;

2262 & 2263 K. globularia; 2264 K. laurina; 3043 H. irya;

3171 H. tomentosa; 3313 H. macrocoma var. canarioides;

3362 K. hookeriana; 3363 M. maxima; 3364 G. bancana;

& 3581 M. cinnamomea; 3701 K. hookeriana; 3705 M. guat-

teriifolia; 3755 K. laurina; 3961 G. eugeniifolia var. griffithii;

4133 H. grandis; 4136 K. malayana; 4162 H. polyspherula;

4422 H. wallichii; 4438 & 4439 H. subglobosa; 4813 K. hoo-

keriana; 4814 H. irya; 4815 G. bancana; 4816 & 4817 K.

globularia; 4818 K. intermedia; 4819 K. glaucescens f. rubens;

4827 H. grandis & K. intermedia; 5061 H. wallichii; 5065 K.

intermedia; 5489 H. tomentosa; 5825 H. irya; 5826 H. crassi-

folia; 5971 (or Mat) H. subglobosa; 6095 H. macrocoma

var. canarioides; 6126 H. crassifolia; 6157 G. forbesii; 6266

M. cinnamomea; 6270 G. forbesii; 6324 & 6355 H. macro-

coma var. canarioides; 6447 K. malayana; 6448 & 6449 H.

subglobosa; 6450 K. latericia; 6451 H. subglobosa & M.

lowiana; 6558 & 6559 H. sucosa; 6693 H. macrocoma var.

canarioides; 6734 K. intermedia; 6735 K. conferta; 6736 H.

bivalvis; 6737 K. latericia; 6909 H. crassifolia; 6920 M.

elliptica; 7205 H. tomentosa; 7206 H. irya; 7627 K. laurina;

7629 H. polyspherula; 8040 H. crassifolia; 8504 probably K.

glaucescens var. cordata; 8906 H. macrocoma var. canarioi-

des; 8930 H. polyspherula; 8957 H. irya; 9222 K. intermedia;

9374 K. curtisii; 9461 H. bracteosa; 9464 G. forbesii; 9587

K. kunstleri; 10227 K. plumulosa; 10240 H. tomentosa;

462

Vol. XVI. (1958).

10695 H. crassifolia; 10786 K. intermedia; 10787 H. tomen-

tosa; 10839 H. subglobosa; 10922 M. guatteriifolia; 11028

K. glaucescens var. patentinervia; 11029 G. eugeniifolia;

11270 H. macrocoma var. canarioides; 11328 H. irya; 11354

& 11355 H. sucosa; 11646 G. eugeniifolia; 11919 & 12155

H. subglobosa; 12541 K. latericia; 13263 G. eugeniifolia var.

griffithii; 13264 K. glaucescens 13342 K. globularia; 14681 K.

curtisii; 14957 H. irya; 14958 & 14959 K. globularia; 15344

M. elliptica; 16177 H. ridleyana; 39355 H. bivalvis; 28th

January, 1891 M. elliptica; date 1892 M. elliptica; K. con-

ferta; K. laurina; H. sucosa & H. wallichii; date 1893 M.

elliptica; date 1894 H. grandis; February 1894 K. conferta;

7th May, 1895 H. macrocoma var. canarioides; May 1896 K.

oblongifolia; November 1900 M. elliptica; date 1906 H.

grandis & K. latericia; March 1915 K. laurina; 4th February,

1917 K. oblongifolia var. monticola.

ROBINSON—6383 K. globularia.

RosE—K.F.N. 73782 M. elliptica.

SALLEH—9785 H. crassifolia.

SALLEH BIN ISA—K.F.N. 63906 M. gigantea.

SCORTECHINI—54b K. laurina; 79b M. elliptica; 175a K. kunstleri;

178a K. scortechinii; 184a H. fulva; 211a H. polyspherula;

246a H. wallichii; 292 K. curtisii; 619b H. macrocoma var.

canarioides; 631 K. intermedia; 763 K. curtisii; 803 K. inter-

media; 830 & 831 K. laurina; 837 H. subglobosa; 862 H. rid-

leyana; 1527 K. stenophylla; 1551 M. lowiana; 1649 H. sub-

globosa var. brachiata; 1855 M. lowiana; 1872 M. maxima;

1949 M. gigantea; 1964 M. elliptica; September 1887 G.

eugeniifolia.

SEIMUND—F.M.S. Mus. Herb. 67 & 1186 K. malayana; 10th

March, 1926 K. stenophylla.

SHARIFF BIN OMAR—K.F.N. 66387 K. hookeriana; K.F.N. 66396

M. iners.

SINCLAIR—AIl numbers preceded by S.F.N.—37936 G. forbesii;

38561 K. latericia; 38704 K. oblongifolia var. monticola;

38913 K. communis; 38914 & 38915 K. latericia; 39240 K.

globularia; 39258 M. fragrans; 39338 M. maxima; 39474 K.

communis; 39475 K. laurina; 39481 K. latericia; 39486 &

39487 H. wallichii; 39488 K. furfuracea; 39490 M. crassa;

39491 K. laurina; 39502 G. bancana; 39503 M. elliptica;

39505. K. conferta; 39518 & 39524 K. stenophylla; 39533 H.

crassifolia; 39538 H. superba; 39539 G. forbesii; 39564 K.

