You are on page 1of 141

ACTA PHYTOGEOGRAPHICA SUECICA 59

EDIDI:r'
SVENSKA V AXTGEOGRAFISKA SALLSKAPET

Dick J ohansson

Ecology of vascular epiphytes in West African


rain forest

Resume:
Ecologie des epiphytes vasculaires dans la foret
dense humide d' Afrique occidentale

UPPSALA 1974
ACTA PHYTOGEOGRAPHICA SUECICA 59
Distributor: Svenska vaxtgeografiska sallskapet, Box 559, S-751 22 Uppsala, Sweden

Dick Johansson

ECOLOGY OF VASCULAR EPIPHYTES IN WEST AFRICAN RAIN FOREST

Ecologie des epiphytes vas�ulaires dans la foret dense humide d'Afrique


occidentale (French summary)

Doctoral dissertation to be publicly discussed in the Botanical auditorium, Upp­


sala University, on May 2, 1974, at 10 a.m., for the degree of Doctor of
Philosophy (according to Royal proclamation No. 327, 1969)

Abstract The ecology of 153 species of vascular epiphytes (101 orchids, 39


pteridophytes and 13 others) in the Nimba Mts in Liberia is described. 29
species are recorded in Liberia for the first time including one new species,
Rhipidoglossum paucifolium, Orchidaceae. Field characteristics, flowering
periods and some pollinators of the orchids are also given.
In high forest with the canopy 30 m or more above the ground, 50.4 %
of the trees (phorophytes) 10 m or higher carried epiphytes, compared to
14.8 % for the phorophytes in regenerating forest. The ratio between fern
and orchid species was 1 :3 at 500-700 m alt. and 1:1 at 1000-1300 m.
Most of the epiphytic species occupy a ± restricted part of the phorophyte,
as judged by their occurrence in the five sectors in which the phorophytes
were subdivided.
Ten different epiphyte communities are recognized. Their floral composi­
tion and development are also described. Certain species of phorophytes,
e.g. Heritiera uti/is, Lophira alata and Parinari excelsa generally carry
an. abundance of epiphytes, while others, e.g. Anthocleista nobilis, Fagara.
tessmannii and Terminalia invorensis are mostly devoid of epiphytes.
Colonization by epiphytes begins late in the life of the phorophyte, which
is indicated by the higher frequency of epiphytes on the larger specimens.
The effect of the epiphytes on their phorophytes is reviewed and obser­
vations from Nimba and East Africa provided. The environmental in­
fluences on the occurrence of the epiphytes on the phorophytes are dis­
cussed. In addition, field observations and experiments concerning water
economy microclimate, substrate properties and light intensities are
presented.

ISBN 91"-7210-059-1 (ISBN 91-7210-459-7) 136 pp.


ACTA PHYTOGEOGRAPHICA SUECICA 59
EDIDIT
SVENSKA VA XTGEOGRAFISKA SA LLSKAPET

Dick J ohansson

Ecology of vascular epiphytes in West African


rain forest

Resume:
Ecologie des epiphytes vasculaires dans la foret
dense humide d' Afrique occidentale

UPPSALA 1 9 7 4
Suggested citation : Johansson, D., 1 9 74, Ecology of vascular eopiphytes in West African rain forest.
Acta Phytogeogr. Suec. 5 9 . Uppsala.

Doctoral dissertation at Uppsala University, Sweden 1 9 74.

ISBN 9 1 -7 2 1 0-059- 1 (paperback)


ISBN 9 1 - 7 2 1 0-45 9-7 (cloth)

© Dick Johansson 1 9 74

Svenska vaxtgeografiska sallskapet


Box 5 5 9 , S- 7 5 1 22 Uppsala, Sweden

Technical editor : Gunnel Sj ors

Printed in Sweden 1 9 74
Textgruppen i Uppsala AB
Abstract Johansson, D. 1974, Ecology of vascular epiphytes in West African rain
forest. (Resume: Ecologie des epiphytes vasculaires dans la foret dense
humide d'Afrique occidentale.) Acta Phytogeogr. Suec. 59. Uppsala. 1 3 6
pp.

The ecology of. 1 5 3 species of vascular epiphytes (10 1 orchids, 3 9


pteridophytes and 1 3 others) i n the Nimba M t s i n Liberia is described. 2 9
species a r e recorded in Liberia for the first time including one new species,
Rhipidoglossum paucifolium, Orchidaceae. Field characteristics, flowering
periods and some pollinators of the orchids are also given .
In high forest with the canopy 30 m or more above the ground, 50.4 %
of the trees (phorophytes) I 0 m or higher carried epiphytes, compared to
14.8% for the phorophytes in regenerating forest. The ratio between fern
and orchid species was I : 3 at 5 00-700 m alt. and 1: 1 at 1 000- 1 3 00 m .
Most o f the epiphytic species occupy a ± restricted part o f the phorophyte,
as judged by their occurrence in the five sectors in w hich the phorophytes
were subdivided.
Ten different epiphyte communities are recognized. Their floral composi­
tion and development are also described. Certain species of phorophytes,
e.g.Heritiera utilis, Lophira alata and Parinari excelsa general l y carry
an abundance of epiphytes, while others, e.g.
A nthocleista nobilis, Fagara
tessmannii and Terminalia invorensis are mostly devoid of epiphytes.
Colonization by epiphytes begins l ate in the life of the phorophyte, which
is indicated by the higher frequency of epiph y tes on the larger specimens.
The effect of the epiphytes on their phorophytes is reviewed and obser­
vations from Nimba and East Africa provided. The environmental in­
fluences on the occurrence of the epiphytes on the phorophytes are dis­
cussed. I n addition, field observations and experiments concerning water
economy microclimate, substrate properties an d light i nten sities are
presented .
Content

Preface 'herbs' 64, O n dead trees 64, O n rocks and


soil 65
I. Introduction Di strib ution o f epiphytes on the p horophytes 66
General notes on vascular epiphytes 2 Ecological subdivision o f the phorophyte 66,
Distribution 2, Occurrence in various Recording and registration of epiphytes 69, Eco-
families 2, Phylogeny 2, Properties 2, logical observations 7 2, General distribution
Ecological subdivisions 4 pattern 7 5
Previou s studies o f vascular epiphyte ecology 5 Epiph y te commu n ities 78
Scope of the present study 5 Terminology 78, Methods 79, Abundance 80, Epi-
F i eld wo rk 6 phyte communities at altitudes below 1 000 m 80,
Ter m inology 6 Epiphyte communities at altitudes above 1000 m
Th e area of investigation 6 85, General observations 8 6
C l im ate 8
Epiphyte succession 86
Veg etation 13
Primary and secondary forest I 3, The Liberian Epiphyte succession in Himalayan Quercus forest

rain forest 1 4, Classification of the forest 8 7, Succession indicated by floristic composi-

in the Nimba area 1 5, The phorophytes (the tion 87, Succession traced by analysis of humus

host trees) 1 7 deposits 87, Effects of environmental changes 89

E ffects o f the epiphy tes on the phorophy tes 91


If. The epiphytic flora
Earl ier ob servations 91
Composi tion and g eograp hical di strib ution 21
Indirect effects 9 I , Direct effects 9 2
Li st of species 23
Author's ob servatio ns 93
Pteridophytes 2 3 , Orchids 2 6, Other vascular
The leafless orchids 9 3
epiphytes 41
V . Environmental influence
Ill. Biology of vascular epiphytes
Climate 95
R eproduc tion biology 43
Macroclimate 9 5, Microclimate I 00, Light inten-
Flowering periods of orchids 43, Observa­
sities 10 I, Daily variations in light intensity
tions of orchid pollination 44, Diaspore
104
size and dispersal 4 5, Germination and
establishment 4 5 Substrate 1 06
Life forms 46 Properties o f the substrate and their influ en c e
Drought tolerance and drought avoidant o n the epiphytes 1 06
epiphytes 47 The bark 106, Bark pH and the epiphytic flora

48 107, Exudates 1 0 8, Humus deposits 109, Corre-


W ater economy
lation between substrate and epiphytic flora
Water uptake 4 8, Storage of water 51,
1 10
Expenditure of water 5 3
Biotic fac tors 1 15
IV. The occurrences o f epiphytes Human influence 115, Animal influence II 5 , Nut-
Epiphy tes on forest trees 55 ritionaJ importance of animals and microorga-
1. Total number of epiphytic species known 5 5, nisms 1 1 5
2. Quotient epiphytes/other vascular plants 5 5,
VI. Distribution on the phorophytes
3. Percentage of tall trees carrying epiphytes 5 5,
4. Highest number of epiphytic species on one General pattern 116, Specific patterns 1 1 6

tree 59, 5 . Number of epiphytic specimens or


Resume 1 21
'stands' on one tree 6 1

Epiphy tes on oth er substrates 62 Literature 1 24


On cultivated trees 62, On tree-ferns 6 3, On

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 1

Preface

My first contact with a tropical rain forest dates M any other Lamco employees have in one way or
back to August 1 9 64 when I com m enced employ­ another helped or simplified m y field work.
ment for Lamco J.V . Op. Co. in Liberia. As a high Help and guidance in the determination o f critical
school teacher at the I nternational school in Yekepa specimens have been received from Jean Bowden,
I used my spare time during the following six years F.N . Hepper, Prances Jarrett, M. Sands and P.
to explore the rain forest. In the vicinity of the newly Taylor of the Kew Herbarium , and from L. Garay of
constructed Yekepa community , at 5 00 m altitude , the Oakes Ames Herbariu m , H arvard Univ. , C am­
virgin rain forest covered the slopes of the Nimba bridge, M ass. Prof. Sanford, Univ. of Ile Ife, Nigeria,
range up to the peaks at 1 300 m altitude. The rain provided an unpublished m anuscript on the use of
forest was very poor in species in the field layer . On epiphytic orchids in general ecology work .
the trees , however, vascular epiphytes were abun­ The determinations of the insects were performed
dant and soon caught m y particular interest. While by M . A . Grog an, S . L . Shute and I . H . H . Yarrow of
trying to learn about the ecology of the epiphytes it the Entomological Dept. of the British Museum
soon became evident that the literature on this sub­ (Natural History).
ject was very scanty. Although the vascular epi­ Aina A im , Gertie Backlund , Margareta Hellman
phytes are a very characteris tic floristic element in and Barbro Enander h ave typed the m anuscript.
the rain forest inaccessible growing sites have M iirta Ekdahl has prepared the large tables and
resulted in few and o ften superficial reports on their Ulla-Britt Sahlstrom h as copied the photographs.
ecology. The map over the investigated area was drawn by
A fter an i ntroductory period in which to become Per Ake S oderman .
acquainted with the tropical flora a prelim inary plan In the preparation of the m anuscript valuable
for the future investigation was outlined in coopera­ criticism has been provided by Doe. Orj an Nilsson.
tion with Prof. H ugo S jors at the Institute of Dr Jerk Hellkvist at the Institute of Physiological
Ecological Botany in Uppsala. Botany in Uppsala has commented on the section
A fter returning to Sweden in 1 9 70 m y work was about the water economy of the epiphytes. The
continued at the newly started Afro-Botanical Dept. linguistic revision has been carried out by Andre
of the Institute of Systematic Botany in Uppsala un­ Brochu and the French resum e was translated by
der the guidance of Prof. Olov H edberg. Renee Perrette.
During an excursion to Kenya and Tanzania in Granges International Mining Division,
January 1 9 7 1 , under the leadersh ip of Prof. Stockholm , has supplied a grant for the photo­
Hedberg, I was able to study epiphytes in several en ­ graphic illustrations.
vironments different from those of the lowland Prof. Hedberg, my teacher and tutor, has through
tropical rain forest. As an indirect result of a three his never failing enthusiasm and encouragement ac­
months ecological study the same year for Sweco in tively contributed to the completion of this work.
the Kidatu area in Tanzania, a knowledge of Finally , Astrid did a considerable amount of work
epiphytes in dry environments was gained . behind the scenes.
Several persons have contributed to the comple­ A ll person s and institutions mentioned above I
tion of this study. H arvey Boettcher, M anuel Diaz want to sincerely thank for valuable cooperation.
and Franz Zimsek o ften accompanied my tours of Uppsala, January 1 9 74
the forest. K urt Eriksson helped in the construction Dick Johansson
of the light-measuring equipment and Miss Anne Institute of Systematic Botany
C ron e carried out some laboratory work in Liberia. Box 5 4 1 , S-75 1 2 1 Uppsala, Sweden

Acta phytogeogr. suec. 59


I. Introduction

GENERAL NOTES ON VASCULAR salis, with numerous epiphyte species. E ricaceae


E PI PH YTES ( 13 50) has a co nsiderable number of epiphytes es­
pecially i n the Vaccinioideae. The families of
Distribution Araceae (2000), A sclepiadaceae (2000),
Epiphyti s m , as defined by Bark m an ( 1 9 5 8 : 9), occurs Gesneriaceae (2000), Melastomataceae (3000), a nd
in m any plant o rders. I t has a very extensive geo ­ R ubi aceae (6000) are also rich i n epiphytes.
graphical distributi o n among non -vascular plants A geographical pattern can be observed in the
s uch as algae, mosses, a nd lichens. composi tion of the epiphyte flora. I n Africa p te rido­
The vasc ular epiphytes, p teridophytes a nd phytes and orchids almost solely form the epiphy te
ph ane rogams , are o n the other h and more or less flora, while i n South America bromeliads and cac­
restric ted to tropical regions. In tropical rain forests taceous plants also are important. I n Asia a nd
especi ally, they form a significant part of the flora, Australia species of Asclepiadaceae a nd R ubiaceae
and their presence a nd abundance are o ften used as are wel l represented together with the p teridophytes
a characteristic of such forests (Schimper 1 903 :260, and o rchids .
Richards 19 64 :8) . Vascular epiphytes are rare i n
temperate clim ates. I n h umid climates , e.g. certain Phylogeny
coastal and high mountain areas, a large number of Schimper ( 1 8 88) mai ntained that the tropical epi­
plant species may occur for varying periods of time phy tes have developed from te rres trial types that
as epiphytes although lacking special adaptations to were g rowing i n dark h umid forests . W e nt (1 895)
such a life (Sharp 1 9 5 7). The somewhat confused claimed th at the epiphytes have evolved from roo t
termi nology of the different types of epiphytes will be climbers. Pittendrigh ( 1 948) suggested that the epi ­
presented later in this chapter. phytic b romeliads o riginated from terrestrial
ancesto rs that lived u nder sem i -desert c o nditions .
Occurrence in various families The same would probably apply to the epiphytic
In the tropical rain forest epiphyti s m amo ng vascular C ac taceae and m aybe the majori ty of the epiphytic
plants occurs in a limited num ber of families , but in a floweri ng plants.
large number of genera. Schimper ( 1 8 88) mentioned Garay (1 9 7 2 :2 1 1) date the developm e nt of the
33 fami lies and 2 3 2 genera. N a tu rally the number of epiphytic mode of life amo ng the o rchids back to the
k nown epiphytic genera has i nc reased since then, but Pl iopleistocene. He also rightly s tates that while the
the number o f families g iven by him is reasonably up terms geophy tic and epiphytic convey the meani ng
to date (R ich ards 1 964 : I I 2). of ecological habi tats , i n reali ty they express disti nct
It i s very di fficul t to estimate the total number of evol u tionary adaptations through morphological
epiphytic species in the world. However, their modifications i n the roo ts.
presence i n certai n families is better k nown. The es­
timated total number of species in the fam ilies men­ Properties

tioned below, in parentheses, is take n from Willis What are the p roperties that make a plant successful
(1 966). The majority of the orchids (20 000) are as an epiph yte? Two condi tio ns must be ful filled. (1)
epiphy tes (cf. H u nt 1 9 6 7, S ummerhayes 1 96 8). The The plant or i ts di aspores m us t i n one way or
b romeliads ( 1 400) play an important part in the another be able to reach a posi tion on the phoro­
epiph yte flo ra of the New World and to a lesser ex­ ph yte. (2) It m ust possess the ability to s urvive
tent this also applies to the C ac taceae (2000). The drough t after its establish ment.
latter family also i ncludes a pantropical genus, Rhip- ( 1 ) The weigh t of the diaspores h as been emphasized

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 3

by Schimper ( 1 888) as a chief limiting factor for


potential epiphytes. He poi n ted out that m any of the
numerous epiphytic bromeliads in South and Central
America have particular adaptations in the seeds for
wind dispersal .
The largest number o f epiphy tes occu rs among the
o rchids and the pteridophytes which both h ave very
minute diaspores. These are so sm all that they can
easily be carried by the wind. 84 % of 50 true
epiphytes in New Zealand exhibit m inute spores o r
seeds (Oiiver 1 9 30: 1 9). T h e weight of a n o rchid seed
(Dendtobium attenuatum) was given as 6 . 5 11g by
Ames ( 1 9 22).
The role of ani m als in the dispersals of the
diaspores of epiph y tes h as been s tressed by several
authors . Holtum ( 19 60) reports dispersal by insects :
"Ants may h ave a n importan t fun c tion in several
ways. They may carry seeds, especially small seeds
which con tain food useful to them. Some of these
seeds, including seeds of o rchids, are sought chiefly
fo r oi l drops in the outer tissues and the emb ryos
may not be dest royed by ants ; the seeds grow in
ants' nests , which provide shel ter and minerals for
the seedlings . " Numerous bird s , bats , monkeys and
squi rrels, living in the trees m ay also contribute to
the dispersal of the epiphytes . F ruits m ay be eaten
and the seeds spread with the exc reta. A somewhat
s tranger way of dispersal i s reported for the
epiphytic bromeliad , Tillandsia usneoides, 'spanish Fig. 1. Ti llandsia usneoides, 'spanish moss'
(Bromeliaceae). New Orleans, La.
moss', that occurs in the South Eastern parts o f USA
(Fig. 1). This lichen-like plant i s used by certain birds D rought is the most serious threat to the life of the
as building materi al for thei r nes ts . The plant su r­ epiph y tes. D rought resistance shows a wide range of
vives the b reeding period and continues to develop variation, from the poorly adapted filmy ferns to the
afterwards. The most common way of dissemination, extremely resistan t epi phytic cactaceae and b rome­
however, occurs during violent gales, when pieces liads. It is reported that bromeliads survive in areas
are b roken off and carried by the wind to new phoro­ with about 400 m m /year in N . A rgentina (Tixier
phytes in the same manner as among certain lichens 196 6 : 2 2).
(Pennfound & Deiler 1 94 7) . The greates t abundance o f epiph y tes is found
( 2 ) Once located on the phorophy te the young plant where there is a yearly rainfall of 1 00 inches c� 2500
must have an effective system of attachment to the mm) or more and where no month has an average o f
substrate so as not to be washed down by the rain­ less than 2 or 3 inches (Holtum 1 9 60). I n Indochina
fall. The root systems are often very extensive, super­ the epiphytes disappear when the rainfall i s less than
ficial or ae rial, and in many cases specially 1 000 mm per year (Tixier 1 9 66: 1 2 1) . The
developed only for ancho ring purpose (certain temperatu re seems also to limit the geographical dis­
bromeliads). tribution of ce rtain epiphytes, as can be observed e.g.
A common problem is the short time of water up­ in the East A frican high mountains where the pte ri­
take due to the limi ted ability of the substrate to dophy tes occur at much higher alti tudes than the
s tore water. The plants are often adapted to a rapid orchids.
up take and high s to rage capac i ty o f water, replacing The epiphytes show an interesting variety of bo th
the s to rage in the substrate. morphological and physiological adap tations for

A cta phytogeogr. suec. 59


4 Dick Johansson

wi thstanding unfavourabl e conditions. It would, I V . Eph em eral epiphytes. Th es e are accidental


how ev er , take too far to discuss thes e adap tations epiphy tes . They are for the most part seedlings of
h er e. A brief d es cr iption of the major life-forms is trees whi ch fail to gain m aturity in their epiphyti c
given in chap ter I I. posi tion, and consequently die after attaining a size
which dem ands more n utrients than th e s ubstrate
Ecological subdivisions can furnish.
S ev eral authors have group ed the epiphytes into Th e term ' eph em eral orch ids' is used by Smith
large units , based upon l ife-forms, l ife-cy cles or their ( 1 925) for plants that carry their flower only one
d ependence on l ight, s ubstrate and h um idity . The day , e.g. Dendrobium crumenatum. In F . W .T.A. 1 : 1
div ergent ter m inology for s uch gro ups m akes com­ ( 1 9 54 :220) the term 'sub-epiphyte' is used for a
parisons between d ifferent investigations difficult. Begonia that is 'creeping over logs , etc'.