463

Gardens Bulletin, S.

intermedia; 39570 K. communis; 39571 K. intermedia; 39581

G. contracta; 39599 H. subglobosa; 39601 H. polyspherula;

39661 M. elliptica; 39697 H. subglobosa; 39937 H. brac-

teosa; 39959 K. glaucescens var. cordata; 39989 H. sub-

globosa; 40026 K. conferta; 40043 H. subglobosa; 40045 G.

bancana; 40046 K. conferta; 40047 H. superba; 40048 M.

iners; 40080 K. furfuracea; 40082 H. punctatifolia; 40155

H. sucosa; 40170 H. fulva; 40174 H. superba; 40202 H.

irya; 40211 H. punctatifolia; 40216 H. wallichii; 40256 H.

crassifolia; 40280 K. malayana; 40287 K. scortechinii; 40288

H. polyspherula; 40304 M. maxima; 40319 M. lowiana;

40320 K. communis; 40362 & 40374 M. cinnamomea; 40480

M. elliptica: 40493 K. malayana; 40522 K. communis;

40567 H. crassifolia; 40570 H. superba; 40591 K. inter-

media; 40599 M. maingayi; 40603 & 40612 K. globularia;

40621 & 40629 H. bracteosa; 40681 & 40711 H. polysphe-

rula; 40713 K. glaucescens f. rubens; 40714 & 40717 K.

malayana; 40739 H. irya; 40753 K. mandaharan; 40772 H.

subglobosa var. brachiata; 40826 K. hookeriana; 40845 K.

glaucescens var. cordata; 40853 M. cinnamomea; 40898 H.

crassifolia; 7th January, 1950 M. elliptica; 21st January, 1953

M. maingayi & H. irya; 19th February, 1953 H. irya; 24th

February, 1953 M. guatteriifolia; 26th February, 1953 K.

intermedia; 18th March, 1953 H. wallichii & H. irya; 8th

May, 1953 H. punctatifolia; 10th June, 1953 K. intermedia;

11th June, 1953 K. malayana; 24th February, 1954 H. sucosa;

8th April, 1954 H. irya.

SINCLAIR & KIAH—S.F.N. 39937 H. bracteosa.

SoH or SOW—AII numbers preceded by F.D.—15091 H. sub-

globosa; 15130 M. lowiana; 15408 H. irya; 15484 H. polys-

pherula; 15728 H. sucosa; 41658 K. malayana. The following

number preceded by K.F.N.—64427 M. malaccensis.

SoH & LINDONG—K.F.N. 70486 H. crassifolia; K.F.N. 70489 K.

intermedia.

SoH & TACHUN—F.D. 16450 H. subglobosa; F.D. 16852 K.

malayana.

SOMERVILLE—F.D. 10456 K. scortechinii; F.D. 14588 K. steno-

phylla.

SPARE—S.F.N. 33267 & 33270 H. irya; S.F.N. 34569 H. sub-

globosa var. brachiata.

STRUGNELL—AIl numbers preceded by F.D.—11151 G. forbesii;

12104 K. kunstleri; 12117 M. gigantea; 12132 M. maxima;

12472 M. cinnamomea; 12477 & 12638 H. wallichii; 12699

464

Vol. XVI. (1958).

H. punctatifolia; 12726 M. cinnamomea; 12729 H. flocculosa;

12895 K. hookeriana; 13391 & 13953 M. elliptica; 22519 K.

rigidifolia; 27906 M. maxima.

SULEIMAN BIN MANJA—AIl numbers preceded by K.F.N.—70179

M. maingayi; 70180 H. wallichii; 70190 G. bancana; 70213

K. hookeriana; 70220 M. iners.

SYED ALI—F.D. 23691 G. eugeniifolia.

SYMINGTON—A// numbers preceded by F.D.—22775 K. laurina;

24081 G. forbesii; 30107 H. ridleyana; 30138 H. macrocoma

var. canarioides; 31061 M. malaccensis; 36245 H. glabra;

40627 K. laurina; 51757; 51845; 51846 & 56704 H. fulva;

57034 M. malaccensis.

SYMINGTON & KIAH—S.F.N. 28755 K. glaucescens f. rubens;

S.F.N. 28775 K. kunstleri.

TAMBEY CHIK—K.F.N. 65681 M. maingayi.

TERUYA—355 H. crassifolia.

WALLICH—6785 & a, b, c, d, e as M. moschata & M. aromatica is

M. fragrans; 6786 as M. dactyloides is M. malabarica; 6787

as M. notha is M. malabarica; 6788 as M. missionis is K. gio-

bularia; 6789 as M. heyneana Wall. is M. malabarica and

leaves of Lauraceae; 6790 as M. glaucescens Wall. is Lau-

raceae; 6791 as M. attenuata Wall. is K. attenuata; 6792 M.

montana Wall. is M. montana; 6793 as M. finlaysoniana Wall.

is Fissistigma fulgens; 6794 as M. lanceolata Wall. is K. glo-

bularia; 6795 as M. farquhariana Wall. is G. eugeniifolia var.

griffithii; 6796 as M. sphaerocarpa Wall. is H. irya; 6797 as M.

amygdalina Wall. is H. amygdalina; 6798 M. elliptica; 6799

as M. integra Wall. is H. subglobosa and an unidentified

plant; 6800 as M. sylvestris is M. philippensis; 6801 as M.

longifolia Wall. is K. linifolia; 6802a as M. hookeriana is

K. hookeriana; 6802b is probably Lauraceae; 6803 as M.

ferruginea Wall. is G. bancana; 6804 as M. exaltata Wall. is

H. macrocoma var. canarioides; H. amygdalina & H. irya;

6805 as M. floribunda Wall. is H. amygdalina; 6806 as M.

horsfieldia is H. crassifolia, H. polyspherula & H. wallichii;

6807 as M. micrantha Wall. is H. irya; 6808 as M. obtusifolia

Wall. is an unidentified plant, not Myristicaceae; 6809 as

M. sesquipedalis is Actinodaphne sesquipedalis; 6810 as

Knema glaucescens Jack is K. plumulosa.

WALTON—F.D. 28396 K. furfuracea; F.D. 44933 H. fulva.

WARBURG—4853 K. intermedia.

465

Gardens Bulletin, S.

WaTson—C.F, 537 H. subglobosa; F.D. 17017 K. communis;

F.D. 32678 H. irya.

WELLS—F.D. 3038 H. subglobosa.

Woop—K.F.N. 76058 K. mandaharan; K.F.N. 76070 M. cinna-

momea; K.F.N. 76076 K. curtisii.