Barkman ( 1 9 5 8 :9- 1 5) has given a v al uabl e s um m ary Th e typical epiphytes, h av e also been divided into
of the terminology con cerning non-vascular epi­ a large num ber of units bas ed on the hab it or
phy ti c cryptogams . d ep enden ce of water, light, a nd s ubstrate.
R egarding non-vascular cryp togams th e term Growth habit
obligate is o ften used for typical epiphytes s en s u S chimper ( 1 888) r ecognized three types : pro to, nest
Oliv er (Table 1 ) . Th e r est of the epiphytes are called or bracket, and tank epiphytes , bas ed on the habit of
facultativ e (Barkman 1 9 5 8 : 1 1) . A s ubdivis ion of the the plant.
facultativ e epiphy tes into three categories, preferen­
tial, indifferent, a nd o ccasional h as been used by Water
Hilitz er ( 1 9 2 5 ) . Both v as cular and non-vascular Th e d ependen ce on water is a com mon cri terion for
plants are incl uded in the m i cro and macro-epiphytes a s ubdivision. Oliver (I 930) distinguishes between
of Tixier ( 1 96 6 :45) and in the Sonn en(lich t), hydrophytes, m esophytes and x erophytes. S chnell
S chatten, and Epiphyllen v eg etation of K napp ( 1 9 5 2 :2 8 2) recogni zes the sam e units as L ebrun
( 1 973). ( 193 7) with one addition, 'les epiphytes hygro -m eso­
Oliv er ( 1 9 3 0) pres ented a s ch em e that will b e used phi l es'. Th e units used by S ch n el l are th us :
h ere as a base for comparing th e terminology of I. Epiphytes hygrophiles d e la base de troncs
different authors (Table 1). I I . Epiphy tes hygrom esophiles d e la region moyenne
I. Typi cal epiphytes. Species which are h abitually d es tron cs
epiphyti c. Ill. Epiphytes m esophiles des grosses bran ch es et
du sommet d es tron cs d es grand arbres
II. O ccasional epiphytes. H er e are i ncluded those
I V. Epiphytes x erophiles d es bran ch es superieures
plants wh ich are under ordinary circumstan ces
terres trial . O ccasionally they are found on trees
Light
wher e they appear to thrive until maturity.
A division into shade-, s un -, and extreme x ero­
I ll. H emi -ep iphytes . These are trees which begin ph ilous epiphytes , has been used by Davis &
th ei r existence on other trees or tree ferns, send down Ri chards ( 1 9 3 3 :3 80-3 8 1 ), and R i ch ards ( 1 9 3 9 :3 1 -
one or more taproots to th e ground and eventually 3 2). R i chards admi ts the diffi culties in disting uishing
es tablish th em selves as indep endent plants. th e latter two gro ups.

Table 1. Classification of epiphytes by some authors.

Oliver 1930 Typical ep. Occasional ep. Hemi-ep. Ephemeral ep.

Schimper 1888 Holo ep. Hemi-ep.


Went 1895 Wahre ep. Hemi-ep.
Pessin 1925 Eu-ep. Pseudo ep. Hemi-ep. Pseudo ep.
Lebrun 1937 Ep. authentique Ep. accidentel Hemi-ep. Ep. accidentel
Hosokawa 1968 Ep. typica Ep. occasionalis Hemi-ep. Ep. ephemeralis
Schnelll970 Ep. hygrophiles Ep. accidentels Lian-ep.
Ep. mesophiles
Ep. xerophiles

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 5

Grubb et al . (1 9 6 3 : 59 2) is also critical of this d iv i­ ford ( 1 9 6 7 , 1 9 68 b, 1 9 6 9 , 1 9 73) has p resented a


s ion. In h is s tudies of moun tain rain forests in series of in teres ting papers concerning the d is tribu­
Eq uador he remarks : "ex treme xerophilous ep i­ tion and ecology of epiphy tic o rchid s in N igeria and
phytes, sensu R ichards, are not markedly more dis­ elsewhere. Several authors e .g . Piers ( 1 96 8), Morris
tinct from e ach o ther in their b iology than crown­ ( 1 9 6 8 , 1 9 7 0), and S tewart (in S tewart & Campbell
base and bole-base skiophytes ." Grubb e t al., use the 1 9 7 0) have reveale d interes ting facts about the
terms photophyte and s k iophy te for ep iphy tes that ecology of orchids in East and Central Africa. In a
l ive in wel l-l it and ill-lit hab itats respec tively. penetratin g study of the epiphytic fern genus Platy­
Guillau me t (1 9 6 7 :6 0) d is tinguishes between a cerium in , the Ivory Coast, Boyer ( 1 9 6 4) assembled
h elioph ile, a mesophile and a sciap h ile group. an impressive amoun t of fac ts , u se ful for the un­
Hosok awa ( 1 9 6 8) groups the ep iphytes into four derstanding of the ecology, not only of the Platy­
ecological types : (a) the sun-type growing in d irec t cerium ferns bu t also of epiphy tes in general .
sunshine, (b) the s tem and/or leaf succulen t xero­ Regarding the New World surprisingly few
ph ilous type, (c) the shade-tolerant type growing in publications are available (Dun n 1 94 1 ). The
diffuse weak l ight, (d) the hygrophilous type. bromeliads have naturally received interest from
many d irec tions. Picado ( 1 9 1 3) publ ished a report
Substrate on the ecology of the bromel iads and their remark­
Wen t ( 1 9 40) considers the demand on the substrate able adap tations in the epiphy tic l ife-forms . Their
as a dividing line between epiphytes. He d ifferen­ anatomy and importance as b reeding s ites for
tiates between two groups : bark epiphy tes and those various insects have been studied by Pennfound &
th at need an accumulation of humus to be able to Dei ler ( 1 9 47 ) and Pittendrigh ( 1 948).
develop, so cal led humus epiphytes.
It is of course impossible to rank the importance S CO PE O F THE PRESENT STUDY
of all the env iron men tal factors without experimen tal It is no ex aggeration to say that the composition and
proof. Which one of the many factors involved that geographical d is tribution of the vascular epiphy tic
is decisive can be very hard to determine s ince they flora bo th in Africa and elsewhere in the tropics are
are o ften in teracting. poorly known . Even less k nown are the presence and
d istribu tion of the various species in the great
PR EVIOUS STUDIES O F V AS CULAR nu mber of habitats that exis t on the trees . This s tudy
EPI PH YTE ECOLOGY has aimed at recording the growing s ites of epi­
Research on epiphy tic plan ts was started by ph y tes in the rain forest and particularly the oc­
Sch imper. In a pioneering and comprehens ive work currence of epiphytes in the h ighest parts of the
'Die Ep iphy tische Vegetation A merikas' (1 8 8 8) he fores t, since such information is extremely scarce.
focuses his atten tion on the anato my, mo rphology This i s partly a quan titative study of the spatial
and ecology of epiphytes. dis tribution, since whenever possible the actu al
Even though the epiphytes for m a ch arac teristic number of epiphytes was counted accord ing to a
part of the rain fores t they are usually men tioned spec ial system. However, the eco logical re­
only casually in ecological reports (see sec tion F ield quiremen ts for the individual species were also in­
work). vestigated .
In Asia specialized stud ies have been carr ied out Special atten tion has thus been paid to the l ight
on 1 ava by Van Oye (I 92 I , 1 92 4 a), and Went and subs trate and their poss ible effect on the
( 1 9 3 1 , 1 9 40), and in M icronesia and elsewhere by presence and abundance of epiph y tes. The o ften
Hosokawa (I 9 52 a, b, I 9 54 a, b, c, 1 9 55, 1 9 5 7, observed tendency of epiphytes to form some k ind of
1 9 68). Hosokawa has also written numerous articles co mmun ity induced a detailed invest�gation of the
on the termino logy of epiphyte commun ities and the abundance, floral composi tion and developmen t of
l ife-forms of epiphy tes. F ro m V ietnam Tixier (1 9 6 6) such co mmunities .
has presented a thorough study of epiphytes includ­ Previous reports of a correlation between the epi­
ing bryophy tes. phyte flora and species of 'h os t tree' offered ano ther
In Africa, Van Oye ( 1 9 2 4 b) and Lebrun (1 9 3 7) aspec t th at was investigated. The many adaptations
have studied epiphytes in the Belgian Congo. San- that occur among the epiphy tes induced s tudies into

Acta phytogeogr. suec. 59


6 Dick Johansson

their water econo my . The use of epiphy tes as i n­ vations from roughly 1 500 trees, pal ms, fer ns and
dicators of microclimate and forest destruc tion was lianas.
also i nves tigated (cf.. Sanford 1 9 7 2). O u tside the specially i nvestigated area I have
A more specialized s tudy regarding the phen­ carried out surveys of the epiphy te flora in every
ology, polli nators a nd special field charac teris tics of cou nty of Liberia. The neighbouring countries of
the orchids was also i ncluded. Gui nea and Ivory Coast have also been visited on
My s tudy i s res tricted to the vascular epiphy tes. several occasio ns e nabl i ng a comparison of the flora.
However, the so-called filmy fer ns, e.g. Trichomanes Epiphytes in e nvironments, different from the rain
and Hymenophyllum, are i n the mai n excluded, but forest types, have been s tudied i n K e nya and
the species observed are listed i n chapter I I (Table 7). Nor thern Tanzania (Jan. 1 9 7 1 ), a nd in Sou thern
In several way s these fer ns resemble the non­ Tanzania (April -June 1 9 7 1 ). Brief v isits i n Zambia
vascular epiphy tes. Their leaves are usually o ne cell (July 1 9 7 1 ) a nd i n Cameroon (M t Cameroo n, July
l ayer thick and without sto mata and can take up 1 9 7 1 ) have supplied addi tional i nfor mation.
water i n the same manner as mosses. The facultative E fforts have been made to get photographs from
(occasional, hemi- and ephemeral) epiphy tes are also specime ns growi ng i n situ, bu t difficul ties in ob­
excl uded but listed (Table 7) and referred to when of tai ni ng good pictures of some of the phorophytes i n
i mportance for the epiphy te flora. the dense forest and o f the epiph y tes i n the crowns,
explai ns why some of them had to be photographed
FIELD W OR K in rather unnatural e nvironments . All pictures, ex­
My field work w a s carried out during 1 9 64- 1 9 7 0, cep t Fig. 3 1 , are taken by the author.
wi th the major part of the distribution recordings
fro m the las t two years of thi s period. TER M INOLOGY
The scanty k nowledge of the presence and the dis­ Epiphyte is here used as defined by Barkman
tribution on the trees of epiph y tes i n the tropics can (1 9 58 : 9). When no thi ng else is stated the term
be traced back to two different causes. The first is epiphyte refers to vascular epiphy tes. The 'host tree'
the very i naccessible growing sites of these plants. (shrub, pal m etc.) of the epiphyte is referred to as the
The second is the difficul ties connected with the phorophyte followi ng Ochsner 1 9 28 ; (cf. Barkman
determi nation of the epiphy tes. Several species are 1 9 58, S jogren 1 9 6 1 , Tixier 1 9 66, Skye 1 9 6 8). This
extremely si milar i n size and shape i n their flowerless term is accordi ng to the defini tio n by Ochsner
s tates (Fig. 3 3). ( 1 9 28). The no me nclature of the phorophytes agrees
To overcome the first obs tacle my research has wi th Voorhoeve (I 96 5) and for species no t treated
been directed i n three different ways: record ing of by him the nome nclature in Hutchinson & Dalziel :
the epiphyte flora, ( 1 ) on felled trees (close obser­ Flora of West Tropical Africa (F .W.T.A .) 2 nd ed., is
vation), (2) of s tandi ng trees (dis tance observation), followed .
(3) on a mi nor part of a tree by close or distance The term high forest is applied to fores ts with a
methods (occasional observati o n). The two first canopy at a height of 3 0 m or more regardless of the
methods give a rather accurate picture of the dis­ origi n. Secondary or rege nerati ng fores t refers to
tribution of epiphy tes on the trees while the third is forests in various s tages of regrowth wi th the canopy
utilized for addi tional i nfor mati o n. below 30 m height.
To get acquai nted with the epiphy tes i n various
stages of their developme nt I s tarted fro m the very
beg inning to collect spec i me ns and grow them in
The area of investigation
s mall wooden boxes in a specially cons tructed 'frame
house'. The phenology of the various species could Th e area i nvestigated, called the N i mba area, is
then easily be studied and minor ch arac teristics si tuated in northern Liberia, near the border be­
observed. The i nves tigation i ncludes the exami nation tween Gui nea and the Ivory Coas t (Fig. 2) a nd i n­
of 22 0 felled trees and 236 trees exami ned accordi ng cludes parts of the Nimba M ts . The cou ntries rele­
to the distance method on 4 7 species of phorophy tes vant to this i nvestigation are shown in Fig. 3 .
i n the high forest. The so-called occasional recor­ Large mountai ns are poorly represe nted i n Wes t
d ings number more than three thousand obser- Africa bu t within the so-called Guinea Highland,

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 7

GUINEA

BONA •

lJMT YULLITON

(/.EETON N

I
I
I

�MT TOKAOEH
I
I
I Mf SOUTH NIMBA
I

I
I
I
I
I 'ZlJLO'Wit
I
I
I
I
I
I

I
I
I
I
/

3km
.______.___....___�


L.::..:.J 800- 1000m j:::::::::::]1000 -1300 m H��h��H 1300 m
>

Fig. 2. Map of the investigated area and its location in Africa.

stretching from Fou ta-Dj allon in Gu inea to the Dans the Ivory Coast the Dans mass if contains the two
massif in the Ivory Coast, a series of r idges and high peaks of Mount Tonko u i, 1 240 m, a nd M t Dou
peaks r ise to altitudes of more tha n 1000 m. I n 1340 m . The N imba Mountains s tretch fro m Guinea
northern S ierra Leone, the Loma Mountains reach into L iberia, where the r idge has a N E-SW d irection
1 94 7 m at the Bintu mane peak . I n Guinea, Mount w ith the h ighest part s l ightl y above 13 00 m (Fig. 4).
Bal a in the Ziama massif is 13 8 7 m h igh, but eve n Here the range is ris ing abruptl y fro m a tableland at
more imposing are the N imba Mountains w ith a around 5 00 m altitude. I ts h ighest part is b u il t up by
number of h igh peaks (e.g. M t M olard 17 5 2 m). In a narrow ridge or crest w ith s teep slopes on each s ide

'
2
/Po r �'7=""�":---+--'-�=--.,.,..-­
Guin
Nigeria

Fig. 3. Some West African


countries of relevance to
I
6N -----

this investigation. The dis­


tribution of rain forest at
low and medium altitudes
is marked by dots. 1 W
2

A cta phytogeogr. suec. 59


8 Dick Johansson

(Fig. 1 1 ). The southernmost part ends in a steep fro m a h igh-p ressure belt above the subcontinent of
mountain, Bele (Bilimu). East of the ridge the two South Africa towards a low-pressure b elt formed
p eaks Mt Gang ra and Mt Yulliton of more than above North Africa. Th is w ind is in W est Africa
1 000 m dominate the landscap e. South of th ese known as the southwest monsoon.
p eaks rises another ridge with th e h igh est part at Mt When th e moist SW -monsoon reach es the
Tokadeh at around 1 000 m altitude. After the dis ­ Lib erian coast h eavy rains start to fall. Th e rainfall
cov ery of iron o re in the Liberian N imba in 1 9 5 5 and in th e coastal strip is v ery h igh, generally mo re than
th e start of mining operations in 1 96 3 , a large influx 4000 mm a y ear. Towards the interior p recipitation
of p eople has tak en place into this p reviously sparse­ dec reases, and in the northern part the p recipitation
ly populated area. The early stages in the exploration is l ess than half of the amount of the coast.
and th e new dev elopments in this area are d escribed In th e southeast of th e country, a short dry p eriod,
in Gardlund (1 9 6 7). 'th e mi ddl e d ry ', often app ears in Jul y -A ugust, which
in c ertain y ears is even felt in N imba (Fig. 5).
I n periods of v arying length the macroclimate at
CLI M A TE
Y ek epa (5 1 0 m) has been measured, i n a traditional
For L ib eria as a whole the y ear can roughly b e way, w ith th e instruments in a meteorological box,
d ivided into a rainy s eason from March/April t o Oc­ 1 .5 m abov e the ground. The rainfall has b een
tober/November, a nd a dry s eason during the re­ measured w ith a standard rain gauge also at 1 .5 m
maining months. These extreme w eather types are above th e ground. The maximum and minimum
c reated by the y early movements of the I nter temperatu res were tak en at s ev en o 'clock for each 2 4
Tropical Front (I .T.F.). This front is formed b e­ hours p eriod. Th e r elativ e h umidity w a s recorded
tween the relativ ely cool, moist air of the equatorial with a hygrograph, wh ich was calibrated to a psy­
zone over the Atlantic Ocea n, the Equatorial chro meter (dry and w et bulb th ermometer) . At the
Maritime A ir M ass, a nd the hot d ry mass of air over p eak of th e ridge, th e mining company , Lamco
th e S ahara, th e Tropical Continental Air Mass. Dur­ J.V . Op . Co., has s ince 1 95 7 p erformed regular
ing th e d ry s eason the movements of thes e air masses meteorological observations. Unfortunately the
are, s imply speaking, regulated by the h igh-p ressure observ ation site has b een mov ed twice, and the data
b elt formed over the Sah ara and th e low-pressure after January 19 6 7 only includ e rainfall and
b elt ov er the area of South Africa. The w ind blow­ cloudiness.
ing fro m th e h igh-pressure b elt to th e low is d eflected Th e mic roclimatical effects o n the d istributio n of
to the west owing to the earth's rotatio n. This wind th e ep iphytic plants w ithin the trees are more con­
wil l th us appear as a north eastern d ry often hot and spicuous than macrocl imatic effects . (Howev er the
dust-laden w ind, co mmonly k nown as the H armat­ d ifference between th e epiphytic flora at the c rest of
tan. During the rainy s eason the wind is blowing the ridge and that in the surrounding lowland is no

Fig. 4 The Nimba Range seen


. .

from Mt Molard (1752 m) in


Guinea in a S W direction. Note
the absence of clouds on the west
side of the Range and on Mt
Gbahm. March 1965.

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 9

Yekepa 510 m
500 ..., ... "',
,--,... \
/ \
30 \
21-22.6 1969
,------
25-26.1 1970

r;' \,
-
25
-
-

250

.-- 20

r--

r-- 15
,----
06 12 18 24 06

n 10 12 14 13 18 20 21 15
I Fig. 6. Temperature variations during a day in the wet and
a day in the dry season. Yekepa 5 1 0 m.

temperature aro u nd 22°C. The daily v ariation of the


temperature in the rainy and dry seasons is show n i n
Nimba Range 1248 m
Fig. 6.
A shower of rai n w ill of course alter this daily
.----f--- pattern. The temperature m ay the n drop 4-5°C in a
-- r-- few minu tes. I n the dry season the difference be­
tween maximu m and minimum temperatures is much
wider than in the rainy season. This can, somewhat
250 ,---- simplified, be correlated to the low frequency or to tal
absence of clouds during that period. This resul ts in
,----
an i ncreased influx of radiation during the day, but
,----
also a larger ou tflow at nigh t.
,.----- With the arrival of the H ar mattan winds fro m the
nor th a few days o fte n show very low nigh t

� 10 11 15 17 21 20 23 10
temperatures. The lowest temperature recorded is
from February 1, 1 9 7 0 when a readi ng of 12. 5°C
was regis tered during the night. The following day
J A N. DEC.
the m aximum temperature reached 3 2 . 5°C which
Fig. 5. Monthly rainfall at Yekepa 5 1 0 m and at Nimba gives a d ifference of 20. 0°C (Fig . 8). The daily v aria­
Range 1 248 m, 1 969. The number of days with rainfall is tion of the te mperature at the higher parts of the
given below each month. ridge is u nk nown but there is a five year (1963-
1967) series of maxim u m a nd minimu m temperature
doubt regulated by the macroclimate.) Unfortu nate­ recordings from the Geologis t's Camp Nimba. The
ly, m ic roclimatical investigations in the treetops and figures from 1966 are given in Table 2. The
along branches, 30-40 meters above the ground, differences between the mea n maxi mu m and
have been impossible to perform. The microcl imate m inim u m temperatures are s mall. The largest
in a rain forest, discussed later in chapter V, has been differences occur i n the dry season.
investigated in Nigeria by Evans ( 193 9) and in the In a h igh forest, at gro u nd level , the daily variation
Ivory Coast by Cachan (I 963), and Cachan & in temperature, as one would expec t, is s maller than
Duval ( 1963). outside the fores t (Table 3). K u nkel (1966), who h as
carried out occasional temperature recordings fro m
Temperature various pl aces in Liberia, esti mates the temperature
In Yekepa the daily maximum temperature during in an open area to be around 5°C higher tha n inside
the rainy seaso n is close to 3 0°C and the minimum the forest.

Acta phytogeogr. suec. 59


10 Dick Johansson

Table 2. Maximum and minimum temperatures (mean ) at Nim­


mm
60
ba Range 1340 m, 1966. ( Source: Lamco J.V.Op. Co. )

Temperature oc
Period Max. Min. Mean
January 22.8 19.2 2 1.0
February 22.4 19.0 20.7
March 21.7 18.7 20.2 30

April 21.0 18.3 19.6


May 20.4 17.9 19. 1
June 20.2 18.3 19.2
July 1 8.8 16.7 17.7
August 19.5 17.7 18.6

I I
September 19. 1 17.4 18.2
October 20.6 18.4 19. 5 . I I I
15.6 30. 6
November 22.7 20.3 2 1. 5 1.6

December 20. 9 1 8.3 19.6


Fig. 7. Daily rainfall (in mm) during June 1 969. Yekepa
5 1 0 m.
Table 3. Temperature one meter above the ground, in a high
forest and in the open, during a day in the wet and a day in the
dry season. Seka V alley 600 m .
tion probably i nfluence the distribu ti o n pattern of
epiphytes on the phorophytes .
Temperature ° C In the beginni ng of the rai ny season, M ay -July,
Date Season M ax. Min. Diff. the rain often falls as a drizzle. In June, particularly,
i t may rai n more or less continuously for half a day ,
High forest 23.6 1969 wet 25 .5 23. 5 2.0
22. 1 1 9 70 dry 26.0 22.0 4.0
which is rare any other time of the year. This is the
Open area 26. 6 1969 wet 29.0 22.0 7.0
ti me when the surface fro nt of the I.T. F. has passed.
22. 1 1970 dry 3 1.0 17. 5 13.5 A t the e nd of the rainy season, Sept. -_N ov. the mor­
nings are often clear but i n the afternoons heavy
thunder-showers are usual. These rainfalls are of a
Rainfall convective nature, e.g. the humid air which is
In an area with differences in alti tude of 500-600 m warmed by the sun rises and co ndensation occurs. I n
with i n a few kilome ters distance, local e ffec ts i n the the afternoon h uge clouds have buil t u p and large
distribu tion of the rainfall will of course be amounts of water are released . These rainshowers
pronounced. From the highest part of the ridge, at are of a short duration.
two different sites, records of the rainfall are avail­ The number of days per month wi th rainfall varies
able. The firs t series of observations are fro m the fro m year to year. The figures for 19 69, at 1 248 m
Geologis t's Camp N i mba, 7 °3 1 ' N , 8°3 1 ' W , at an and 5 1 0 m are shown in Fig. 5. The period Augu s t­
al ti tude of 1 340 m, and _i nclude the years 1 9 57 -1 966. October has the larges t number of rai ny days.
The second observation poi nt was at the M i ne Of­ I n Yekepa, no rai n was recorded during the period
fice, 7°3 8' N, 8°29' W , at an al titude of 1 248 m, fro m December 1 2, 1 9 6 9 - February 8, 1 9 7 0 (58 days of
the years 1 96 8 - 1 9 7 0. The average rainfall for the drought), and at the M i ne Office, 1 248 m, no rai n
four years 1 9 63 - 1 9 6 6 at Geologis t's Camp N i mba was recorded duri ng the period November 1 6 , 1 969 -
was 3 1 06 mm compared to 2487 mm for the three­ Febru ary 1 6, 1 9 70 (92 days of drough t).
year-series at M i ne Office. Dur i ng 19 62, the rai nfall
in Yekepa was o nly 54 . 0 % of th at at Geol. Camp. Humidity
(Adam 1 9 7 1 : 1 5). At nigh tti me, the relative h u midity is 1 00 % or there­
The distribution of the rai nfall i n one month abouts duri ng the whole year. The absolu te content
reveals some i nteres ti ng details. Even i n a month of water in the air varies naturally with the seasons.
duri ng the rai ny season, June 1 9 69, wi th a total of Even i n the dry season the air i s on the verge of
3 1 6 mm, 1 4 days h ad no rai nfall, a nd at the middle saturation duri ng the nigh ts, a fact wh ich the low
of the month it only rained 1 0 mm during a 1 2 -day­ night temperatures can account for. From the
period (Fig. 7). These 'dry periods' of shorter dura- saturated level at nigh tti me the humidi ty level is

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 1 1

i 6 8 10 1214 16 18Z0 22t� · 6 8 '" '"/ t¥ ' 6 8l12 " 16 18 Z022?Y.· 6 8 10 "" 16 18 Z0 2�· 6 8 10 1214 16 18 Z0 2!� · i 8 10/l 16 1820 22?� ·; 8l ""l 1820 22?� .
mi.J-J-1.1- LU�IIJ_H.)
Hontag Monday Lundi Dit:nsfoq rue:sdoy Mordi Hillwoch Wednesday Hercr�i Donner&tag Thunday Jeudi I Freitag Friday Vendredi I Sonnobend Saturday Som�i I Sonnlag Suttd_
oy Dimonche IMonfaa

. UII 1 I I U_/
1,[ I I
,
u
.
- r e ke p a 5 1 0 m
7 0 I I 'Jj I I
L// /// ril l 1/1/ 1/ l rlllll rt /Tlr/1 ! rUt/ lrt l I TT /ll rt / 11
J an . 2 6- Feb. 1 1 9 7 0: I V" I I •� .,.. I J" A-T 1" I
. �H
1 -H�-1+��
r -+ r V V
/ I
I 1 7
��-��.++ � -l-
t--� ·

20.0 19-s 1 8 .0 1 6.s 1 4. 5 12.5


.· Honlog HOliday Lundi Dit:rutaq lut:sdo Hard(
1 14 16 1820 21�,· 6 8 10l14 16 18Z022f'�i· 6 8 1 /" 16 181 ?�.· n61 10 12 "l "t22?�,·;
6i7""/6 "2flt?� · 6 810/"; ; �¥,· 6 8 10 12 14 16 1820 22?;� ' 6 8 ,0/ / Hillwoch Wednesday Hercredi / Donnerdog Thursday Jeudi / Freilag Friday Vendredi / Sonn !Hnd
a Solurdoy Sam�i f So nlog Sunday Dimonche fHonloo

V e ke P a '5 to 'rind� · '


l .f- {Il
.
l!
'
I I I I I ',[ I I .L
.! I 11 11/ l rlltrr TTTTTr! l / 11 1 1 / l l l ri i i i i i / /JI -('v/ / 1 1
i. J U n,e, l, 1 6 - 22 1 9 6 9l l I I I 1//1
I
I I I I I I
I !
I I I I I 1 1 I
;:
I�
. ./ I -- --�

20 . 5 2 1 .o 19.5 19.5 2 1 .o 2 2 .o 21. 0

Fig. 8. Relative humidity during a week in the wet and a imum and minimum temperature (°C) are given for each
week in the dry season. Yekepa 5 1 0 m. Figures for max� 24 hr period.

lowered during the day as the temperature rises. The small (F ig. 9). The weekly average of the absolute
absol ute minim um rel ative h um idity (Table 4), al­ m in im um relative h um id ity is as high as 84 %.
though only lasting for a short per iod in the after­
noon, can be used to ill ustrate the yearly variation in
Table 4. Absolute minimum relative h u m idity. Mean for the
the desiccation effect on the plants.
periods. Yekepa 5 1 0 m.
Naturally large variations are found in the daily
changes in the relative h um idity. This applies to Absolute minimum
variat ions between individ ual days of the same Period relative h u m idity (o/o)

season as well as between whole periods in different


seasons. The daily variation of the relative hum idity 1 6. 6 - 2 3 . 6 1 9 69 58.8
during a week in the rainy season (July 22-27, 1 9 69) 2 2 . 7 -2 7 . 7 1 9 69 68.0

and the dry season (February 2-8, I 9 7 0) illustrates 1 1 .8- 1 7.8 1 969 73. 1
2 2 . 9- 2 8 . 9 1 969 55.7
thi s (Fig. 8). The impact of the dry H armattan winds
4 . 1 1 -9. i I 1 96 9 48.0
is clearly visible in the graph fro m the dry season.
8 . 1 2- 1 4 . 1 2 1 9 69 46.4
During the period 3 1 . I -5.2 I 970 the mean absolute 1 2. 1 - 1 8. 1 1 9 70 4 7.4
minimum relative h um idity was only 2 3 . 2 %. 2.2-8.2 1 9 70 35.7

At higher levels, e.g. at Mt Gbahm 1 3 00 m , where 2.3-8.3 1 9 70 3 7. 3


6.4- 1 4.4 1 9 70 46.6
mist and rainfall particul arly from June to A ug ust
2 7.4-3.5 1 9 70 48.6
keep the air saturated, the daily variation is very

A c t a phytogeogr. suec. 59
12 Dick Johansson

Monlog Hondoy l.undl Oi�nslaq Tuesday Mardi Hilfwoch Wedn�sday M�rm:di Donn�rslog Thrmdoy J�udl FreUog FrldtJy Vendr�l SonnabMd Saturday Sam�i Sonnlag Sunday Oimanthe Honlag

1i . '{jjj,h rlt:��, �·�jfr� . . . .7�i


11/'-�. _.LI I�. f-1>rt �� bf/.J th� I rv I J
. . . . . . , .. .. ..,H . . . . .,.; ··"W , . . . "j¥i ·;"""�{
. . .

I I rl 1 1 1 1 1 1 1 1 1 L TLI I II I I/ 1 U n l l l l l / 1 11! rl ! ! 1 1 1 1 1 1 1 1 '/ I l l


· ·

!"- 'f-'f'- +' V.. -In

Fig. 9. Relative h u midity during a week in the wet season. Mt Gbahm 1 300 m.

Cloudiness Wind
The cloudiness has been recorded at the Geologist's In general wind velocities are low. Thunderstorm s
Camp Nimba and Mine Office at 1 248 m . The day are accompanied or rather p receeded by strong
was div ided into two recording periods, 06°0- 1 2°0 winds that usual ly last for only a few min utes.
and 1 2°0- 1 8°0 (Fig. 1 0). The ratio between cloudy Kunkel (1 9 6 6 : 5 9) reports that, at the end of
and sunny days in 1 9 69 reveals that June, July and February 1 962, a very strong thunderstorm passed
August were the cloudiest months, although the John - Davis Town (Putu district) in eastern Liberia.
period September to October h ad the same mean The wind had a hu rricane-like strength and created
rainfall as the three earlier months. January 1 9 70 an alm ost 1 00 m gap in the forest east of the settle­
showed only five mornings with more than 50 % ment. Only around 20 % of the trees were still stan­
c loud cover and all the afternoons were cloudless. ding after the storm h ad passed. In the N imba area
During the period January 1 2 to Feb ruary 6, 1 9 70 no such storms of s igns of earl ier ones have been
no clouds whatsoever were reported at any period of observed.
the day. The local cloud o r mist system that fre­ For certain short periods the dry Harmattan
quently can be observed around the highest parts of winds are strongly felt in the N i mba area. The wind
the ridge du ring the d ry season seems to be cor­ is weak , merely as a breeze, but is characterized by
related with the N E -winds (Fig. 1 1 ). its dust-filled, very d ry ai r. The v is ibility is low (a few

I v o ry C o a st L i b e ri a
DAY S Altitude
30
r-r- r-r- r-r--
r-r--1-r-
-_ -r-_r-

-
-r- 1---


-r-
I JOO

?

\�
= Air sinking

\
wn


15 -

-
A i r fo rced jA
Clouds diss �
u p w a rds /
�-·

.• 0<g tx'x
�Rx
_.

0 � �R:><�
A M PM
600

J F M A M J J A s 0 N D

Fig. 1 0. Annual distribution of cloudiness (hatched) and


sun at Nimba Range 1 248 m, 1 9 69. The days are divided Fig. 1 1. Cloud formation over the crest of the Nimba
into two parts, AM and PM, and cloudiness means that Range under easterly winds during the dry season.
more than 50 % of the sky was covered with clouds. (Schematic cross-section of the Range 1.1 km SW of the
(Source : Lamco J . V. Op. C o.). Three Border Point, in a N W-SW direction.)

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 13

k ilom eters only). The sun d isappears as a r ed ­


glowing ball in th e a fternoon, a n d th e relative
humidity r each es its absolute y early m inimum.

VEGETATION
Earlier botanical exploration
Th e N imba Mts in what today is part of Guin ea and
th e Ivory Coast wer e during the French adm in is tra­
tion subj ect to a s er ies of studies cover ing s ev eral
aspects of natural h istory (e.g. L amotte et al. 1 9 5 5).
A r emarkabl e work on the vegetation of the N imba
Mountains was pres ented by Sch n ell ( 1 9 5 2).
From th e Liber ian part of the N imba Mountains
botan ical collections have b een p erform ed by W .
Harl ey a t and around M t Bilumi (M t Bele). Und er
th e sponsorsh ip of th e N imba R es earch C o m m ittee,
h eaded by K . C urry-Lindahl, bo tanical collections
w er e carried out by P. Adams ( 1 964) and J. Adam
( 1 964- 1 9 65 and 1 9 70). Based upon h is ex ten s iv e
coll ections from all parts of t h e ridge, Adam ( 1 9 7 1 )
publish ed a work entitled 'Flare d escriptiv e d es
Mon ts N im ba'. During h is study of Lib erian h igh
forest trees, Voorhoev e ( 1 965) v isited th e N imba
M ts and mad e collec tions there.