Wray—122 H. subglobosa; 176 M. malayana; 285 K. scortechinii;

467 H. subalpina; 606 K. glaucescens; 646 M. crassa; 993;

1077 & 1087 K. oblongifolia var. monticola; 1212 K. curtisii;

1214 M. iners; 1422 K. scortechinii; 1429 G. forbesii; 1436

G. eugeniifolia; 1736 M. elliptica; 2006 K. glaucescens; 2056

K. kunstleri; 2064 H. subglobosa; 2084 G. eugeniifolia; 2088

H. polyspherula; 2112 K. curtisii; 2130 K. intermedia; 2218

H. subglobosa; 2345 M. elliptica; 2377 K. conferta; 2399 &

2695 G. eugeniifolia var. griffithii; 2700 K. plumulosa; 2705 H.

subglobosa; 3010 K. intermedia; 3026 M. cinnamomea; 3071

H. crassifolia; 3126 K. plumulosa; 3985 K. kunstleri.

WYATT-SMITH—AIl numbers preceded by K.F.N.—63176 M.

malaccensis; 63706 K. scortechinii; 64076 K. hookeriana;

64086 G. forbesii; 64095 K. laurina; 64264 H. irya; 65531

M. maingayi; 65536 M. iners; 66456 H. punctatifolia; 66540

M. maxima; 66607 K. laurina; 66608 G. eugeniifolia var.

griffithii; 68301 K. laurina; 68679 M. gigantea; 70357 H. wal-

lichii; 71179 H. irya; 71188 M. iners; 71198 K. globularia;

71200 K. laurina; 71228 K. laurina; 71271 M. maingayi;

71371 H. bracteosa; 76520 K. stenophylla; 76531 H. fulva;

76532 K. furfuracea; 15th September, 1950 H. fulva; 76525

H. irya.

WyYATT-SMITH & SOW—K.F.N. 52149 M. malaccensis. ’

YAHAYA—S.F.N. 35768 H. sucosa. ;

YAKIM—C.F. 505 M. gigantea; F.D. 7723 H. subglobosa var. !

brachiata.

YATTIM—K.F.N. 53610 M. maingayi.

YEOB—F.D. 5920 G. forbesii.

YEOP—F.D. 3233 M. cinnamomea; F.D. 93524 K. stenophylla.

YEOB BIN ABDUL RAHIM—F.D. 3103 M. cinnamomea.

ZABID—K.F.N. 54691 H. punctatifolia.

466

Index

New species, new varieties, new names and new combinations in bold-faced type.

Brochoneura, 213, 219, 223, 224, 225,

22s. 230, 232, 236, 237,240, 241.

chapelieri (Baill.) H. Perr., 222,

232; 235, 242:

Cephalosphaera, 213, 223, 224, 230,

236, 237, 240.

Coelocaryon, 213, 223, 224, 230, 236,

Za7.

Compsoneura, 213, 222, 224, 227,

mo, 232, 236, 237, 240, 242,

244, 245.

debils (A. DC.) Warb., 214.

Dialyanthera, 213, 219, 222, 224, 227,

Zoe, 252,. 235, 236, 237, 244,

245.

gordoniifolia (A. DC.) Warb., 219.

latialata Pittier, 219.

lemannii A. Cc. Smith, 219.

Embelia ridleyi K. et G., 389.

Gymnacranthera, 434.

apiculata Warb., 447.

bancana (Mig.) J. Sinclair.

var. bancana, 436.

var. borneensis (Warb.) J. Sin-

clair, 439.

canarica (King) Warb., 444.

contracta Warb., 439.

eugeniifolia (A. DC.) J. Sinclair.

var. eugeniifolia, 444.

var. griffithii (Warb.) J. Sinclair,

447.

farquhariana (Hk. f. et Th.) Warb.,

447

farquhariana (Hk. f. et Th.) Warb.,

sensu Warb., 444, 447.

var. griffithii (Hk. f.) Warb., 447.

var. major (King) Gamble, 447.

forbesii (King) Warb., 441.

467

ibutii Holth., 393, 435.

murtonii (Hk. f.) Warb., 436.

var. borneensis Warb., 439.

paniculata (A. DC.) Warb., 435,

stenophylla Warb., 300.

Horsfieldia, 368.

amklaal Kanehira, 383.

amygdalina (Wall.) Warb., 370,

413, 419, 422.

ardisiifolia (A. DC.) Warb., 218.

bivalvis (Hk. f.) Merr., 379.

bracteosa Hend., 419.

canarioides (King) Warb., 389.

congestifiora A. C. Sm., 383.

costulata (Mig.) Warb., 411.

crassifolia (Hk. f. et Th.) Warb.,

386.

flocculosa (King) Warb., 398.

fulva (King) Warb., 396.

var. paludicola (King) Warb.,

386.

glabra (Bl.) Warb., 411.

globularia (Bl.) Warb., 379.

grandis (Hk. f.) Warb., 400.

irya (Gaertn.) Warb., 382.

iryaghedhi (Gaertn.) Warb., 227,

244, 369.

kingii (Hk. f.) Warb., 370, 413,

422

laevigata (Bl.) Warb., 282.

lemanniana (A. DC.) Warb., 383,

422.

leptocarpa Warb., 393.

macrocoma (Mig.) Warb.

var. canarioides (King) J. Sin-

clair, 389.

var. macrocoma, 393.

var. rufirachis J. Sinclair, 393.

macrothyrsa (Miq.) Warb., 411.

majuscula (King) Warb., 425.

merrillii Warb., 389.

montana Airy-Shaw, 219.

motleyi Warb., 400.

novo-guineensis Warb., 218.

oblongata Merr., 392, 389.

odorata Willd., 368.

papillosa Warb., 389.

penangiana J. Sinclair, 408.

polyspherula (Hk. f. emend. King)

J. Sinclair, 422.

prainii (King) Warb., 389.

punctatifolia J. Sinclair, 413.

racemosa (King) Warb., 389.

ridleyana (King) Warb., 432.

subalpina J. Sinclair, 410.

subglobosa (Mig.) Warb.

var. hams (King) J. Sinclair,

var. subglobosa, 426.

sucosa (King) Warb., 416.

superba (Hk. f. et Th.) Warb., 393.

sylvestris (Houtt.) Warb., 218, 219.

thorelii Lecomte, 422.

tomentosa Warb., 403.

trifida A. C. Sm., 392.

valida (Miq.) Warb., 411.

wallichii (Hk. f. et Th.) Warb.,

405.