Primary and secondary forest


Gen erally speak ing the rain for ests can be d iv id ed
into two main groups, prim ary and s econdary. Th e Fig. 1 2. The canopy of the high forest. Mt Gbahm 800 m.
prim ary for es t does not b ear any s ign of human in­ 1 969.
flu ence (Fig. 1 2), whil e the s econdary has b een
affected in one way or the oth er b y human ac tiviti es for est. It is in many cases ex tremely d ifficult to dis­
(Figs. 1 3 and 1 4 ). Ther e are diverging opinions as to tinguish a mature s econdary forest, e.g. a s econdary
th e pres ence of true prim ary for est in W est Africa forest of som e hundred y ears age, from a prim ary
and Liberia. The t erm ' h igh for est' is o ften used one.
(Voorhoeve 1 9 65 : 1 9) to d escrib e a for est without It has b een estimat ed that it takes 80 y ears for a
stating an opin ion of its origin. I n French l iterature moist forest species of tree to reach maturity (Cole
the term 'foret d ense' (Aubrev ille 1 93 8) is o ften used. 1 9 6 8) and that th e cycl e towards a m atur e h igh
It includes the evergreen rain fores t as well as the for est lasts 300 to 400 y ears (Voorhoev e 1 9 65 : 20).
mix ed d ec iduous forest. 'Clos ed for est' (Richards It is v ery hard to character iz e th e plant com ­
1 9 64) is equ ival en t to th e 'foret d ense' of modern m u n ities in tropical for es ts b ecause they are very
F rench writers, and includes all types of lowland rich in spec ies wh en compar ed to temperate forest
for est in wh ich th e trees form a clos ed canopy and stands of eq u ivalent s iz es (cf. R ichards et al. 1 940).
the ground i :; not cover ed w ith grass. Th e rain fores t Som e ecologis ts name a closed for est using th e
is o ften divid ed into m ixed and s ingle dom inant com­ em ergents in th e A -s tratu m (th e canopy), but th es e
mun ities. Th e greater part of the prim ary forest in emergents are gen erally few and far b etween and
the Nimba area consists of mixed for es t composed of dom inance by a few spec ies is the exception rather
a v ery large number of spec ies . than the rul e. Ev en in the for es t as a whole the ma­
The s econdary forest (or d egraded fores t) is jority of tree spec ies are rare or occasional. Aubrev il­
characteriz ed by its physiognomy and th e abun­ l e ( 1 9 3 8) does not cons id er any tru e associations in
dance of trees from the lower strata of the prim ary th e mix ed rain forest of W est Africa. The whole m ix-

Acta phytogeogr. suec. 59


1 4 Dick Johansson

Fig. 1 3 . 'Farmland' in high forest.


The forest is cut and burned. Rice,
and cassawa (manioc) are
cultivated in the clearing for a few
years after which the farmland is
abandoned. Mt Yullition 700 m.
1 966.

Fig. 1 4. 'Farmland' in various


stages of regrowth near a native
village. West of Bona, 1 966.

ed forest in this case could be regarded as a single would be favourable for it throughout most of the
association of fluctuating composition. Richards country.
( 1 964 : 262) states : "The failure of any one species to The Liberian rain forest is p art of a rain forest belt
gain the upper hand in the m ixture may be due, as th at stretches from Ghana to Sierra Leone, and
Au breville suggests, to all the species having very which is a western extension of the huge lowland rain
similar ecological requirements and responding in a forest of the Congo basin (Fig. 1 5). A region with
very similar way to slight variations in the environ­ dry climate from Western Nigeria to Eastern Ghana
ment. As would be expected on grounds of probabili­ separates the two forested areas from each other.
ty, under these conditions the composition of these This break is commonly known as the 'Togo­
mixed communities will fluctuate in space and prob­ Dahom ey gap'. The high forests of Liberia exhibit
ably in time as well." several features typical of tropical rain forests such
as a large number of species which are mostly
The Liberian rain forest woody plants, e.g. many tree species (Cooper &
Through human interference only 35 % of Liberia is Record 1 9 3 1 , Kunkel 1 96 5 , Voorhoeve 1 9 65). The
at present covered by high forest (Voorhoeve undergrowth is also mainly composed of woody
1 9 65 : 1 9), although climatic and edaphic conditions plants. The herbaceous plants are mainly represented

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 1 5

Concerning the development of the Liberian forest


interesting speculations have been presented : "It is
possible that as recently as 300 years ago there was
considerably less high forest area in Liberia than is
found today. The subsequent sharp decline in pop­
ulation, as heavy tribal warfare and slaving activities
exerted their drain, possibly coupled with the ravages
of new diseases, would have permitted many cleared
areas to complete their reversion to high forest. Over
a period of several centuries, stands such as now oc­
cur throughout much of Liberia would probably
develop. Their composition would differ notably
from original stands in the same general area, but
their density would be very similar. Only by such an
hypothesis can the writer explain the over-all com­
positional inferiority of Liberian high forests as com­
pared with those considered typical of the Ivory
Coast and the Gold Coast." (Mayer 1 95 1 : 25.)
A similar view is reflected in a report of the Ger­
man Forestry Mission to Liberia, which express the
opinion that all the forests of Liberia were once
destroyed by shifting cultivation since remnants of
Fig. 1 5. Distribution of rain forests at low and medium villages have been found even in the most distant
altitudes. (From Keay 1 959.) places of the forest.
The human influence on the fauna may indirectly
by epiphytes. A striking abundance of climbers, both affect species' composition of the rain forest. The
herbaceous and woody (lianas) is also observed. elephants that nowadays are almost extinct are said
Voorhoeve ( 1 965) divides the Liberian rain forest to promote the reproduction and distribution of
into three categories : primary forest, old secondary Sacoglottis and Parinari trees. They eat the trees'
forest that has reached climax, and old secondary fruits, and after passing through the alimentary canal
forest which has not yet reached the climax. Concer­ the seed germinates extraordinarily well and at the
ning the distribution he remarks ( 1 9 6 5 : 1 9) : "The same time is transported over a considerable dis­
first group appears to be extremely rare in Liberia, it tance.
is restricted to remote and limited localities of which The high forest in the Nimba area that previously
only the gorges in the Nimba Mountains are known was sparsely populated is nowadays subject to a
to me. Steep slopes prevented all agriculture there rapid deterioration (Adam 1 966, 1 970 ; Coe &
and the belief in numerous spirits made the people Curry-Lindahl 1 9 65). However, precautions recom­
afraid to penetrate the area." mended by I.U.C.N. are being taken by the Liberian
In neighbouring Sierra Leone human activity government to spare certain areas from the
started to take its toll in the forest very early. It is in­ devastating effects of shifting cultivation (Curry­
teresting to note that Adam Afzelius, in his Sierra Lindahl 1 969).
Leone Journal (ed. by A.P. Kup, Uppsala 1 96 7), Classification of the forest in the Nimba area
already observed this. In his notes from May 28 Schnell ( 1 95 2 :444) divided the high forests into four
1 7 95 he states : "Fleming complained that there was group s :
now not to be found any timber on the Company ( 1 ) C limax mesophile, caracterise par Triplochiton
ground, complaint which I have long expected, and scleroxylon et Chrysophyllum perpulchrum.
from the manner of treating the wood, I am sure they (2) Climax ombrophile a Lophira procera
won't 10 years hence have any timber in Sierra (3) Climax ombrophile a Tarrietia u ti/is (Heritiera
Leone, but in the highest and inaccessible moun- uti/is)
tains." (4) Climax montagnard a Parinari excelsa

A cta phytogeogr. suec. 59


1 6 Dick Johansson

Table 5. The number of trees, 40 cm or larger in diameter, in a Table 6. The number of 15 different timber trees, 45 cm or
forest at 525 m altitude east of Grassfield. Total area in­ larger in diameter in East Nimba National Forest. Total area in­
vestigated, 4 ha. (From Adam 1969.) vestigated, 565.2 acres. (Source: Lamco J . V. Op. Co.)

Piptadeniastrum 33 Ongokea Genera or species Trade names Number


Parinari excelsa 17 Terminalia
Piptadeniastrum D ahoma 703
Lophira 12 Erythroxylum
Chlorophora lroko 289
Tarrietia (Heritiera) 10 Newtonia duparquetiana
Lophira alata Azobe 28 1
Calpocalyx 7 Funtumia
Fagara B ahe 1 78
Chlorophora 4 Nauclea
Tarrietia (Heritiera) Whismore 169
Parkia 4 Lovoa
Terminalia ivorensis Frimare 1 17
Amp him as 2 Anthonotha fragrans
Parkia Parkia 1 17
Uapaca guineensis 2 Newtonia aubrevillei
Erythrophleum Tali 101
Anopyxis
Nauclea Bilinga 48
Klainedoxa Total 103
Entandrophragma Mahogany 30
Canarium White mahogany 29
Terminalia superba Limb a 21
Khaya Mahogany 15
Adam (1 969 : 3 84) recognizes two alliances in the
Guarea Mahogany 6
rain forest in the lowlands surrounding the Nimba Lovoa Mahogany 5
Mountain s. On drained soils on peaks and slopes,
Piptadeniastrum + Parinari excelsa + Parkia. I n the very common Parinari excelsa and several other
valleys and levelled areas a n alliance with Lophira + species, it gives valuable information of the propor­
Tarrietia + Uapaca + Erythroxylum + Newtonia. tions between the enumerated species.
Farming activities in the high forest favour the
Forest at altitudes below 800 m development of m any small and m iddle sized trees
The species' composition of a forest east of the typical of the secondary forest, e.g. Musanga
Grassfield Airstrip at 5 2 5 m has been investigated by cecropioides (Moraceae), (C oombe & Hadfield
Adam (1 9 6 9 : 3 80-3 85). The canopy of this forest 1 96 2), A nthocleista nobilis (Loganiaceae),
.

lies at approximately 40-45 m height above the Harungana madagascariensis (Guttiferae). Among
ground (Table 5). the taller species easily observed in older secondary
R em arkable is the dominance by the family forests, Ceiba pentandra, Pycnanthus angolensis and
Leguminosae. It is represented by seven genera, eight Elaeis guineensis m ay be mentioned.
species and 49 individuals in the inventory. The
species of this family constitute nearly 48 % of the Forest at altitudes above 1 000 m
total number of trees. Four fam ilies, Leguminosae, In the higher parts of the Nimba R ange a distinct
Rosaceae, Ochnaceae and Sterculiaceae together change is noticed in the physiognomy and floral
represent 85 % of the trees in the inventory. Adam composition of the forest. These parts, which are
( 1 969) also emphasizes the l arge differences in the very small and m ainly consist of the narrow ridges,
flora composition betw�en the Kitoma and the Nim­ are covered by a low single dominant Parinari ex­
ba forests. The K itom a forest i s located 30 k m south celsa forest, reaching a height of 1 0-20 m . The
of the Nim ba Mts. In Kitoma 1 7 7 individuals of smaller trees have their ramifications close to the
Calpocalyx aubrevillei were found in an area of 1 5 . 2 ground, an d the field layer is usually built up by
ha, which gives 1 1 . 6 trees pe r h a. In N im b a 7 in ­ herbs. At 1 200- 1 300 m altitude the Aframomum
dividual s were found which gives 1 . 7 trees pe r ha. herbs are typical. The single dominant Parinari
The figures for the Piptadeniastrum africanum are forest is by no means something specific for the
0.9 trees per ha in Kitoma, and 8 . 2 trees per ha in Nimba Mts but can be found at similar altitudes at
Nim ba. e.g. Mount Wutivi in western Liberia ( Kunkel
An inventory of the East Nimba N ational Forest 1 9 62 b), Mt Tonkoui in the Ivory Coast and Fouta
and adj acent parts h as been undertaken under the Dj allon in Guinea. From Uganda a rain forest com­
leadership of T.A. Gorgla (1 9 6 9). This study in­ munity occurring between 4500 and 5000 feet i n.
cluded 15 species of trees that could be of commer­ which Parinari excelsa forms about 80 % of the
cial i nterest (Table 6). Although the study excludes canopy is reported by Eggeling ( 1 94 7).
A cta phytogeogr. suec. 59
Ecology of vascular epiphytes in West African rain forest 1 7

The canopy is not closed except in very minor sec­ species. When a doubt in the determination of
tions. The light intensity in the lower parts of the species has arisen, the epiphyte flora h as simply been
forest and at the ground is therefore considerably recorded without correlation to a specific
higher than in the forest at the base of the mountains. phorophyte. The list of phorophytes that briefly will
Schnell ( 1 9 5 2: 338) remarks : "On retrouve dans les be presented is dominated by medium to large sized
forets a Parinari excelsa du Nimba et des massifs trees. For each phorophyte a few properties that
voisins, un certain nombre de caracteres deja might be of importance for the epiphyte flora have
signales dans les forets montagnardes, par J. Lebrun, been added. The m ajority of these phorophytes are
H. Scaetta (1 9 7 3) , Schluter, Eidmann : faible hauteur included in the excellent · study of the Liberian rain
des arbres, tronc ramifie a quelques metres du sol, forest by Voorhoeve ( 1 96 5 ) .
reduction du nombre des especes, reduction du nom­ The properties of the phorophytes refer to tall
bre de s lianes, abondance des epiphytes (surtout specimens and it must be emphasized that m any of
C ryptogames), feuillage generalement persistant." these properties are rather vague and dependent on
On the crests at 1 200- 1 300 m altitude the environmental influence. The roughness of the bark
Parinari excelsa forest is in total dominance, but a refers to the basal parts of large branches, and the
little lower down at 1 000- 1 1 00 m trees of a rather terms low, medium and tall refer to the classes
large number of species can be found (Adam 1 0-20, 2 1 - 35 , 36 m or more, respectively. The
1 9 70 :204-207). phorophytes are presented in alphabetical order
Seasonal changes in the forest without regard to family.
In the Nimba forest, not much seasonal changes in
leaf-fall are noticed. The evergreen trees at the Phorophytes in high forest
various strata appear luxuriant at all times of the A lbizia glaberrima (Schum. & Thonn.) Benth.
(Leguminosae)
year because progressive leaf-fall occurs, while the
Large. Bark smooth. Crown open, spreading. Deciduous.
few deciduous trees appear dormant only for a
(DC.) MacBride (Leguminosae)
A lbizia zygia
rather short time. These deciduous trees are so few Medium. Bark rather smooth. Crown open, spreading.
and scattered that their presence does not alter the Deciduous.
face of the forest. Bernhard-Reversat et al. A lstonia booneiDe Wild. (Apocynaceae)
( 1 972 :2 1 8) reveal that the maximum leaf-fall of Large. Bark rather smooth, soft. C rown fairly open,
various species in the Ivory Coast do no coincide, heavily branched. Evergreen.
and there is no correlation with the precipitation, but A mphimas pterocarpoides Harms (Leguminosae)
rather with the incident radiation-a view that earlier Medium-large. Bark scaly. Crown dense. Deciduous. (Fig.
had been stressed by Hopkins ( 1 970) in a study of 1 6).
the seasonal growth in S Nigeria. Furthermore, the A nthocleista nobilis G.Don (Loganiaceae)
interspecific variation of the start and the length of Small. B ark with thorns. Crown open. Evergreen.
the leaf-fall period at the same locality also tend to A nthonotha fragrans (Bak.f.) Exell & Hillcoat
obscure the presence of these trees. Some of the (Leguminosae)
more common deciduous trees in the forest are .{Jom­ Medium-large. Bark scaly. Crown dense. Deciduous.
bax buonopozense, Chlorophora regia, Ceiba pen­ A ntiaris toxicaria(Rumph. ex Pers.) Lesch. var.
tandra, Daniellia ogea, Entandrophragma utile, Pip­ welwitschii (Engl.) Corner (Moraceae)
Large. B ark smooth, thick. Crown fairly open. Briefly
tadeniastrum africanum, R hodoghaphalon brevicus­ deciduous.
pe, Terminalia ivorensis, and Triplochiton scleroxy­
Berlinia confusaHoyle (Leguminosae)
lon. In a phenological study of Chlorophora regia at Medium-large. Bark smooth or scaly. Crown rather dense.
Njala, Sierra Leone, Cole (1 968 : 76) found not only Evergreen.
an interspecific variation in leaf shedding time Bombax buonopozense Beauv. (Bombacaceae)
between trees growing within a radius of 1 00 yards, Medium. R ather smooth bark. C rown heavily branched.
but also a difference in the length of the leafless Deciduous.
period between male and female trees. Hutch. & Dalziel (Leguminosae)
Bussea occidentalis
Medium . Bark smooth. Crown dense. Evergreen.
The phorophytes (the host trees) Calpocalyx aubrevillei Pellegr. (Leguminosae)
As stated earlier the rain forest is very rich in tree Medium. Bark smooth. Crown small. Evergreen.

A cta phytogeogr. suec. 59


18 Dick Johansson

A. �. . . -

Fig. 1 6. Trees common in the Nimba area. (A) Heritiera pterocarpoides, (D) Lophira alata, (E) Ceiba pentandra,
uti/is, (B) Piptadeniastrum africanum, (C) Amphimas (F) Entandrophragma utile.

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 1 9

Canarium schweinfurthii Engl. (Burseraceae)


Large, often emergent. Crown rather open. Bark smooth.
Briefly deciduous.
Ceiba pentandra (L.) Gaertn. (Bombacaceae)
Large to emergent. Bark smooth. Crown rather open.
Deciduous. (Fig. 1 6).
Chidlowia sanguinea Hoyle (Leguminosae)
Small-medium. Bark smooth. Crown small, rather dense.
Evergreen.
Chlorophora regiaA. C hev. (Moraceae)
Large. Bark thick and rough. Crown dense. Briefly
-
deciduous.
Chrysophyllum perpulchrum Mildbr. ex Hutch. & Dalziel
(Sapotaceae)
Medium. Bark smooth. Crown dense. Evergreen.
Combretodendron macrocarpum (P. Beauv.) Keay
(Lecythidaceae)
Large. Bark thick, deeply grooved. Crown fairly dense.
Briefly deciduous.
Coula edulis Baill. (Olacaceae)
Medium. B ark rough and scaly. Crown dense. Evergreen.
Cryptosepalum tetraphyllum (Hook. f.) Benth.
(Leguminosae)
Medium. Bark smooth, superficially cracked. Crown
open. Briefly deciduous.
(Harms) Rolfe ex. Holl. (Leguminosae)
Dan iellia ogea
Large to emergent. B ark smooth. Crown small, fairly
open. Deciduous.
Distemonanthus benthamianus Bail. (Leguminosae)
Medium-large. Bark smooth, defoliating. Crown fairly Fig. 1 7. Some trees that seldom carry epiphytes. (A) Ter­
open. Deciduous. minalia ivorensis, (B) Terminalia superba, (C) Fagara
tessmannii.
Entandrophragma utile (Dawe & Sprauge) Sprauge Mammea africana Sabine (Guttiferae)
(Meliaceae) Large. Bark rather rough, scaly. Crown small, dense.
Large-emergent. Bark grooved, thick. Crown dense. Brief­ Evergreen.
ly deciduous. (Fig. 1 6).
Aubrev. & Pellegr. (Rubiaceae)
Mitragyna ciliata
Erythrophleum ivorenseA. Chev. (Leguminosae) Medium-large. Bark thin, scaly. Crown rather dense.
Large. Bark rough. Crown very dense. Evergreen. Evergreen.
Fagara tessmannii Engl. (Rutaceae) Nauclea diderrichii (De Wild) Merrill (Rubiaceae)
Medium . Bark rough, with thorns. Crown open, translu­ Large. Bark rather smooth, or somewhat scaly. Crown
cent. Evergreen. (Fig. 1 7). smal l . Evergreen.
Guarea cedrata (A. Chev.) Pellgr. (Meliaceae) Parinari aubrevillei Pellegr. (Rosaceae)
Medium-large. Bark smooth, scaly. Crown fairly open. Medium. Bark smooth or somewhat scaly. Crown dense.
Evergreen. Evergreen.
(Sprauge) Sprauge (Sterculiaceae)
Heritiera uti/is Parinari excelsa Sabine (Rosaceae)
Medium . Bark thin, peeling. Crown dense. Evergreen. Medium-large. Bark scaly. Crown fairly open. Evergreen.
(Fig. 1 6). (Fig. 1 8).
Khaya anthotheca (Welw.) C. DC. (Meliaceae) Parinari glabra Oliv. (Rosaceae)
Large-emergent. Bark smooth, sometimes peeling. Crown Large. Bark smooth, scaly. Crown rather open.
rather dense. Deciduous. Evergreen.

Lophira alata Banks ex Gaertn. f. (Ochnaceae) Parkia bicolor A. Chev. (Leguminosae)


Large-emergent. Bark scaly or grooved. Crown fairly Large. Bark thinly scaly. Crown fairly open. Deciduous.
open. Briefly deciduous. (Fig. 1 6). Pentaclethra macrophyllaBenth. (Leguminosae)
Lovoa trichilioides Harms (Meliaceae) Medium-large. Bark irregularly scaly. Crown dense.
Large. Bark scaly. Crown fairly open. Evergreen. Evergreen.

A cta phytogeogr. suec. 59


20 Dick Johansson

Triplochiton scleroxylon K. Schum. (Sterculiaceae)


Large-emergent. B ark rather smooth, more seldom scaly.
Crown dense. Deciduous. (Fig. 7 7).
Uapaca guineensis MUll. Arg. (Euphorbiaceae)
Medium. Bark rough, cracked, scaly or flaky. Crown
dense. Evergreen. (Fig. 7 6).
Vitex micrantha Giirke (Verbenaceae)
Small-medium . Bark grooved. Crown rather dense. Ever­
green.

Phorophytes around villages and in farmlands


This group includes cultivated trees of various origins
as well as trees of the young secondary forest and
grasslands. The very common oil palm is also listed
here because of its importance for epiphytic ferns.
Coffea sp. (Rubiaceae)
The coffee trees in cultivation are of heterogeneous
origin s.
C. liberica Bull ex Hiern. and C. canephora Pierre ex
Froehner. Low. Bark smooth or somewhat rough. Crown
open. Evergreen.
Cola nitida (Vent.) Schott & Endl. (Sterculiaceae)
The common cola tree, cultivated around the villages for
its edible caffein-rich seeds. Low-medium. Bark fissured,
soft. C rown dense. Evergreen.
Cussonia barteri Seemann (Arali'aceae)
Low savanna tree. Bark deeply furrowed. C rown open,
spreading. Deciduous.
Delonix regia (Boj. ex Hook.) Raf. (Leguminosae)
Fig. 1 8. Parinari excelsa. Grassfield 5 50 m. This tree is grown as an ornamental ("Flame of the forest"
or "Flamboyant"). Low-medium. Bark smooth. Crown
open, spreading. Foliage translucent. Deciduous.
Pentadesma butyraceaSabine (Guttiferae) Elaeis guineensis Jacq. (Palmae)
Medium. Bark rough. Crown rather dense. Evergreen. The oil pal m . Low-medium.
Piptadeniastrum africanum (Hook. f.) Brenan (Legumi­ Hevea brasiliensis (A. Juss.) Mull. Arg. (Euphorbiaceae)
nosae) The rubber tree. It is estimated (Kunkel 1 965 : 1 2) that this
Large-emergent. Bark smooth. Crown open and covers I % of the total land surface of Liberia. However,
spreading. Briefly deciduous. (Fig. 1 6). there are only minor rubber plantations within the Nimba
Pycnanthus angolensis (Welw.) Warb. (Myristicaceae) area.
Medium. Bark fissured or scaly. Crown open. Evergreen. Mangifera indicaL. (Anacardiaceae)
R hodognaphalon brevicuspe (Sprauge) Roberty (Bom­ The well-known Mango tree (locally called Plum tree).
bacacae) Medium. Bark rough. Crown dense. Evergreen.
Large-emergent. Bark scaly, soft. Crown rather dense. Musanga cecropioides R. Br. (Moraceae)
Deciduous. The common umbrella tree that rapidly colonizes old
Sacoglottis gabonensis(Baill.) Urb. (Humiriaceae) farmlands etc. Low-medium. Bark smooth. Crown rather
Large. Bark scaly, untidy. Crown very large, open, open. Evergreen.
spreading. Evergreen. Terminalia catappa L. (Combretaceae)

Terminalia ivorensis A. Chev. (Combretaceae) Cultivated for their edible seeds ("Almond tree") or as an
Large. Bark deeply grooved. Crown rather open. ornamental. Low-medium. Bark fissured. Crown
Deciduous. (Fig. I 7). spreading, open. Deciduous.

Terminalia superba Engl. & Diels (Combretaceae) Theobroma cacao L. (Sterculiaceae)

Large. Bark grooved and scaly. Crown open. Deciduous. The cacao tree. Low. Bark smooth. C rown dense.
(Fig. 1 7). Evergreen.

A cta phytogeogr. suec. 59


11. The epiphytic flora

Composition and geographical distribu­ 5000 in Malaysia alone (Van Steenis 1 93 8). The
tion general poverty of the African rain forest has been
suggested to be a result of past climatic changes
The flora of Africa · is generally considered to be (Mildbraed 1 922). Stewart (Stewart & C ampbell
poorer in species than those of other tropical regions. 1 970: 1 1 ) explains the relative low number of orchid
The orchids, of which the majority are epiphytes, species in Africa : "Vast tracts of the land mass are
have been estimated to comprise two to three thou­ unsuitable as orchid habitats."
sand species in Africa south of the Sahara, including However, the poverty of epiphytic plants in the
the island of the Malagasy republic (Madagascar), African rain forest is by no means an exception.
and the islands of the Seychelles, Comoro, and the From Sarawak, Borneo, Richards ( 1 93 6 : 1 5- 1 6)
Mascarene groups (Stewart & Campbell 1 970). This remark s : "One of the most striking features of the
figure may be compared to the more than three thou­ Sarawak rain forest, especially when compared with
sand species of orchids recorded in the South that of tropical South America, is the poverty of the
American republic of Colombia alone. Another ex­ epiphytic vegetation both in species and individuals."
ample from the orchid family may illustrate the The vascular epiphytic flora of the investigated
relative poverty of species. In the Zaire (former area includes a total of 1 5 3 species (the facultative
Belgian Congo) there are a total of 4 1 4 species epiphytes excluded). The epiphytes may be divided
(Nihoul et al. 1 964), compared with an estimated into three major groups : ( 1) the Pteridophytes with

Fig. 1 9. Records of
Angraecum birrimense (O)
and Cyrtorchis monteiroae
(e).

Fig. 20. Records of


Polystachya saccata (O)
and Rangai!'ris rhip­
salisocia (e).

A cta phytogeogr. suec. 59


22 Dick Johansson

Fig. 21. of
Records
Polystachya (0)
dalzielii
and P. leonensis ( •). (A)
Picket Hill (B) Mt
Kakouima (C) Loma Mts
(D) Wologisi Range (E)
Nimba Mts (F) Massif de
Dans (G) Mt C ameroon

Fig. 22. Records of


A ngraecum classensii <•),
A ngraecopsis elliptica (e),
Diaphananthe dens iflora
(.&) and Stolzia repens (0).

Fig. 23. Records of


Bulbophyllum magnibrac­
teatum(.&), B. pavimen­
(0) and B. schimpe­
tatum
ranum (e).

Fig. 24. Records of


Begonia rubro-marginata
(0) and Elaphoglossum
isabelense (e).

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 23

39 spp., (2) the Orchids with 1 0 1 spp. and (3) the rare. The flowering time is expressed in two months
remammg species, called the Other vascular periods (when the most frequent flowering occurs).
epiphytes, with 1 3 spp. This grouping is followed in A brief description of the plant is added for the
the list and analyses. orchids only. The particulars of the flowers are, due
The orchid flora seems to be composed of three to their limited duration, mostly of minor importance
different elements : (A) Species of the rain forest, e.g. in the field work. In most cases the determination
A ngraecum birrimense and Cyrtorchis monteiroae, has to be based on vegetative characteristics only.
(B) Species of the deciduous forest, e.g. A nsellia The inflorescences which often remain on the plant
africana and Rangaiiris rhipsalisocia, (C) and for several years have proved useful. (A field key to
species restricted to high altitudes, e.g. Polystachya the epiphytic orchids of Ghana based on vegetative
dalzielii and Polystachya leonensis (Figs. 1 9 -2 1 ). A characteristics has been presented by Hall & Bowl­
few species, e.g. Bulbophyllum inflatum, have only ing 1 96 9 , and in a key to the orchids of Zaire, former
been reported from the Nimba Mountains in Liberia. Belgian Congo, Tournay (1 9 55) used a similar
However this endemic character may only result technique.) Some species are difficult to determine
from insufficient knowledge (cf. Schnell 1 9 52 :407 even in a flowering state. The methods used to dis­
and Curry-Lindahl 1 96 9 : 23). tinguish such species are discussed in connection
Some species with a ± wide distribution in tropical with the presentation of the species. The size of the
Africa, e.g. A ntrophyum immersum and plants may vary widely due to environment.
Nephrangisfiliformis are in West Africa restricted to In the following list the grouping previously
the Nimba region. Several other species have had presented is followed. Within each of the three
their previously known area of distribution extended groups the names of the genera and their species are
westwards, e.g. A ngraecum classensii, A ngraecopsis given in alphabetical order. The first set of my collec­
elliptica, Begonia rubro-marginata, Bulbophyllum tions (with few exceptions), is kept in the Herbarium
magnibracteatum, B. pavimentatum, B. schimpera­ of the Institute of Systematic Botany at the Universi­
num � Diaphananthe densijlora, Elaphoglossum ty of Uppsala (UPS). A second (almost complete) set
isabelense, and Stolzia repens (Figs. 2 2 -24). A total has been given to the Kew Herbarium (K). For each
of thirty-two species were recorded in Liberia for the species a voucher is quoted, e.g. D .J. 43 1 (UPS).
first time.
Pteridophytes

Antrophyum Kau lf.


List of species A. mannianum Hook.
Guin., S.L., Lib., S.Nig., Cam.
The following list for the Nimba area includes for A species of the high forest. Moss-covered trunks high
above the ground or on the outer branches. "Occasionally
each species ( I ) their known geographical distribu­
on mossy boulders" (Harley 1 95 5 :9 I ). Open shade, bark.
tion in West Africa, (2) a brief autecologic descrip­ Rare. D.J. 5 1 6 (K).
tion, and (3) (for the orchids) some field A. immersum (Bory ex Willd.) Mett.
characteristics. Lib.
The notes on geographical distribution are based In West Africa only recorded from the Nimba area. In
on F. W.T.A. 2nd ed. (1 9 54, 1 9 59 , 1 96 8) on Schnell humid habitats, almost exclusively on the trunks of large
trees. Heavy - open shade, bark - minor humus deposits.
1 9 52 , Harley 1 9 55, 1 9 56 , 1 9 59 , Alston 1 959 ,
Rare. D.J. 5 5 4 (K), 582 (UPS).
Kunkel 1 96 2 a, Ake Assi 1 96 3, Guillaumet 1 96 7 ,
A dam 1 9 7 1 , and m y own collections. The ab­ Arthropteris J. S m .
breviations of the countries follow F.W.T.A. 2nd ed. A . monocarpa (Cordem.) C . Chr.
1 963, 2. Guin., S.L., Lib., lv.C ., Ghana, C am.
The autecological notes are based upon the quan­ Basal parts of trunks (particularly on the 'trunks' of the
tree fern Cyathea manniana). Damp habitats in heavy
titative studies (Tables 2 7 -29) and general obser­
shade, minor humus deposits. Rather common, similar to
vations, and the terminology for light and substrate A. orientalis, which, however, occurs in better lit habitats.
is in accordance with the one used during the field D.J. 48 1 (K, UPS).
work. The abundance is given in three classes, sub­ A. palisoti (Desv.) Alston
jectively estimated, common, rather common and Lib., Iv.C., Ghana, S.Nig., Cam.

A cta phytogeogr. suec. 59


24 Dick Johansson

Fig. 25. A splenium barteri.

Harley { 1 95 5 ) reports three localities in Liberia. Middle


part of branches of tall trees. Open shade, bark. Rare. D .J.
558 (K).
A. orientalis(Gmel.) Posth.
Guin., S.L., Lib., Jv.C., Ghana, N.Nig., S.Nig., C am.
Basal parts of large branches. Open shade, humus
deposits. Rather common. D.J. 806 (K, UPS).

Asplenium L.
A. aethiopicum (Burm.) Becherer
Guin., S.L., Lib., N.Nig., Cam.
Harley { 1 95 5 :84) and Kunkel ( 1 962 :40) report it from
various parts of Liberia. Trunks and branches. Open
shade, humus deposits. Rare. D.J. 782 (K, UPS).
A. africanum Desv. Fig. 26. Drynaria laurentii, sterile and fertile leaves.
Guin., S.L., Lib., Jv.C. , Ghana, S.Nig.
Basal parts of large branches. Open shade, humus
deposits. Rare. D.J. 852 (K), 834 (UPS). The roots hold A. megalura Hieron. ex Brause
humus in a fashion similar to Microsorium punctatum, Guin., S.L., Lib., Jv.C., Ghana, Togo.
which it resembles at a distance. Middle - outer parts of large branches. Full sun, humus
deposits. Common. D.J. 780 (K, UPS).
A . barteri Hook.
Guin., S.L., Lib., Jv.C ., Ghana, C am. A. variable Hook. var. paucijugum (B allard) Alston.
Basal parts of the trunks or moss-covered rocks. Humid Guin ., S.L., Lib., Jv.C., Ghana, C am.
habitat. Heavy shade, varying substrate. Common. D.J. This taxon is often referred to as A splenium paucijugum
5 5 5 (K), 629 (UPS). Fig. 25. Ballard (Harley 1 95 5 :86, Kunkel 1 96 2 :43). Basal parts of
trunks of tall trees or on moss-covered rocks. Wet and
A. dregeanum K unze
humid habitats. Heavy shade, humus deposits. Rather
Guin., S.L., Lib., Jv.C ., Ghana, Togo, S.Nig., N.Nig.,
common. D.J. 5 89 K, 786 (UPS).
Cam.
On trunks in humid habitats. Heavy shade and humus
deposits. Rather common at altitudes above 1000 m. D.J. Da vallia S m .

5 1 3 (K, UPS). D . chaerophylloides (Poir.) Steud.


Guin., S.L., Lib., Jv.C., Ghana, Togo, N.Nig., S.Nig.,
A . geppiiC arruth. Cam.
Syn : A. anisophyllum Kunze Middle parts of large branches. Open shade - full sun,
Guin., S.L., Lib., Jv.C., Ghana, S.Nig., Cam. bark , - minor humus deposits. Rather common. D.J. 660
Basal ramifications of tall trees. Open shade, large humus (K, U PS).
deposits. Rather common. D.J. 624 (K), 5 70 (UPS).
A. hemitomum Hieron. Drynaria ( B ory) J . S m .
Guin., S.L., Lib., Jv.C., Ghana, N .Nig., S.Nig., Cam. D. laurentii (C hrist.) Hieron.
On trunks, often rather close to the ground. Open shade, Guin., S. L., Lib., Jv.C., Ghana, Togo, S.Nig.
humus deposits. Rare. D.J. 1 060 (K, UPS). Basal - middle parts of large branches. Open shade - full

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 25

sun. Colonizes bark surfaces. This is probably the most


common epiphyte of the Nimba area. Several other
epiphytes are associated to this fern in all types of forests.
D.J. 43 1 (UPS). Fig. 26.

Elaphog/ossum Schott
E. barteri (Bak.) C .C hr.
Guin., S.L., Lib., lv.C., S.Nig., Cam.
Basal parts of trunks. Heavy - open shade, large humus
deposits. Rare. D.J. 5 5 3 (K), 697 (UPS).
E. chevalieri C hrist.
Guin., S.L., Lib., lv.C.
B asal parts of moss-covered trunks. Heavy - open shade,
humus deposits. Restricted to altitudes above 1 1 00 m.
Rare. D.J. 642 (K, UPS).
E. isabelense Brause
Lib. (new), F.Po.
Middle - basal parts of large branches of tall trees. Open
shade - full sun, bark or minor humus deposits. Common.
D.J. 495 (K , UPS).
E. kuhnii Hieron.
S.L., Lib., Cam.
Basal parts of tall trees, or rotten logs. Humid habitats.
Heavy shade, humus deposits. Rare. D.J. 493 (K), 646
(UPS).
E. salicifolium (Willd. ex Kaulf.) Alston
Guin., Lib. (new), C am.
Widely distributed over the phorophyte, mostly in the
basal parts of the large branches. Full sun - open shade,
bark. Common. D.J. 644 (K), 825 (K, UPS).
Fig. 27. Nephrolepis biserrata.
Lomariopsis Fee
L . guineensis (Underw.) Alston humus deposits. Rather common. Often associated with
Guin., S.L., Lib., Iv.C . , Ghana, Togo, D ah., S.Nig., C am. Drynaria laurentii.D.J. 5 3 5 (K, UPS).
Reported from various parts of Liberia (Harley 1 95 5 :87,
Kunkel 1 9 6 2 : 5 2). Basal parts of trunks. Heavy - open Microgramma Presl
shade, bark. Rare. Rooted at the base of the phorophyte,
M. o wariensis (Desv.) Alston
which it climbs. Very tightly attached to the bark by short
Guin., S.L., Lib., Jv.C., Ghana, Togo, N.Nig., S.Nig.,
rootlets of the rhizome. After some time it normally loses
Cam.
contact with the ground. D.J. 7 7 7 (K), 482 (UPS). Fig.
On trunks and branches. Open shade, minor humus
1 1 6.
deposits. Common on trees and palms around villages,
less frequent in the high forest. From Liberia a new form,
Loxogramme ( B i u m e) Presl f. nana, has been described (Kunkel 1 962:37- 3 8) . D .J.
L. lanceolata (Sw.) Presl 5 6 7 (K, UPS).
Guin., S.L., Lib., Iv.C ., Ghana, N.Nig., S.Nig., Cam.
Mainly on the trunks. Open shade, minor humus deposits. Microsorium L i n k
Rare. D.J. 8 3 3 (K), 742 (UPS).
M . punctatum (L.) Cope!.
Guin., S.L., Lib., Ghana, S.Nig., C am.
Lycopodium L. Basal - middle parts of large branches. Open shade - full
L . mildbraedii Hert. Syn : L. dacrydioides Bak. sun, minor humus deposits. Common. The debris held by
Guin., S.L., Lib. (new), Cam. the root system is often utilized by other epiphytes. D.J.
On moss-covered branches and trunks. Open shade, 794 (UPS).
minor humus deposits. Common at altitudes above 1 000
m. D.J. 503 (K, UPS).
Nephrolepis Sc hott
L . warneckei (Hert.) Alston N. biserrata (Sw.) Schott
Guin., S.L., Lib., Jv.C., Cam. Guin., S.L., Lib., Jv. C ., Ghana, Togo, S .Nig.
In the outer - middle parts of large branches. Full sun, Epiphytic or terrestrial, equally common in both habitats.