Iryanthera, 213, 222, 224, 225, 227,

229 23252555 2a Lae eee

2A0, 241,242) 244,245.

macrophylla (Benth.) Warb., 219,

220.

paradoxa (Schwacke) Warb., 221.

sagotina (Benth.) Warb., 226.

Knema, 254.

alvarezii Merr., 293.

angustifolia (Roxb.) Warb., 328.

attenuata (Hk. f. et Th.) Warb.,

328.

badia Merr., 281.

cantleyi (Hk. f.) Warb. sensu

Warb., 312.

communis J. Sinclair, 297.

conferta (King) Warb., 286.

var. scortechinii Warb., 288.

coriacea Warb., 291.

corticosa Lour., 325.

var. tonkinensis Warb., 325.

curtisii (King) Warb., 281.

elegans Warb., 332.

furfuracea (Hk. f. et Th.) Warb.,

275.

geminata (Miq.) Warb., 302.

glauca (Bl.) Warb., 302.

var. andamanica Warb., 303.

var. nicobarica Warb., 303.

var. sumatrana Warb., 302.

var. typica Warb., 302.

glaucescens Jack, 302.

var. cordata J. Sinclair, 310.

var. glaucescens, 304.

f. rubens J. Sinclair, 307.

var. patentinervium J. Sinclair,

308.

Gardens Bulletin, S.

globularia (Lamk) Warb., 325.

eee (Hk. f. et Th.) Warb.,

intermedia (Bl.) Warb., 315.

var. dubia Warb., 312.

kunstleri (King) Warb., 291.

latericia Elm., 278.

laurina (Bl.) Warb., 329.

lenta Warb., 288.

malayana Warb., 293.

mandaharan (Miq.) Warb., 319.

membranifolia Winkler, 283.

meridionalis J. Sinclair, 278.

missionis (King) Warb., 325.

oblongata Merr., 329.

oblongifolia (King) Warb.

var. Peres (King) Warb.,

23.

var. oblongifolia, 321.

obovoidea Mertr., 329.

palembanica (Miq.) Warb., 302.

parvifolia Merr., 291.

pectinata Warb., 288.

petelotii Merr., 325.

plumulosa J. Sinclair, 312.

pulchra (Miq.) Warb., 310.

retusa (King) Warb., 318.

rigidifolia J. Sinclair, 284.

rufa Warb., 318.

scortechinii (King) J. Sinclair, 288.

sphaerula (Hk. f.) Airy-Shaw, 325.

stenophylla (Warb.) J. Sinclair, 300.

tomentella Warb., 255.

wangii H. H. Hu, 328.

wrayi (King) Warb., 303.

yunnanensis H. H. Hu, 328.

Mauloutchia, 213.

Myristica, 333.

amboinensis Gandoger, 361.

amplifolia Warb., 436.

apiculata (Warb.) Boerl., 447.

argentea Warb., 215, 239.

aromatica Lamk, 361.

avis-paradisiacae Warb., 213.

bancana Miq., 436.

bivalvis Hk. f., 379.

borneensis Warb., 356.

brachiata King, 431.

bracteata King non DC., 339.

calocarpa Migq., 353.

canarica King, 444, 446.

canarioides King, 389.

cantleyi Hk. f. sensu King, 312.

cantleyi Hk. f. non sensu King, 329.

celebica Miq., 356.

ceylanica A. DC., 335.

cinnamomea King, 358.

collettiana King, 425.

conferta King, 286.

Vol. XVI. (1958).

contracta (Warb.) Boerl., 439.

cookii Warb., 350.

‘coriacea (Warb.) Boerl., 291.

corticosa (Lour.) Hk. f. et Th.,

294, 302, 325.

var. decipiens Miq., 315.

var. lanceolata Miq., 302.

var. sumatrana Miq., 302.

crassa King, 366.

crassifolia Hk. f. et Th., 386.

curtisii King, 281.

elliptica Hk. f. et Th.

var. celebica (Mig.) J. Sinclair,

356.

var. elliptica, 353.

var. simiarum (A. DC.) J. Sin-

clair, 356. ;

eugeniifolia A. DC., 444.

‘exaltata Wall. Cat., 382, 386.

farquhariana Hk. f. et Th., 444.

farquhariana Wall. Cat., 447.

var. major King, 447.

fatua Houtt., 215, 220.

ferruginea King, 436.

flocculosa King, 398.

forbesii King, 441.

fragrans Houtt., 361.

furfuracea Hk. f. et Th., 275.

furfuracea Hk. f. et Th. sensu A.

Be... 312:

furfurascens Gandoger, 329.

fulva King, 396.

geminata Miq., 302.

gigantea King, 343.

glabra Bl., 411.

var. sumatrana Migq., 425.

glabra Miq., 425.

glabra de Vriese, 315.

glauca Bl., 302.

glaucescens Wall. Cat., 312.

glaucescens Hk. f. et Th., 293, 302,

2k ae

globosa Warb., 335.

globularia Bl., 379.

var. subglobosa Miq., 425.

globularia Bl., sensu Hk. f. et Th.,

422.

globularia Lamk, 293, 325.

globularia Miq., 411.

grandis Hk. f., 400.

griffithii Hk. f., 447.

guatteriifolia A. DC., 350.

sie: aan Pierre ex Lecomte,

63.

hollrungii Warb., 218, 335.

hookeriana Hk. f. et Th., 272.

ae Wall. Cat., 386, 403,

425.

‘ 469

iners Bl., 363.

intermedia Bl., 315.

iteophylla Miq., 315.

javanica Bl., 383.

kunstleri King, 291.

laevigata Miq., 411.

lanceolata Wall. Cat., 325.

laurella Gandoger, 361.

laurina Bl., 329.

var. borneensis Miq., 289.

lemanniana A. DC., 383.

lepidota BI., 335.

littoralis Miq., 350.

longifolia Wall. ex. Bl. sensu Hk. f.

ef "Pn. 275.

lowiana King, 345.

maingayi Hk. f., 348.

majuscula King, 425.

malabarica Lamk, 239.

malaccensis Hk. f., 356.

malaccensis Gandoger, 356.

malayana (Warb.) Boerl., 294.

mandaharan Migq., 319.

maxima Warb., 339.

micrantha Wall. Cat., 383.

missionis King, 325.

moschata Thumb., 361.

murtonii Hk. f., 436.

var. borneensis (Warb.) Boerl.,

439.

oblongifolia King

var. monticola King, 323.

var. oblongifolia, 321.

officinalis L. f., 361.

palembanica, Miq., 322.

paludicola King, 386.

pandurifolia Winkler, 356.

papillosa (Warb.) Boerl., 389.

philippensis Lamk, 221, 341.

philippinensis Gandoger, 361.

polyspherula Hk. f., 426.

prainii King, 389.

pulchra Migq., 310.

racemosa King, 389.

retusa King, 318.

ridleyana King, 432.

ridleyi Gandoger, 278.

riedelii Warb., 351.

rubiginosa King, 400.

schleinitzii Engl., 219.

scortechinii King, 288.

speciosa Warb., 215.

sphaerula Hk. f., 325.

sphaerocarpa Wall., 383.

stenophylla (Warb.) Boerl., 300.

suavis King, 366.

subalulata Miq., 218, 219,

subglobosa Miq., 425.

sublanceolata Miq., 363.

sucosa King, 416.

superba Hk. f. et Th., 393.

sycocarpa Miq., 353.

teijsmannii Miq., 368.

tomentosa Bl. non Thb., 329.

tomentosa Hk. f. et Th., 403.

tristis Warb., 218.

tubiflora Bl., 234, 235.

velutina Mef, 218.

villosa Warb., 256.

vordermannii Warb., 363.

vriesiana Migq., 383. ~*

wallichii Hk. f. et Th., 405.

wrayi King, 303.

wyatt-smithii Airy- Shaw, 356.

Myristicophyllum, 246.

Osteophloeum, 213, 214, 222, 224,

227, Z29,, 232, 235, cae ee.

Gardens Bulletin, S.

Pycnanthus, 213, 223, 224, 227, 230,

232, .230, ay: 245. -

angolensis (Welw.) Exell, 219.

Scyphocephalium, 213, 223, 224, 228,

230, 234, 236,°237.

Staudtia, 213, 223, 224, 230, 236, 237,

240.

— — «= _-s

Virola, 213, 221, 222, 224.227.2290

232, 233.236.2000

bicuhyba (Schott) Warb., 214.

calophylla Warb., 221, 222.

rugulosa Warb., 219.

sessilis (A. DC.) Warb., 211, 214,

219; 221.

subsessilis (Benth.) Warb., 212.

surinamensis (Rol.) Warb., 239.

Names which are synonyms are printed in italics.

237.

Arundinaria glaucescens _(Willd.)

Beauv., 67.

Arundo bambos Linn., 59.

Arundo multiplex Lour., 68.

Bambos arundinacea Retz., 52, 59.

Bambusa, 52.

Bambusa apus Schult., 112.

arundinacea (Retz.) Willd., 59.

aspera Schult., 100.

bambos auctt., 59, 61.

blumeana Schult., 57.

burmanica Gamble, 62.

elegans Ridl., 75.

glaucescens (Willd.) Sieb., 67.

heterostachya (Munro) Holtt., 16,

klossii Ridl., 23, 79.

latispiculata (Gamble) Holtt., 65.

levis Blanco, 119.

longinodis Miquel, 42.

magica Ridl., 75.

montana (Ridl.) Holtt.,

multiplex auctt., 68.

nana Roxb., 67.

pauciflora Ridl., 76.

ridleyi Gamble, 20, 22, 23, 71.

spinosa Roxb., 57, 62.

spinosa Blume, 57.

stricta Roxb., 86.

ventricosa McClure, 70.

vulgaris Schrad., 63.

uae

470

var. stricta (Lodd.), 63.

wrayi Stapf, 73.

Cephalostachyum, 33.

Cephalostachyum malayanum Ridl.,

90.

virgatum, 22.

Dendrocalamus, 86.

Dendrocalamus asper (Schult.) Backer,

100, 122.

dumosus (Ridl.) Holtt., 96.

elegans (Ridl.) Holtt., 95.

flagellifer Munro, 100.

giganteus Munro, 103.

hirtellus Ridl., 93.

pendulus Ridl., 11, 90.

sinuatus (Gamble) Holtt., 97.

strictus (Roxb.) Nees, 98. .

Dendrochloa, 34, 50. +f

Dinochloa, 81. 4

Dinochloa montana Ridl., 77.

scandens (Bl.) O. Ktze., 84. *

tijankorreh (Schult.) Buse, 84. 7

sp., 85

Gigantochloa, 104.

sage a: apus (Schult.) Kurz,

1 £2;

aspera (Schult.) Kurz, 100.

hasskarliana (Kurz) Backer, 118.

heterostachya Munro, 65.

kurzii Gamble, 112, 124,

Vol. XVI. (1958).

latifolia Ridl., 132.

var. fimbriata Holtt., 133.

var. alba Holtt., 133.

latispiculata Gamble, 65.

levis (Blanco) Merrill, 119.

ligulata Gamble, 129.

maxima Kurz, 114.

var. viridis Holtt., 115.

var. minor Holtt., 116.

ridleyi Holtt., 127.

scortechinii Gamble, 11, 23, 122.

var. albovestita Holtt., 124.

scribneriana Merrill, 119.

verticillata auctt., 114, 116, 119.

wrayi Gamble, 23, 124.

Ludolfia glaucescens Willd., 67.

Melocanna gracilis Munro, 37.

Nastus ? tjankorreh Schult., 84.

Neohouzeana, 34.

Ochlandra ridleyi Gamble, 48.

Oxytenanthera sinuata Gamble

nigrociliata sensu Munro, 118.

Racemobambos, 134.

Racemobambos setifera Holtt., 134.