A cta phytogeogr. suec. 59


26 Dick Johansson

As epiphyte mainly at the basal part of the large branches. Tectaria Cav.
Open shade - full sun, large humus deposits. Common in T. angelicifolia (Sebum.) C opel.
secondary forests (particularly at the leaf bases of oil Guin., S.L., Lib., Iv.C., Ghana, N .Nig., S.Nig., C am.
palms). Rare in the high forest. Used by the natives in In the basal parts of tall trees (or on rocks). Frequently on
treatment of snake bites and urinary complaints (Harley the trunks of Cyathea manniana. Heavy shade, on
1 94 1 ). D.J. 765 (K). Fig. 27. various kinds of substrate. Common. D.J. 1 05 9 (UPS).
N. undulata (Afzel. ex Sw.) J.Sm. T. fernandensis (Bak .) C.Chr.
Guin., S.L., Lib., Iv.C., Ghana, Togo, Dab., N.Nig., Guin., S.L., Lib., Iv.C . , Ghana, Togo, S.Nig., C am.
S.Nig., Cam. Basal parts of large trees. Heavy shade, humus deposits.
Basal - middle parts of large branches. Open shade - full Only in very humid habitats. Rare. D.J. 795 (K).
sun, minor - large humus deposits. Occurs in several
epiphyte communities, particularly with Platycerium Vittaria S m .
ferns. Common. D.J. 707 (K, UPS). V. guineensis Desv.
Guin., S.L., Lib., Jv.C., Ghana, Togo, S.Nig., C am.
On all parts of the trees, but mainly in the basal and
Oleandra Cav.
0. distenta Kunze
middle parts of large branches. Open shade, minor humus
deposits. Often associated with Microsorium punctatum.
Guin., S.L., Lib., Iv. C ., Ghana, Togo, S.Nig., Cam.
Common. D.J. 762 (UPS).
Around the basal ramifications of the tall trees. Open
shade, humus deposits. Common in the high forest and on Xiphopteris K a u lf.
the trunks of the oil palm in farmlands. The humus held X. oosora (Bak.) Alston
by this species is utilized by several other epiphytes. D.J. Guin., S.L., Lib. (new), C am., F.Po.
805 (K), 427 (UPS). Fig. 80. Basal - central parts of large branches. Open shade,
humus deposits. At altitudes above 1 200 m. Rather com­
Phymatodes Presl mon. D.J. 826 (K), 5 1 4 (UPS).
P. scolopendria (Burm.) Ching.
X. serrulata (Sw.) Kaulf.
Guin:, S.L., Lib., Iv.C., Ghana, Dab., Togo, N .Nig.,
S.L., Guin., Lib., Jv.C., F.Po.
S.Nig., Cam., F.Po.
Middle parts of large branches. Open shade, minor humus
Basal - middle parts of large branches in high forest. On
deposits. R are. D.J. 592 a (UPS).
trunks of trees and oil palms around villages. In high
forest in open shade, bark - minor humus deposits. In X. villosissima (Hook.) Alston
secondary forests it is found in a wider range of habitats. Guin., S.L., Lib., Jv.C., C am., F.Po.
Common. D.J. 5 3 2 (K), 8 3 1 (UPS). Basal and middle parts of large branches. Open shade,
minor humus deposits. At altitudes above I 1 00 m. Rare.
Platycerium Desv. D.J. 5 1 8 (K), 827 (UPS).
P. angolense Welw. ex Hook.
Guin., S.L., Lib., Iv.C., Ghana, N.Nig., S.Nig. Orchids
Middle parts of large branches of emergent trees or on the Aerangis Rchb.
trunks of trees left in farmlands. Full sun, bark. Common A . biloba (Lindl.) Schltr.
on large Triplochiton scleroxylon trees in the Yekepa Sen., Guin., S.L., Lib., Iv.C., Ghana, Togo, N.Nig.,
community. Rare in the high forest. In Iv.C . this species S. Nig., Cam.
occurs mainly in the deciduous forests in the northern part Basal or middle parts of large branches. Full sun - open
of the country (Boyer 1 964 ). D.J. 809 (UPS). Fig. 85. shade, bark - minor humus deposits. �Common. Growing
P. stemaria (P. Beauv.) Desv. tips of roots orangish-pink (Piers 1 95 9, Moir 1 963, San­
Guin., S.L., Lib., Jv.C., Ghana, Dab., N.Nig., S.Nig., ford 1 968). Stem short, woody. Leaves crowded at the
C am. apex of the stem, 8- 1 8 cm, oblanceolate, curved and un­
In its ecology very similar to P. angolense. Rare in high equally bilobed, dark green. Inflorescence longer than the
forest, more common on tall trees in secondary forest. In leaves, many flowered, somewhat pendulous. Flowers
Iv.C., this species occurs mainly in the secondary white, star-like. June-July. D.J. 5 62 (UPS). Fig. 28.
evergreen forest in the southern part of the country, and is A . laurentii (De Wild.) Schltr.
± absent in the primary rain forest (Boyer 1 964). D.J. 79 1 Lib., Ghana.
(UPS). Fig. 84. Anywhere on the large branches. Full sun, bark. Rare.
Erect, 10- 1 5 cm. Stem short, woody. Leaves strap-shaped,
Pyrrosia M i rb. arranged in a fan-shaped manner, acutely v-shaped in
P. mechowii (Hieron.) Alston cross-section, rather thick and stiff. Inflorescence longer
Guin., Lib. (new), N.Nig., C am . than the leaves, zigzag-shaped rachis. Flowers white, with
Middle parts o f large branches. Open shade - full sun, a long spur. Jan.-Febr. The zigzag rachis distinguishes it
minor humus deposits. Humid habitats, e.g. swamp from Rangae'ris muscicola. Plectrelminthus caudatus has
forests. Almost exclusively on Mitragyna ciliata. Rare. also a zigzag rachis, but leaves of a different shape. D J .
D.J. 808 (K, UPS). 4 4 9 (K), 4 1 1 (UPS).

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 2 7

Ancistrochilus Ro lfe
A. rothschildianus O'Brien
Guin., S.L., Lib., Iv.C., S.Nig., C am.
Basal or central parts of large branches, occasionally on
moss-covered trunks. Open shade, humus deposits.
Rather common. Erect, 20-30 cm. Pseudobulbs crowded,
conical - pyriform, 2-3 cm high, up to 5 cm in diameter,
longitudinally sulcated, green. Leaf 1 5-30 cm, lanceolate ­
oblanceolate, thin, ribbed, dark green. Inflorescence long,
curved. Flower large (8 cm in diam.). Oct.-Nov. D.J. 703
(UPS). Fig. 29.

Ancistrorhynchus F i net
A. capitatus (Lindl.) Summerh.
S.L., Lib., S.Nig.
Basal parts of large branches or on trunks. Open shade -
full sun, bark - minor humus deposits. Common. Erect,
25-30 cm. Stem short, woody. Leaves crowded, curved,
20-50 x 1 .5-3 cm, with a few teeth at apex. Inflorescences
short, dense, on the stem below the leaves, long persistant.
Flowers small, white. July-Aug. D.J. 5 3 1 (K, UPS).
A. cephalotes (Rchb.f.) Summerh.
Guin., S.L., Lib., Iv.C., Ghana, N.Nig., S.Nig.
Trunks or the middle parts of large branches. Full sun,
bark. Common. Erect or scrambling. Stem long, woody,
twisted, covered with old leatbases and inflorescences.
Leaves 8-20 cm, linear, apex bilobed. Inflorescences on
the stem below the leaves, short, with a very large number
of flowers. Flower white, small. June-July. D .J. 524 (K,
Fig. 28. Aiirangis biloba. UPS). Fig. 1 1 0.
A. clandestinus (Lindl.) Schltr.
S.L., Lib., lv.C., Ghana, N.Nig., S.Nig., C am.
Trunks or basal parts of larger branches. Full sun, bark.
Rare. Erect, in the Nimba area a small to medium-sized
plant, stem short, woody. Leaves 5-8 x 0.5- 1 .0 cm, stiff
and fleshy, leathery, apical lobes of the leaf very acute and

Fig.29. A ncistrochilus
rothschildianus.

A cta phytogeogr. suec. 59


28 Dick Johansson

unequal in length. (F.W.T.A. 2nd ed. 3 : 1 , 1 96 8 :2 70 gives Angraecum Bory


the length of the leaves up to 1 80 cm.) Inflorescence very A. birrimense Rolfe
short, dense, from the stem below the leaves. Flowers S.L., Lib., Iv.C., Ghana, S.Nig., C am.
white, small. May-June. The leaves are similar to Trunks, basal or middle parts of large branches. Open
Chamaeangis vesicata, but the species is distinguished by shade, humus deposits. Common. A climbing and pen­
the inflorescence. D .J. 804 (UPS). dulous plant, 1 -2 m long. Stem woody. Leaves 6- 1 2 cm,
scattered, lanceolate - elliptical, bilobed, fleshy, light
A . recurvus Finet
green. Inflorescence short. Flowers large. Tepals light
Guin., Lib. (new), Ghana, S.Nig., C am.
green, lip white with a green centre. July-Aug. D.J. 436
Occurs in the same habitats as A . clandestinus. Rare.
(K), 568 (UPS). Fig. 30.
Leaves 8- 1 5 x 1 .2- 1 . 7 cm, not particularly stiff. Floral
parts as A . clandestinus, but leaves of a different shape. A . chevalieri Summerh.
D.J. 8 10 (UPS). Guin., Lib., Jv.C ., Ghana, N.Nig., S.Nig.
B asal parts of large branches. Open shade - full sun, bark
Angraecopsis S u m merh. - minor humus deposits. Rare. Climbing, somewhat pen­
A. elliptica Summerh. dulous, stem woody. Leaves 3-7 cm, narrowly strap­
Lib. (new), N.Nig., C am. shaped, fleshy, leathery. Inflorescence short. Flowers
Basal or central parts of large branches. Open shade, yellowish, small. June-July. D.J. 1 05 5 (UPS).
humus deposits. Rare. Erect, 6- 1 0 cm. Stem very short. A. classensii De Wild.
Leaves 4- 1 0 x 1 -2 cm, ligulate, curved. Inflorescence Lib. (new), Nig.
filiform, ± as long as the leaves. Flowers very small and Trunks and branches of small trees, close to the ground.
tiny, yellow-green. Aug.-Sept. D .J. 6 1 1 (K, UPS). Full sun - open shade, bark. Rare. Ascending, 1 0-30 cm
long. Stem woody. Leaves 6- 1 0 x 1 . 5-3 cm, oblong -
obovate, bilobed at apex. Inflorescences erect, 5-25 cm,
usually much longer than the leaves. Flower ochra­
coloured. May-June. D.J. 5 1 2 (K, UPS).
A . distichum Lindl.
Guin., S.L., Lib., Iv.C., Ghana, D ah., N.Nig., S.Nig.,
Cam.
Trunk s or basal parts of large branches. Full sun - open
shade, bark - minor humus deposits. Common. A dense
tuft of branching stems, ± erect, 8-20 ·cm. Leaves 1 .0 x
0.7 cm, fleshy, triangular, distichuous. Flower sessile,
white, small. May-June. D .J. 5 25 (UPS). Fig. 5 9.
A . podochiloides Schltr.
Lib., Iv.C., S.Nig., C am.
Trunks of tall trees. Open shade - full sun, · bark. Rare.
Pendulous, 20-60 cm. Leaves crowded, 1 -2 cm, narrowly
)anceolate - oblong, fleshy, hard. Flower small, sessile,
white. June-July. D.J. 6 1 0 (UPS). Fig. 5 9.
A . subulatum Lindl.
S.L., Lib., lv.C., Ghana, S.Nig., Cam.
Basal or middle parts of large branches. Open shade,
minor humus deposits. Rare. Erect - pendulous. Leaves 3-
8 x 0. 1 -0.3 cm, subulate - terete, falcate. Flowers small,
sessile, white. June-July. Two other species with terete
leaves are Nephrangisfiliformis and Tridactyle tridentata.
They are distinguished by a marked pendulous growth.
D.J. 5 5 7 (K), 766 (UPS).
Ansellia Li ndl.
A. africana Lindl.
Lib., Jv.C., Ghana, S.Nig., F.Po.
Basal or middle part of large branches. Full sun, large
humus deposits. Erect 50 - 80 cm. Pseudobulb 30-50 cm.
Leaves lanceolate, from the upper part of the pseudo bulb.
Flower yellow witl:I large chocolate brown spots. Feb.­
March. Numerous erect aerial roots occur at the base of
the plant (cf. Graphorchis lurida). D.J. 790 (UPS). Figs.
Fig. 30. A ngraecum birrimense. 3 1 and 62.
A cta phytogeogr. suec. 59
Ecology of vascular epiphytes in West African rain forest 29

Fig. 32. Brachycorythis kalbreyeri.

line, depressed, obtusely 3-4 angled, pale green. Leaf 3 -6


cm, elliptical - oblong, flat, stiff, green. Inflorescence 2-4
times longer than the leaf, descending. Bracts large,
spreading. Flower small, dark purple. Lip mobile, densely
hairy. Longer hairs with club-shaped swollen ends. Oct.­
Nov. D.J. 6 5 8 (UPS).
B. bifarium Hook. f.
Lib. (new), lv.C., Cam.
Middle or outer parts of branche's. Open shade, minor
Fig. 3 1 . A nsellia qfricana.Photo: Harvey Boettcher. humus deposits. At altitudes above 1 000 m. Rare. Erect,
5 - 1 0 cm. Pseudobulbs crowded, 1 . 5 -2.0 cm, ellipsoid or
ovoid with sharp angles or protruding ridges, green with
Bolusiella Sch ltr. purple stripes and spots. Leaf 4-8 cm, strap shaped, light
B. ta/botii (Rendle) Summerh. green. Inflorescence equating or longer than the leaves,
S.L., Lib., Iv.C . , Ghana, S.Nig. bracts large, orange-red. Flower small, pale green. Oct.­
Outer parts of large branches. Full sun, bark. Rather com­ Nov. D.J. 706 (K, UPS).
mon. Erect, 5-8 cm. Stem very short, woody. Leaves 2-4
cm, arranged in a fan-like manner, narrowly lanceolate, B. bufo (Lindl.) Rchb.f.
distinctly narrowed in the upper half, compressed, fleshy. Guin., S.L., Lib., Iv.C., Ghana, S.Nig., C am.
Inflorescence thin, considerably longer than the leaves. Basal parts of large branches. Open shade, humus
Flowers scattered, minute, white. June-July. D.J. 438 (K), deposits. Rather common. Erect, 1 5-25 cm. Pseudobulb
1 054 (UPS). Fig. 1 1 8. 3-5 cm, ovoid or conical ovoid, quadrangular. Leaf 8- 1 5
cm, oblanceolate, green. Inflorescence erect, longer than
the leaves. Rachis flattened up to one cm wide. Flower
Brachycorythis Li ndl.
small. Sept.-Oct. This and some other B. spp. (e.g. B.
B. kalbreyeri Rchb. f.
falcatum) form a group in which it is difficult to dis-
Guin., S.L., Lib., C am.
tinguish the species on vegetative characteristics only.
Basal parts of the trunks. Open shade, large humus
D.J. 7 1 1 (K, U PS).
deposits. Rare. Erect, up to 40 cm. Stem slender, spotted
purple. Leaves evenly distributed along the stem, 6-9 cm, B. buntingii Rendle
lanceolate, thin in texture. Flower large, up to 5 cm in Guin., Lib., lv.C., N.Nig., S.Nig., Cam.
diam, lilac-lavender. June-July. In the dry season the Middle parts of large branches. Full sun, bark. Rather
epigeal parts of the plant wilt and then it is difficult to common . Erect, 5 - 1 0 cm. Pseudobulb 1 .5-2.0 cm, bluntly
trace but occasionally one finds the tubers in the humus. triangular, hard, red-brown, with a smooth or slightly
D.J. 497 (UPS). Fig. 3 2. reticulate surface. Leaf 4-7 cm, elliptical, obtuse, bluish­
green, purplish beneath. Inflorescence erect, double the
Bulbophyl/um Thou. (Fig. 3 3 ) length of the leaf, densely flowered almost to the base.
B. barbigerum Lindl. Flower small, white. Oct.-Nov. D.J. 688 (K, UPS).
S.L., Lib., Iv.C., Ghana, S.Nig., C am. B. calamarium Lindl.
Middle parts of large branches. Open shade, bark. Rare. S.L., Lib., S.Nig.
Prostrate 4- 1 0 cm. Pseudobulb 1 -3 cm, elliptical in out- Middle or outer parts of large branches. Open shade,

Acta phytogeogr. suec. 59


Fig. 3 3 . Simplified drawing of the Bulbophyllum species in barbigerum, 1 6. B. inflatum, 1 7. B. winkleri, 1 8. B.
the Nimba area. (The species are not in proportion to each linderi, 1 9. B. magnibracteatum, 20. B. imbricatum, 2 1 .
other.) 1. Bulbophyllum josephii, 2. B. cocoinum, 3. B. B. falcatum, 22. B. melanorrachis, 23. B. lucifugum, 24.
flavidum, 4. B. pavimentatum, 5. B. recurvum, 6. B. B. bufo, 25. B. congolanum, 26. B. maximum, 27. B.
nigritianum, 7. B. schimperanum, 8. B. buntingii, 9. B. lupulinum, 28. B. cochleatum, 29. B. oreonastes, 30. B.
schinzianum, 1 0. B. phaeopogon, 1 1 . B. distans, 1 2. B. rhizophorae, 3 1 . B. scariosum, 32. B. bifarium.
calamarium, 1 3. B. intertextum, 1 4. B. saltatorium, 1 5 . B.