Schizostachyum, 31.

Schizostachyum aciculare Gamble, 39.

blumii auctt., 40.

brachycladum Kurz, 22, 25, 45.

var. auriculatum Holtt., 47.

chilianthum auctt., 42.

dumosum Ridl., 96.

elegans Ridl., 95.

gracile (Munro) Holtt., 25, 37.

grande Ridl., 49.

? hasskarilianum Kurz, 118.

insulare Ridl., 47.

jaculans Holtt., 40, 74.

latifolium Gamble, 48.

longispiculatum Kurz, 48.

ridleyi (Gamble) Holtt., 48.

subcordatum Ridl., 90.

tenue Gamble, 37.

terminale Holtt., 51.

zollingeri Steud., 11, 42.

Teinostachyum, 33.

Thyrsostachys, 80.

Thyrsostachys siamensis Gamble, 80.

Synonyms in italics. New names in bold face.

Astasiaceae, 188.

Key to the genera, 188.

Astasia

variabilis, 189, 190.

Colacium, 156.

vesiculosum, 153, 156.

Cyclidiopsidaceae, 187.

Cycliodiopsis

acus, 188, 189.

Distigma

curvatum, 189.

f minor, 192.

proteus, 189, 191.

Entosiphon, 200.

obliquum, 189, 201.

sulcatum, 198, 201.

Euglenaceae, Key to Genera, 138.

Euglena, Key to Malayan species, 139.

acus, 142, 144.

var. hyalina, 144.

caudata, 142, 147.

chlorophoenicia, 142, 144.

cyclopicola, 142, 143.

ehrenbergii, 142, 145.

elongata, 140, 142.

fusca, 142, 145.

granulata, 142, 148.

mainxii, 141, 142.

oxyuris

var. charkowiensis, 142, 146.

proxima, 142, 143.

refringens, 142, 147.

reticulata, 141.

Sanguinea, 142, 149.

splendens, 142, 148.

subehrenbergii, 142, 145.

synchlora, 141, 152.

tripteris, 142, 147.

Eugleninae, Colourless:

Key to the families, 187.

Heteronema spp. Key, 196.

invaginatum, 197, 198.

leptosomum, 197, 198.

polymorphum, 197, 198.

Lepocinclis, Key to Species, 150.

Lepocinclis

acicularis, 153, 155.

marsonni, 153, 155.

var. angustata, 153, 154.

butschii, 152, 153.

colliferum, 153, 154.

conica, 152, 153.

deflandreana, 152, 153.

dimidio-minor, 152, 153.

globula, 152, 153.

gracelicauda, 152, 153.

major, 152, 153.

verrucosum, 153, 154.

salina, 153, <155.

steinii, 153, 155,

texta

var. mamillata, 151, 153.

Menoidium

obtusum, 189, 191.

pellucidum, 189, 190.

Notosolenus, 199.

lens, 198, 200.

orbicularis, 198, 200.

similis, 198, 199.

stenochismos, 198, 200.

Petalomonas spp. Key, 194.

abcissa

v. pellucida, 189, 195.

asymmetrica, 189, 196.

heptaptera, 189, 196.

mediocanellata, 189, 195.

platyrrhynca, 189, 196.

Peranemaceae

Key to the genera, 192.

Peranema, Key to the species, 193.

Peranema

cuneatum, 189, 193.

curvicauda, 189, 194.

kupfferi, 189, 193.

Phacus, Key to Sections and species,

157.

Phacus

acuminatus, 161, 165.

v. javana, 161, 166.

discifera, 161, 165.

agilis, 161, 164.

angulatus, 161, 167.

anomalus, 161, 167.

caudatus, 161, 167.

circulatus, 161, 168.

curvicauda, 161, 167.

dangeardi, 163, 164.

glaber, 162, 173.

granum, 163, 164.

hamatus, 161, 169.

helikoides, 162, 172.

horridus, 162, 173.

lefevrei, 160, 162.

longicauda

var. rotunda, 162, 171.

myersi, 161, 170.

onyx, 161, 171.

oscillans, 161, 164, 165.

platalea, 161, 167.

pleuronectes, 161, 168.

polytrophos, 163, 164.

pusillus, 163, 164.

Gardens Bulletin, S.

pseudonordstedstii, 162, sei

pyrum, -162, 172.

raciborskii, 161, 173.

ramula, 162, 172.

stokesii; 160, 162.

suecicus, 162,°173. -

tortus, 162, 171.

tricarinatus, 163, 170.

trifacialis, 161, 162, 164.

triqueter, 161, 170. —

unidentatus, 164, 169.

wettsteinii, 161, 163.

ProwsE, G. A., The Eoglentnnd of

“Malaya, 136.

Strombomonas spp., Key, 184.

australica, 164, 186.

deflandrei, 164, 185.

fluviatilis, 164, 186.

gibberosa, 164, 187.

girardiana, 164, 185.

praeliaris

v. brevicollaris, 164, 185.

schaunslandii, 164, 186.

Trachelomonas, Key to Species, 174.

Trachelomonas, 184.

Trachelomanas

allia, 178, 180.

armata, 178, 181.

conica, 178, 179.

crebea, 178, 182.

curta; £77, 178:

dangeardiana

v. glabra, 178, 182.

hispida, 178, 179.

v. elongata, 178, 179.

hystrix

V. paucispina, 178, 183.

klebsii, 178, 180.

lacustris, 178, 179.

lismorensis, 177, 178.

megalacantha, 178, 181.

mirabilis

v. affinis, 178, 180.

oblonga, 177, 178.

v. attenuata, 178, 179.

rugulosa, 177, 178.

scabra

v. labiata, 17, 182. -

similis, 178, 183.

v. hyalina, 178, 183.

superba, 178, 181.

v. oblonga, 178, 181.

volvocina

v. minuta, 176, 178.

volzii

v. cylindracea, 178, 183.

zingeri, 178, 180.

Yard,

(Hk. f.

Knema furfurace

39488.

bark. A from tree in Pottin

arb. B,

Knema hookeriana

ardens, Singapore. B from Sinclair S.F.N.