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 31

minor humus deposits. Rare. Prostrate, 1 5 -20 cm. deposits. R are. Erect, 8- 1 6 cm. Pseudobulb 1 -2 cm, light
Pseudobulb 4-5 cm, conical or ovoid, 3-4 angled, velvety, green, velvety. Leaf 6- 1 2 cm, oblanceolate-narrowly
light green. Leaf 1 2- 1 5 cm, flat, stiff, oblong, green. oblong, acute, green. Inflorescence about the same length
Inflorescence 40-60 cm, bracts less than 1 5 mm, overlap­ as the pseudobulb and the leaf, arching, densely flowered.
ping each other. Flower small, lip hairy, sepals strongly Flower small, yellow or pink. June-July. D.J. 5 8 5 (K), 5 8 6
reflexed, yellow with purple spots. July-Aug. Difficult to (UPS).
distinguish from B. schinzianum and B. phaeopogon B. imbricatum Lindl.
without flowers. The size of the bracts gives a possibility. S.L., Lib., Jv.C., Ghana, S.Nig., C am.
D.J. 8 5 6 (UPS). In the middle parts of large branches. Open shade, minor
B. cochleatum Lindl. humus deposits. Rare. Erect, 1 5-25 cm, stout. Pseudo­
Guin., S.L., Lib., lv.C ., N.Nig., S.Nig., Cam. bulbs 3-4 cm, ovoid, green, scattered along the rhizome.
Basal or middle parts of large branches. Open shade, bark Leaf 1 2-20 cm, linear - oblong. Inflorescence longer than
or in minor humus deposits. Common. Erect, 1 2-24 cm. the leaves, erect stout. Rachis fleshy, dark purple. Flower
Pseudo bulbs scattered, 3 - 1 0 cm, narrowly cylindrical, small with purple spots. Sept.-Oct. D .J. 7 1 0 (K), 678
bluntly quadrangular, brownish-green with brown or pur­ (UPS).
ple stripes or dots. Leaf 8- 1 4 x 0.5- 1 . 2 cm, strap-shaped, B. injlatum Rolfe
green, sometimes tinged purple. Inflorescence up to 30 cm S.L., Lib.
long, curved. Flowers small, orange-red. Sept.-Oct. D .J. Trunks or in the basal parts of large branches. Heavy -
63 1 (K, UPS). _ open shade, minor humus deposits. At altitudes above
1000 m. Rather common. Pendulous, 8-20 cm. Pseudo­
B. cocoinum Batem. ex Lindl. bulb 1 -3 cm, ovoid quadrangular, rhomboid in cross sec­
S.L., Lib., Ghana. tion, green. Leaf 7-20 x 1 .5-5 cm, lanceolate - oblong or
B asal and middle parts of large branches. Open shade, elliptical, leathery, green. Flower small, yellow-green.
humus deposits. Rather common. Erect, 1 5-30 cm. Jan.-Feb. Easy to recognize. The plant grow upside down.
Pseudobulbs crowded, 2-4 cm high, 1 . 5-2.0 cm broad, This species is very similar to B. comatum Lindl., in a ster­
ovoid, sharply 3-4 angled, pale yellow. Leaf lanceolate or ile as well as fertile state. The latter has not been recorded
oblanceolate, 1 0-20 cm, acute, not stiff. Inflorescence in the Nimba Mts but from Mt Momi in the Ivory Coast.
long, arching, densely flowered. Flowers small, creamy In the field work specimens without flowers that could be
white. Aug.-Sept. D .J. 54 1 (K, UPS). any of the two species have been recorded as B. injlatum.
B. congolanum Schltr. D.J. 7 1 7 (K, UPS).
Guin., S.L., Lib., Iv.C., Ghana, N . Nig., S.Nig. B. intertextum Lindl.
Middle parts of large branches. Full sun, bark or minor S.L., Lib., Ghana, N .Nig., S.Nig., Cam.
humus deposits. Rather common. Erect 8- 1 4 cm. Middle parts of large branches. Full sun, bark. Common.
Pseudobulb 2-4 cm, ovoid - conical, 3-4 angled, light Erect, 2-5 cm. Rhizome slender, forming mats. Pseudo­
green-yellow. Leaf 6- 10 cm, strap-shaped, green. bulbs scattered, up to one cm, elliptical - ovoid, green.
Inflorescence erect, somewhat curved, of about the same Leaf 1 -2.5 x 0.5- 1 .0 cm, elliptical, light green.
length as the leaves. Flower, small creamy-white. Oct.­ Inflorescence 1 4-20 cm, slender. Flower minute, creamy­
Nov. D.J. 7 1 4 (K, UPS). Fig. 58. white. Oct.-Nov. A somewhat different form of this
B. distans Lindl. species occurs : Pseudobulb of similar shape and size as
Lib., Jv.C ., Ghana, S.Nig., C am. above, but with a purplish colour. Leaf strap-shaped, 1 -4
Middle or basal parts of large branches. Full sun - open x 0.3-0.5 cm, dark greenish-purple. Inflorescence much
shade� minor humus deposits. Rather common. Ascen­ shorter, 4-8 cm. Flower minute, purplish. Oct.-Nov. D.J.
ding, 1 0- 1 5 cm. Pseudobulb depressed, 2-3 cm, conical­ 687, 6 9 1 (K, UPS).
ovoid, yellow-green. Leaf 8- 1 2 cm, oblong, flat, stiff, B. josephii (K unze) Summerh.
green. Inflorescences (often two) erect, slender, 30-50 cm. Guin., S.L., Lib. (new), lv.C., S.Nig., C am.
Bracts 6- 10 mm. Flower small, dark purple, lip densely Basal parts of large branches. Open shade and humus
hairy. June-July. D.J. 8 5 5 (K), 857 (UPS). deposits. Rare. Erect, 1 2-20 cm high. Pseudobulbs
B. falcatum (Lindl.) Rchb.f. crowded, 2-3 cm, narrowly ovoid-conical ovoid, velvety,
Guin., S.L., Lib., lv.C., Ghana, S.Nig., C am., F.Po. redbrown. Inflorescence long, slender, arching. Flower
Middle part uf large branches. Open shade, minor humus purplish-violet. July-Aug. Other species with redbrown
deposits. Common. Erect. Pseudobulb 1 . 5-5 cm long, pseudobulbs are B. winklerii, and B. buntingii D.J. 44 1
ovoid, 3-4 angled, green. Leaves 3- 1 2 x 0.6-2 cm, oblong­ (K), 699 (UPS).
lanceolate or oblanceolate. Inflorescence very long, up to B. linderi Summerh.
50 cm. Rachis 0.5-0.8 cm broad. Flowers small, yellow­ S. L., Lib., Jv.C .
purple. Oct.-Nov. D.J. 668, 670 (K), 6 7 1 (UPS). Basal o r middle parts o f large branches. Open shade,
B. jlavidum Lindl. bark. Common. Erect, 1 2-20 cm. Pseudobulb 2-3 cm,
Guin ., S.L., Lib., Iv.C., Ghana, S.Nig., Cam. ovoid, triangular, with a shiny texture, green. Leaf 8- 1 6
Middle part of large branches. Open shade, minor humus cm, oblong - lanceolate, green, often red coloured along

Acta phytogeogr. suec. 59


32 Dick Johansson

the margin. Inflorescence erect, the apical part' with the B. oreonastes Rchb. f.
white fleshy rachis is held in a horizontal position. Guin., S.L., Lib., Iv.C ., Ghana, N . Nig., S.Nig., C am.
Flowers white. Oct.-Nov. D .J. 6 7 5 (K, UPS). Middle part of large branches. Open shade - full sun, bark.
The most common epiphytic orchid in the Nimba area.
B. lucifugum Summerh.
Erect, 6- 1 0 cm. Pseudobulbs spaced out along the
S . L., Lib. (new).
rhizome, 1 -3 cm, ovoid, quadrangular, yellowish-green.
Basal or middle parts of large branches. Open shade,
Leaves 2-6 cm, oblong, leathery, green. Inflorescence
minor humus deposits or on bark. Rare. Erect, 1 0-20 cm.
longer than the bulb and leaves together. Rachis narrowly
Pseudobulb 2-4 cm, conical - ovoid, obtusely
winged, flowers yellowish with purple lihes, lip purple.
quadrangular, surface wavy or verrucate, greenish with
March-April. This species is vegetatively similar to B.
purple spots. Leaf strap-shaped, green, 1 0-22 x 0.6-0.9
zenkeranum Kraenzl. The latter species has been recorded
cm. Inflorescence recurved, rachis flattened, green with
from the Kitoma Range, south of the Nimba area.
purple dots, crinkled at the edges. Flower small. Feb.­
Specimens without flowers that could be any of the two
March. D.J. 744 (K), 730 (UPS).
species have been recorded in the field work as B.
B. lupulinum Lindl. oreonastes. D.J. 640 (K, UPS).
Guin., S.L., Lib. (new), Ghana, N.Nig., S.Nig. B. pavimentatum Lindl.
Middle parts of large branches. Full sun, bark - minor Lib. (new), S.Nig.
humus deposits. Rare. Erect, 1 5-20 cm. Pseudobulb 4-6 Middle parts of large branches. Open shade, bark. Rare.
cm, rectangular, yellow green. Leaf 8- 1 8 cm, oblong, Erect, 1 2- 1 6 cm. Pseudobulbs crowded, about 2 cm long,
green. Inflorescence erect. Bracts large, with numerous ovoid, obscurely 2-3 angled, velvety green. Leaf 6- 1 4 cm,
short black or purplish hairs. Flower small, yellow with oblong- elliptical, dark green. Inflorescence equal with or
red spots. Jan.-Feb. D.J. 45 1 (K, UPS). longer than the leaf, thin, peduncle longer than the rachis.
B. magnibracteatum Summerh. Flower small, purplish. Oct.-Nov. D.J. 679 (UPS).
Lib. (new), Ghana, S.Nig. B. phaeopogon Schltr.
Middle parts of large branches. Open shade, bark. Rare. Lib. (new), Iv.C., Ghana, S.Nig., Cam.
Erect. Pseudobulbs 1 .5 -2.0 cm long, ovoid, 3 -angled. Outer parts of large branches. Open shade, minor humus
Leaves 5-7 x 1 .5 cm, stiff. Inflorescence of about the same deposits. Rare. Erect, 1 2- 1 6 cm. , Pseudobulbs erect,
length as the plant. Rachis 0.4-0. 7 cm broad with bracts crowded, 2-3 cm, ovoid, 3-4 angled, light green. Leaf
as broad as the rachis. When old and dry the rachis has a oblong- lanceo1ate, 7- 1 4 cm, stiff, green. Inflorescence 30-
characteristic obovate shape ending in a short acute apex. 60 cm. Bracts more than 1 5 mm long, overlapping.
Flowers small, purplish. April-May. D.J. 489 (UPS). Flower small, tepals yellow with red spots. Lip with long
hairs, mobile. July-Aug. This species is hardly dis­
B. maximum (Lindl.) Rchb. f. tinguishable from B. schinzianum when in a flowerless
S.L., Lib., Iv.C ., Ghana, S.Nig. state. D .J. 8 1 1 (K ), 106 1 (UPS). Fig. 54.
Middle part of large branches. Open shade - full sun, bark.
B. recurvum Lindl.
Rather common. Erect, 1 2-20 cm, stout. Pseudo bulbs 3-7
S.L., Lib., Cam., F.Po.
cm, ellipsoid - conical ovoid, acutely quadrangular, green,
Middle parts of large branches. Open shade, minor humus
scattered along the woody and stout rhizome. Leaf 7- 1 0
deposits-bark. Rare. Erect. Pseudobulbs 1 - 1 .5 cm long,
cm, oblong - elliptical, stiff, dull green. Inflorescence 1 2-40
ovoid or conical ovoid, obscurely 4-angled, green to brow­
cm, rachis 1 .0- 1 .5 cm broad, margins markedly undulate.
nish green. Leaves elliptical, 3-6 x 1 -2 cm, acute or obtuse.
Flowers minute, arranged along a central line of the
Inflorescence erect, 8- 1 3 cm, longer than the leaves,
rachis. Sept.-Oct. D.J. 656 (K, UPS).
recurved in the upper flowering � half. Flower small,
greenish-yellow. Oct.-Nov. D.J. 659 (UPS).
B. melanorrachis (Rchb. f.) Rchb. f. ex De Wild.
Guin., S.L., Lib., Ghana, S.Nig. B. rhizophorae Lindl.
Middle part of large branches. Open shade and minor S.L., Lib. (new), Jv.C., Ghana, S.Nig., C am.
humus deposits. Rare. Erect. Pseudobulbs 1 .5-3.5 cm On branches of small trees close to water. Open shade -
long, narrowly ovoid. Leaves 4- 1 0 x 0.7- 1 . 2 cm, oblong­ full sun, bark . Rare. Pseudobulb 2-3 cm long, elliptical.
oblanceolate, obtuse. Inflorescence 4-6 cm, rachis dark Leaves oblong, ± as long as the pseuodobulbs.
purple. Flowers crowded, small. Oct.-Nov. D.J. 648 (K, Inflorescence 6- 1 0 cm, erect. Flower small, yellowish­
UPS). white. Oct.-Nov. D.J. 748 (K).
B. nigritianum Rendle B. saltatorium Lindl.
S.L., Lib., Iv.C., Ghana, S.Nig. -S.L., Lib. (new), Ghana.
B asal or middle parts of large branches. Open shade, Basal or middle parts of large branches. Open shade, bark
minor humus deposits. Rare. Erect. Pseudobulbs 2-4 cm - minor humus deposits. Rare. Prostrate, lying on the sub­
long, conical ovoid, velvety, dark green. Leaf narrowly strate, 4-7 cm. Pseudobulbs crowded, 1 -2 cm, ovoid,
strap-shaped, I 0- 1 5 x 0. 7- 1 .0 cm. Inflorescence erect, 10- flattened, obtusely 3-4 angled, green. Leaf 3-5 cm, falcate,
1 5 cm long, flowering almost to the base. Flower small, oblong-elliptical, dark green. Inflorescence shorter than
white. Oct.-Nov. D.J. 693 (K), 669 (UPS). the plant. Flowers small, dark purple, lip hairy, almost all

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 33

flowering simultaneous. March-April. D.J. 468 (K, UPS).


B. scariosum Summerh.
Guin., S.L., Lib.
Middle or outer parts of large branches. Full sun, bark -
minor humus deposits. At altitudes above 1 000 m. Com­
mon. Erect, 8- 1 6 cm. Pseudobulb 2-4 cm, ovoid,
quadrangular, light green-yellow, turning redbrown when
old. Leaf 4- 1 2 cm, oblong, light green. Inflorescence erect,
arching in the apical part. Flower creamy white. March­
April. D.J. 7 1 6 (K, UPS).
B. schimperanum Kraenzl.
Lib. (new), S.Nig., Cam.
Middle or outer parts of large branches. Full sun, bark.
Rather common. Erect. Pseudobulbs 1 -2 cm long, ovoid,
triangular, light green-yellow, often tinged with red. Leaf
5 - 1 0 x 1 .5-2.0 cm, oblanceolate - oblong, obtuse.
Inflorescence erect, 1 0- 1 5 cm long, carrying flowers
almost to the base. Flowers small, white. Sept.-Oct. D.J.
890 (K), 694 (UPS).
B. schinzianum Kraenzl.
Lib., lv.C., N .Nig.
Middle parts of large branches. Open shade - full sun,
bark. Rather common. Erect, I 0-25 cm. Pseudobulbs
crowded, erect, 2-4 cm, ovoid, sharply quadrangular, light
yellow-green. Leaf 1 0-20 cm, oblong-elliptical, stiff, green.
Inflorescence 20-60 cm. Flower small, dark purple; lip
with short dense hairs. Sept.-Oct. In flowerless state
similar to B. phaeopogon. D.J. 435 (K), 64 1 (UPS). Fig.
53.
B . winkleri Schltr.
Lib., N.Nig., S.Nig., Cam.
Upper parts of trunks or basal parts of large branches. Fig. 34. Calyptrochilum emarginatum.
Open shade, bark. Rather common. Prostrate, 8- 1 5 cm. Chamaeangis Sch ltr.
Pseudobulbs crowded, 1 -2.5 cm, ovoid, obtusely
C. vesicata (Lindl.) Schltr.
quadrangular, surface wrinkled, red-brown. Leaf 5- 1 2 cm
Guin., S.L., Lib., lv.C., Ghana, S.Nig.
long, lanceolate - oblanceolate, acute, flat, dark green.
Middle parts of large branches. Full sun, bark - minor
Inflorescence shorter than the leaf. Flower very small.
humus deposits. Common. Stem short� woody. Leaves 1 5-
April-May. D.J. 756 (K, UPS).
40 x 1 -3 cm, linear or narrowly lanceolate, often much
curved, fleshy, v-shaped in cross-section, olive green or
Calyptrochilum Kraenzl.
pale yellow. Inflorescences numerous, longer than the
C. christyanum (Rchb. f.) Summerh. leaves, arching or pendulous, many flowered. Flower
Port.G., Guin., S.L., Lib., Iv.C., Mali, Ghana, N.Nig,, small, pale yellow. May-June. D.J. 5 30 (K), 430 (UPS).
S.Nig.
Basal parts of large branches. Full sun, bark. Common. Cyrtorchis Sch ltr.
Might grow to several meters in length. Stem woody. Leaf
C. arcuata Lindl. subsp. variabilis Summerh.
6- 1 2 cm, strap-shaped, bilobed, leathery. Inflorescence 3-
Guin., Lib., Iv.C., Ghana, D ah., N.Nig., S.Nig.
4 cm, many flowered, rather lax, rachis ± zigzag. Flower
Basal or middle parts of large branches. Open shade - full
white. Dec.-Jan. D.J. 7 1 9 (K, UPS).
sun, minor humus deposits or bark. Common. Erect, 30-
C. emarginatum (Sw.) Schltr. 60 cm. Stem long, woody, thick, covered with old leaf
Guin., S. L., Lib., Iv.C., Ghana, S.Nig., C am. sheaths. Leaf 1 0-20 9n long, oblong or slight_ly obovate,
On trunks of trees in villages. Rare in the primary forest. unequally bilobed at the apex. Inflorescence horizontal,
Full sun, bark. This species can grow to several meters' equaling the --length of the leaves. Flowers white, rather
length. Stem woody. Leaf 8- 1 5 x 2.5-5 cm, ovate - strap­ scattered on the raceme. Aug.-Sept. F.W.T.A., Vol. 3 : 1
shaped, bilobed, fleshy, green. Inflorescence short, many 1 968, recognizes three subspecies of C. arcuata. The sub­
flowered, very dense. Flower whitish - pale purple. May­ spp. variabilis and /eonensis would be impossible to
June. This plant might be confused with the closely related separate in the field. Specimens that could be any of the
Calyptrochilum christyanum, which, however, has a zig­ two subspp. have been recorded as C. arcuata in the field­
zag rachis. D.J. 829 (K). Fig. 34. work. D.J. 627 (K, UPS).

A cta phytogeogr. suec, 59


34 Dick Johansson

Fig. 36. Eurychone rothschildiana.

unequally bilobed, flat, stiff, leathery, dark green.