B, Knema furfuracea

et Th.) Warb. Both show similar flaky

anic G

Knema hookeriana (Hk. f. et Th.) W

Bot

Plate I. A

ERLE:

icia

Knema later

Knema intermed

a &

o-5

On”

ro)

~~ &

Co:

aA 2

Mo &

= oe

SEe

sae

SEE

ees

= oO

fen

SO.y

s<s

<2.

ark:

. oo

bee

o~ o

Ess

Wot

suc

S c

= <-=

S52

Sts

S“¢€

Ses

<ocm

Plate III. Myristica maxima Warb.

View showing the whorled branches. From the tree on Lawn Z,

Botanic Gardens, Singapore.

Myristica maxima

Myristica lowiana

brittle bark

Botanic Gardens

’

g the slightly rough bark without any

Myristica lowiana King showing the rough

Sinclair S.F.N. 40319.

with longitudinal fissures. A from the tree on Lawn Z,

Myristica maxima Warb. showin

Met

cP)

ZEO

eee

Qc op

Om .S

"oS & WW

fae}

lair S.F.N.

. . + a“ tt ale

' * beh .

Beate «nee 5" «s 9

ge a eer a

a a as ail rs ete od “

pee .

meted © San ‘

dinal fissures. A from Sinc

Myristica iners (Bl.) Warb. Bark is small

40599. B from Sinclair S.F.N. 40048.

ae oS

a mf ea

ie has y

Myristica main

bod

—_

ory

in both cases, black with longitu

ac Pa en

Plate V. A, Myristica maingayi Hk. f. B,

A, Myristica elliptica

istica elliptica

, My

ing into Peirce

inclair S.F.N.

rough bark of an old tree from S$

tilt roots from a tree at a stream flow

—

cD)

Hw Say

Mv d

Tr 2

n ©

Ss Q,

Los

~ Cc oO)

Qeag

ea,

_~ 4

=" Go

caares

So &

Balt 2

~ )

> 8

= 2

Plate VI.

39661.

vk 4 :

B, Myristica fragrans

Plate VII. A, Myristica cinnamomea King from a young tree in

Gardens’ Jungle, Singapore. B, Myristica fragrans Houtt. from the tree

behind the Office, Botanic Gardens, Singapore. Both show a slightly

rough bark.

* i

ristica crassa

, My

Horsfieldia bivalvis

howing bark

lair S.F.N

shallow longitudinal fissures. A from Sinc

9490. B from the tree in Botanic Gardens

, shallow, longi-

, Singapore.

showing bark with numerous

, Horsfieldia bivalvis (Hk. f.) Merr. s

fot t)

AOS

S&o

2° nO

seg

ates eS

pes

hati cl

Pn ee

SEBS

Plate VIII. A

B, Knema communis

Plate IX. A, Horsfieldia irya (Gaertn.) Warb. B, Knema communis J. Sinclair. Both

show dents where portions of bark have fallen out but the bark of A is

hard while that of B is thinner and more flaky. A from tree in Arbo-

retum, Botanic Gardens, Singapore. B from Sinclair S.F.N. 40522.

. Horsfieldia crassifolia

superba

Horsfieldia

arb. B, Horsfieldia superba

rb. Both show longitudinally fissured bark but in B

B from Sinclair

(Hk. f. et Th.) “We

the ridges are flatter. A from Sinclair S.F.N. 40256.

Ss}

—

<tan

x ™

. 2

PR aS

3 S

Sess

~ 3 “

Ue :

= ce

x x :

ot os

Tose

Plate X. A,

all

*N.

~];

Sinc

Sif

punctatifolia J.

showing hard bark with

ark. A from Sinclair

sie es

re AS) Sr

3 a o's .

= ht

Sj <= os faa

8 = 3 : —

— an—_

3 x oDy

"D 3 we FOS

Send ~

S — apie. toe

Sa a = O84

~_ ete fn fia

‘ S .w8.3

me oe ©

< See

a So

SEBE

Ske

32 wo

v :

i) BoC

sEox

SoD

mea

Plate XI.

A, Horsfieldia-sucosa

B, Horsfieldia polyspherula

Plate XII. A, Horsfieldia sucosa (King) Watrb. showing the thin flaky bark with

some shallow longitudinal fissures. B, Horsfieldia polyspherula (Hk.

f. emend. King) J. Sinclair showing the non-flaky, hard bark with

longitudinal depressions and raised ridges. A from tree in Botanic

Gardens’ Jungle near Potting Yard. B from Sinclair S.F.N. 40681.

B, Gymnacranthera forbesii

Plate XIII. A, Gymnacranthera bancana (Mig.) J. Sinclair. B, Gymnacranthera

forbesii (King) Warb. Both show a slightly rough, non-flaky bark.

A from Sinclair S.F.N. 40045. B from Sinclair S.F.N. 37936.

. DC.) J. Sinclair showing a slightly

ilar to that of G. bancana and G. forbesii.

mnacranthera eugeniifolia (A

rough bark sim

Plate XIV. Gy

"itet

;

tro

‘oo

. =

he Ne ae

mw Nie, ot ea

yf ies, or)

Lah

ViTItn 2 RE EEE UF IFT DULANIL GA

10.

v1;

12.

SINGAPORE

Annual Reports. 1875*.

Reports for many years 1886 onward remain available.

Prices variable.

. The Agricultural Bulletin of the Malay Peninsula (Series D.

Nos. 1-9, 1891-1900.

Only Nos. 3, 5 and 7 available at 20 cents each.

. The Agricultural Bulletin of the Straits and F.M.S. (Series II).

Vols. 1-10, 1901-12, monthly issues.

All are available except Vol. 1 (6) and 1 a? at 50 cents

each or $5 per volume.

. The Gardens’ Bulletin, Straits Settlements (Series III).

Vol. 1 (1-5) Jan.-May 1912, issued under title of

Agricultural Bulletin of the Straits and F.M.S. and

Vol. 1 (6-12) Dec. 1913—March 1917 as The Gardens’

Bulletin, S.S.

Vols. 2-11, July 1918—August 1941. 3

All parts available except Vol. 2, 6 and 11 (3).

Prices variable according to size.

. The Gardens’ Bulletin, Singapore (Series IV).

Vol. 11 (4) September 1947*.

All parts available. Prices variable according to size.

Present subscription per volume payable in advance,

$20 in Malaya, or $21 outside.

. Malayan Garden Plants: a series of six booklets each

illustrating 10 useful garden plants.

10 useful garden plants.

All available. Price 50 cents each.

. An illustrated guide to the Botanic Gardens Singapore.

Price $1.

. Materials for a Flora of the Malay Peninsula.

Pt. I, If and III remain available.

. A Revised Flora of Malaya, Vol. 1 Orchids by R. E. Holttum.

Government Printing Office, Singapore, 1953 ed. I,

1957 ed. II. Price $20, second edition.

A Revised Flora of Malaya, Vol. 2, Ferns by R. E. Holttum.

Government Printing Office, Singapore 1955. Price $20.

Malayan Orchid Hybrids, by M. R. Henderson and G. H.

Addison. '

Government Printing Office, Singapore 1956. Price $21.

Wayside Trees of Malaya by E. J. H. Corner, 2 Vols.

Government Printing Office, Singapore. Price $25. —

Items 1—8 can be obtained from the Botanic Gardens,

Singapore.

Items 9-12 can be obtained from the Government m ; e

Printer, Singapore.

Prices quoted in’ Malayan Dollars.

In certain cases postage is extra.

cL? ey

rt riled s.

bia

eta

bcs Se <

ree.

aha

agit

av)

y ry :

Set erreacnen ees

Sm GORE ent mc oy HD

} ; b seks ye mi > pha ees

ifs Silhe fart : : ace iste

sets tpikued . aioh nt

eee te at so pareldtabursthiadn ri

i “alieh elena tate Lae en

iS slclavbessdeliiabtannan

uLe py t

tit bn ah Cla ||

- ~ "> bb seledys

fms 4, he 24”

spiais: pete im Gauedi Gm Pb pe

Setrtir tats ito ti tile 4

dy beget» lta aad Mies

"¢ ib pad Minty fy eape tien fe 1

*-saplns 4a orodenalnysisanesieh af MRE

Wek iene t oo ctehelsbe Noman lel aie athinneanl

vein sirens sicerery ten metal te Sil A ied pie

Ce i PL dt dy el 4

9 Bow

- “ ay) - Nee < is +9 nko

tot dede) wad os oil pe ‘ ; Son reset a gides sin bts 4, ety

4 . 4 rH ay Per ared bea al a oF hee {>t EP % , a Oh ote Od eS = Oe,

ni a bbl eletabed ebkbenel al aha Ek ral nee ; ‘ ‘ ra treet Ss ea,

yee, beter erat 4 bee ot) ogee ers 4 * bids oe ae fl eS kl . ‘ on ex . . ; . Z

PIT) Oe dae Sadalade

Mabel Gl dal til re teas 4

5 > 7 $a teted 00) ops bela ven paren it heagus

M ) MUN Po rdat 4 e gin ead dant,

i dy Fey vey ised ue:

tes : Aa,

eee we

Bs + ease 34g

petty cee py

ro tn Teh e gms

eT LP aden

“ 4 . 5 Tey baba : : i ne r

sipisinses ; Ms or " 1 ey) ags Alene + ¢ n . ' fad it ee

re! mate esanal np meemh

Heh ot eben tubes ty |

rt Thr FM beds a

LY HF pd at

Yea: ry

Lrinnne ehh

j : -

Sales > silent a ne fy ti . ; "

rea becalanecel Lidy “e / Jey vn)

rd tee ddd»

Lawes @ hakase Ett et Eee

ee OM DME Log Gy

ee bathe diadelebebetens ita

slel bien Pin

eR an ete S one Seer eieeeor’

en ee ee ee el

Lobe didi BoP. . (

BHP E TPM my ae ete oe

Ph Read Ko te rote een

- oe, ,

Ninkeneheheithedalnkebaks $ IO! Reha 5M) Do Ol a > ral mys oe i pea 0/8 he , bem te Te ‘ | = a " * | : |

mip HMM i : i

nthe Le : ebay

ij dloimialehel degbaiainicheielicnthitelalcalan ie Ter

ial iababahniehabalehsisdelciil: Aciebe keholel, ae bekeiel)

ia tal olives afiadbadal aiiintin Uiedaiin Hinde ie tale inet tie meen an)

G~hapalicons eines tet oh -& seeded otk Lniniedebehinee habietabe tan ‘ dr diet Bed vn

jn) Oh af ingagel Dafedabeledeted be leGnimbadale.el wears: ‘ roy eben : : ‘ ; Ne on Pat 2 4 J ae <r OW bee

Uinlidy ere Bf aiev@lor Ohedubede al eleanen =/)-penteamgs ’ hel ‘ ON bs l 4 : ' > : an oer - ‘ ‘ bya : 2 ot NE RP 9

1d te- ted 5 aided 94 ob eit etd Lk ed et eden ‘ : iat Subad 4 . . ei | on | | : | | =

Py telig he Phunk ee ae ee) dna 4 4 - ' - ‘ : 2 yo)

6 ol ania ahaltllebellasaih alk adele te pe 4 , ) r . sited os ‘ ‘ < ee a Art meer wie + si qpalbedn 26 f yoayehoges een

ma td Tesleselbacdetem tek ee Leen ne ene i ‘ \ \ |

~ 7

ponds j erotn -. Ww dered apo aPebgbainds>

Tdi wbotek ohabeh chal cn eRe eee FR e Si paroles Stina: ; 7 2 a 0 ; . ; . oy webadal

POEMS T LE ery Al detedltel iby 9 rat : habe 118 ON) : ; Y ya nemyeeet aye)

per Dy dibtat nd angie id Mae aera baba . +4 - 4 n a mr by eer eoe se oe ees + Ser e=p-

ON EP Fee! OPM tom lenegeveiele eben | 4 q