Inflorescence long, pendulous, many flowered. Flower
small, salmon-pink, semi-transparent. July-Aug. D .J. 5 69
(UPS). Fig. 3 5 .
D. densiflora (Summerh.) Summerh.
Lib. (new), C am .
O n branches o f small shrubs and trees, more seldom on
taller trees. Open shade, bark. Rare. Semi-pendulous. At­
tached to the substrate by a large number of roots at the
Fig. 3 5 . Diaphananthe bidens.
base of a 5-40 cm long stem. Aerial roots numerous along
the stem. Leaves 5 - 1 0 x 1 -2 cm, lanceolate - oblong, un­
equally bilobed. Inflorescences 2-5 cm. Flowers crowded,
C. aschersonii (Kraenzl.) Schltr. small, pale green. J uly-Aug. D.J. 5 7 1 (K, UPS).
S.L., Lib. (new), Ghana, S.Nig., C am. D. pellucida (Lindl.) Schltr.
Middle parts of large branches. Full sun, bark. Rather Guin., S.L., Lib., Iv.C., Ghana, S.Nig., C am.
common. Erect, 20-30 cm. Stem woody, thick, covered B asal parts of large branches. Open shade, bark or minor
with old leafsheaths. Leaves 6- 1 5 x 0.5- 1 .0 cm, almost humus deposits. Common. Stem short, woody, bearing
equally bilobed at the apex, very fleshy, rather stiff, strap­ the leaves in a dense tuft. Leaves arching, 1 5-70 cm,
shaped. Infl o rescences much shorter than the leaves. oblanceolate, unequally bilobed or almost entire at the
Flowers crowded on the raceme, white. Aug.-Sept. D.J. apex, fleshy. Inflorescences numerous, pendulous, many
6 1 3 (K), 5 76 (UPS). flowered. Flower pale yellow-pinkish. July-Aug. D.J. 6 1 2
C. monteiroae (Rchb. f.) Schltr. (UPS). Fig. 1 05 .
S.L., Lib., Iv.C . , Ghana, S.Nig., C am. D . rutila (Rchb. f.) Summerh.
Basal or middle parts of large branches. Open shade, Guin., S.L., Lib., Iv.C., Ghana, N.Nig., S.Nig., Cam.
minor humus deposits. Rare. Erect or pendulous, 30-60 On trunks or branches of small trees or shrubs. Full sun -
cm. Stem long, woody. Leaf 5-20 x 2-4 cm, oblong­ open shade, bark. Rather common. Growth habit as D.
elliptical or oblanceolate, unequally bilobed, often with bidens. Stem woody. Leaves 6- 1 5 cm, unequally and ob­
wrinkled margin. Inflorescence arching, exceeding the tusely bilobed. Inflorescence long, pendulous, with a large
leaves. Flowers scattered, white. Oct.-Nov. D.J. 674 number of small, translucent, pale yellow-purplish flowers.
(UPS). July-Aug. D.J. 560 (K, UPS).

Diaphananthe Sch ltr. Eurychone Schltr.


D. bidens (Sw.) Schltr. E. rothschildiana (O'Brien) Schltr.
Port.G., Guin., S.L., Lib., Iv.C., Ghana, S.Nig., N.Nig., Guin ., S.L., Lib. (new), Iv.C ., Ghana, S.Nig.
Cam. Trunks or branches of smaller trees or shrubs. Full sun -
On trunks, or in high forest in the middle parts of large open shade, bark. Rare. Stem very short, woody. Leaves
branches. Open shade - full sun. Bark or minor humus few, 7- 1 7 cm, broadly oblanceolate, unequally bilobed at
deposits. Common. Up to several meters length. Stem the apex, leathery. Inflorescence shorter than the leaves.
woody. The plant grows out away from the substratum. Flowers large, up to 5 cm in diam., 3-6, white and green.
Leaf 5- 1 4 cm, oblong-lanceolate to narrowly ovate, apex June-July. D.J. 8 5 3 (K, UPS). Fig. 36.

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 35

Habenaria Wi l l d .
H. leonensis Dur. & Schinz
Guin., S.L., Lib., Iv. C .
Basal parts of large trees. Full sun - open shade, humus
deposits. Rare. In the investigated area epiphytic, but in
the grasslands of Guineean Nimba common as terrestrial.
Erect, up to 30 cm high. Leaves 6- 1 2 x 1 -2.5 cm, narrowly
lanceolate. Inflorescence with a few flowers. Flower white.
July-Aug. In a sterile state hard to separate from other
Habenarias. D.J. 5 80 (K, UPS).
H. procera (Sw.) Lindl.
S.L., Lib., Iv.C., Ghana, S.Nig., Cam.
On trunks of trees, but more often of oil palms. Full sun -
open shade, humus deposits. Rare. Erect, up to 60 cm
high. Leaves lanceolate, 1 2-20 x 2-3 .5 cm. Inflorescence
densely flowered. Flowers white. July-Aug. In a sterile
state hard to separate from other Habenarias. D.J. 595
(K, UPS).

Liparis L. C. R i ch.
L. caillei Finet
Guin., S.L., Lib., N.Nig., C am.
Basal parts of trunks. Open shade, humus deposits. Rare.
Pendulous. A minute plant. Total size 3-6 cm, in­
florescence excluded. Pseudobulb 1 2-20 mm, globose,
green. Leaves 2-4 cm, ovate or lanceolate, green.
Inflorescence 1 4- 1 8 cm long, pendulous, rachis winged,
many flowered. Flower pale yellow. June-July. The leaf
and pseudobulb disappear among the debris in which the
Fig. 3 7. L iparis caillei. plant occurs. Even when flowering the inflorescence is
hard to observe. D.J. 5 2 1 (UPS). Fig. 3 7 .
L. nervosa (Thun.) Lindl.
Genyorchis S c h ltr. Syn. : L. guineensis Lindl., and L. rufina (Ridl.) Rchb. f.
G. pumila (Sw.) Schltr. ex Rolfe.
S.L., Lib., Iv.C., Ghana. Sen., Port.G., Guin., S.L., Lib., Iv.C., Ghana, D ah.,
Middle or outer parts of branches on small trees or N.Nig., S.Nig., Cam., F.Po.
shrubs. Full sun, bark. Rare. Erect, 2-3 cm. Pseudobulb In the forest only as an epiphyte, elsewhere often
0.6- 1 .5 cm, ovoid, quadrangular. Leaf 1 -2 cm, ligulate - terrestrial. In the lowest parts of trunks. Open shade - full
oblong, green. Inflorescence slender, 5 -6 cm. Flower sun, humus deposits. Rare. Erect, 1 5-25 cm. Stem with a
minute, Polystachya-like, white-pale green. March-April. ± swollen base. Leaves 5-20 cm, lanceolate, thin, with un­
This small inconspicuous plant is similar to a Bulbo­ dulating margins, yellow-green. Inflorescence erect, longer
phyllum sp. D.J. 485 (K, UPS). than the leaves. Flowers yellow. July-Aug. D.J. 5 5 6 (K,
UPS).
Graphorchis Tho u.
G. lurida (Sw.) 0 . Ktze.
Sen., Port.G., Guin., S.L., Lib., Iv.C., Ghana, N .Nig., Listrostachys Rchb. f.
S.Nig., C am. L. pertusa (Lindl.) Rchb. f.
Middle parts of large branches. Full sun, bark. Common. S.L., Lib., Iv.C., Ghana, S.Nig.
Erect, up to 50 cm. Pseudobulbs crowded, 5 - 1 0 cm, con­ Middle parts of large branches. Full sun - open shade,
ical - ovoid, yellow, longitudinally wrinkled, with black bark . C ommon. Erect, 1 5-25 cm. Leaves arranged in a
transverse stripes. Leaves 4-6, in a dense tuft at the tip of fan-shaped manner on a short, woody stem, 1 5 -20 x 1 -2
the mature pseudobulb, lanceolate, up to 40 cm, thin, cm, strap shaped, unequally bilobed at apex, leathery, v­
ribbed. Inflorescence from the base of the pseudobulb shaped in cross-section. Inflorescence 1 0-25 cm long,
appearing before the leaves, 1 5-50 cm, paniculate. Flower often longer than the leaves. Flowers small, white,
small, yellow and brown. Jan.-Feb. The yellow pseudo­ arranged in two dense rows along the rachis. Aug.-Sept.
bulb, surrounded by a dense tuft of white erect aerial This plant is of a shape similar to A i!"rangis laurentii and
roots, makes identification easy even when leaves or in­ Rangae'ris muscicola but has much more fleshy leaves.
florescences are missing. D J . 722 (UPS). Fig. 6 1 . D.J. 700 (K), 443 (UPS). Fig. 3 8 .

A cta phytogeogr. suec. 59


36 Dick Johansson

Fig. 38. Listrostachys pertusa.

Nephrangis S u m merh.
N. filiformis (Kraenzl.) Summerh.
Lib.
Middle or outer : parts of large branches. Open shade -
full sun, bark. Common. Erect when small, pendulous
with increasing size, 30-40 cm. Stem slender, often
branched. Leaves 2-8 cm long but only 1 -2 mm broad,
terete, much curved, acute. Inflorescence very short, few
flowered. Flower small, with a white bilobed lip. April­
May. Two other species have terete leaves : T. tridentata
and A ngraecum subulatum. The shape of the leaves
separates it from Tridactyle tridentata, and the pendulous
growth separates it from Angraecum subulatum. D.J. 529
(K, UPS). Fig. 5 9.
Fig. 3 9 . Plectrelminthus caudatus.
P/ectrelminthus R afi n .
Polystachya H ook.
P. caudatus (Lindl.) Summerh.
P. adansoniae Rchb. f.
Guin., S.L., Lib., Iv.C., Ghana, N.Nig., S.Nig . , Cam.
Guin., S.L., Lib., Iv.C ., Ghana, N . Nig., S.Nig., C am.
Middle or outer parts of large branches. Full sun, bark.
Middle or outer parts of large branches. Full sun, bark -
Rather common. Erect, stout, 20-30 cm . Stem short,
minor humus deposits. Common. Erect, up to 20 cm high.
woody. Leaves 1 0-30 cm, arranged in a fan-like manner,
Pseudobulb ellipsoid, 2-4 cm, light green. Leaves strap­
oblong or elliptical oblong, almost equally bilobed at apex.
shaped, 8- 1 6 cm, light green. Flowers small, yellow,
Inflorescence longer than the leaves, erect, rachis zigzag­
arranged in a spike. May-June. D.J . 4 8 5 , 8 5 8 (K), 437
shaped. Flowers large, white, spur long, spirally twisted.
(UPS).
Large capsules. April-May (Oct.-Nov.). D.J. 650 (UPS).
Fig. 39. P. affinis Lindl.
Guin., S.L., Lib., Iv.C., Ghana, S.Nig.
More or less restricted to trunks. Open shade - full sun.
Podangis Sch ltr.
P. dactyloceras (Rchb. f.) Schltr. Rather common. Growing in a vertical position on trunks.
Guin., S.L., Lib (new), Ghana, N .Nig., S.Nig., C am.
.
Pseudobulbs depressed, crowded, 2.5-4.5 cm in diameter,
Middle parts of large branches. Full sun, bark minor -
orbicular or broadly elliptical, green. Leaf 1 0-20 cm,
humus deposits. At altitudes above 800 m. Rare. Erect, 7- oblanceolate, dark green. Leaf and paniculate in­
12 cm. Leaves in a fan-like arrangement on top of a short florescence arching. Flower yellow with red markings.
woody stem, 6- 1 0 cm, lanceolate, fleshy, green. Jan-Feb. D.J. 450. (K, UPS).
Inflorescence arises below the leaves, and does not exceed P. dalzielii Summerh.
the leaves in length. Flowers 5 - 1 0, semi-transparent, Guin ., S.L., Lib., Jv.C.
white. Feb.-March. C an be mistaken for a Bolusiella Middle or outer parts of large branches. Full sun, minor
talbotii or a Rangaeris rhipsalisocia. The first, however, humus deposits. At altitudes over I 000 m . Common. A
has a long thin inflorescence that exceeds the length of the small plant. Pseudobulb around I cm high, conical, very
leaves, the latter has sickle-shaped leaves. D .J. 42 1 (UPS). hard, green. Leaf 2-6 cm, ovate-lanceolate, thin, purplish-

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 3 7

Fig. 40. Polystachya elastica.

green. Inflorescence up to 1 5 cm high but usually less, ris­


ing from a leafless pseudobulb. Flower white - lilac.Feb.­
March. Inconspicuous in a flowerless state. Wet roots in
cross-section purple-coloured. Polystachya pobeguinii, P.
elastica and P. saccata have leaves and pseudobulbs of
similar shape and sizes. D.J. 420 (K, UPS).
Fig. 4 1 . Polystachya laxiflora.
P. elastica Lindl.
S.L., Lib. Inflorescence either a spreading panicle with long slender
Outer parts of large branches. Full sun, bark. Rare. Erect. branches or, especially at higher altitudes, branches
Pseudobulbs crowded, 1 -2 cm 'high. Leaves narrowly reduced to short tufts. When flowering, a large drop of
lanceolate 2-5 cm long. Inflorescence 5 - 1 5 cm, born on a clear viscid fluid is often present below each bract. Flower
leafless pseudobulb. Flower rose· coloured. March-April. white - yellow. Feb.-March. This species turns black when
In a flowerless state similar to the more common P. dried and pressed. D.J. 6 1 4, 740 (K), 704 (UPS). Fig. 4 1 .
pobeguinii. DJ. 74 1 (K), 7 5 3 (UPS). Fig. 40.
P. leonensis Rchb. f.
P. galeata (Sw.) Rchb. f. Guin., S.L., Lib., Iv.C., C am.
Guin., S.L., Lib., lv.C., S.Nig. On trunks, or in the basal and middle parts of large
Middle and basal parts of large branches. Full sun - open branches. Open shade, minor humus deposits. At altitudes
shade, humus deposits. Common. Erect, 20-40 cm. above 1 000 m. Common. Erect, 1 2-20 cm. Pseudobulbs
Pseudobulbs crowded, 5 - 1 5 cm, narrowly cylindrical, crowded, 1 cm, conical, covered with sheets of old leaves.
green. Leaf 1 0-25 cm, at the apex of the pseudobulb, Leaves 8-20 cm, lanceolate or oblanceolate, thin, green,
elliptical - oblong, fleshy, stiff, persistant for many years. arranged in a fan-like manner. Inflorescence paniculate.
Inflorescence erect, shorter than the leaf and pseudobulb. Flowers yellow-green, lip white. Feb.-March. D.J. 423
Flower large compared to most Polystachya species, (K), 739 (UPS).
yellow-green, with red or purplish markings. March-April.
P. microbambusa Kraenzl.
DJ. 429 (K, UPS).
Guin ., S.L., Lib., Iv.C.
P. laxiflora Lindl. On trunks. Full sun, minor humus deposits. At 1 200- 1 300
Guin., S.L., Lib., Jv.C., Ghana, C am. m altitudes. Rare. Common on rocks in the grasslands of
Basal parts of large branches. Open shade, humus the Nimba Ridge in Guinea. Erect, 5- 1 5 cm. Pseudobulb
deposits. Common at altitudes above 1 000 m, lower down 1 - 1 . 5 cm, cylindrical, arising from upper part of the
uncommon. A large erect plant, up to 50 cm. Pseudobulb preceeding one. Leaves 2-4 cm, lanceolate. Flower small,
narrowly cylindrical. Leaves 8-20 cm long, oblanceolate. yellow. Jan.-Feb. D.J. 459 a (UPS).

A cta phytogeogr. suec. 59


38 Dick Johansson

Fig. 43. Polystachya saccata.

from a leafless pseudo bulb. Flowering scape 7- 1 5 cm


long. Flowers rose or lilac, yellow hairs on the lip. Feb.­
March. Inconspicuous in a flowerless state. The pseudo­
bulb and leaves similar to those of P. saccata and P.
Fig. 42. Polystachya polychaete. elastica. When flowering the rose flowers attracts atten­
tion . D .J. 45 7 (K), 788 (UPS). Fig. 50.
P. obanensis Rendle P. polychaete Kraenzl.
Lib., Iv.C . , S.Nig. S.L., Lib., Iv.C., Ghana, S.Nig., C am.
Basal parts of large branches. Open shade, humus Middle or basal parts of large branches. Full sun - open
deposits. Rare. Erect, 1 0-20 cm. Pseudobulbs com­ shade, humus deposits. Common. Erect, 1 5-30 cm high.
pressed, 2-4 cm, elliptical, shiny yellow, arranged in a Pseudobulbs 1 -2 cm in cross-sect., thin, stem like, slightly
more or less erect row, giving the impression that they are inflated at the base. Leaf 8-25 cm, lanceolate - elliptic.
climbing on top of each other. Leaf 5- 1 5 cm, oblong, Flowers small, arranged in a very dense simple raceme,
green. One single leaf at the apex of the pseudobulb and creamy white. Oct.-Nov. D.J. 708 (K), 4 1 8 (UPS). Fig.
two at the base. Inflorescence paniculate. Flower small, 42.
yellow. Scent strong, citrus-like. Sept.-Oct. D.J. 6 1 6 (K), P. puberula Lindl.
82 1 (UPS). Port.G., Guinea, S.L., Lib., lv.C.
P. paniculata (Sw.) Rolfe Middle and outer parts of large branches. Full sun, minor
Guin., S.L., Lib., GHANAf3 S.Nig., C am. humus deposits. Common. Erect, 30-40 cm high. Pseudo­
Middle parts of large branches. Full sun - open shade, bulbs crowded, 3-6 cm, conical, somewhat flattened,
bark. Common. Erect, 1 5 -30 cm. Pseudobulbs 5- 1 5 cm greenish-yellow with black transverse lines. Leaf 1 0-30
high, ellipsoid to long cylindrical, flattened, standing in a cm, oblanceolate - linear, thin, light green. Flowers small,
dense row, with the flattened sides towards each other, yellow, in a rather large and loose panicle. May-June. D.J.
green with minute purple dots. Leaves 8-20 cm, oblong ­ 5 1 1 (K, UPS).
elliptical, bluish green with a waxy layer above, purplish­
P. ramulosa Lindl.
green below. Flower small, orange with red markings. S.L., Lib., Ghana, S.Nig.
Oct.-Nov. D.J. 633 (K), 632 (UPS). On the lowest parts of trunks. Open shade, humus
P. pobeguinii (Finet) Rolfe deposits. Rare. Erect, 5- 1 0 cm. Pseudobulbs slightly
Guin., S.L., Lib., Iv.C. swollen at the base. Leaf 4-8 cm, oblong - elliptical, dark
Middle parts of large branches. Open shade, humus green-somewhat purplish, often damaged by insects.
deposits. Rare. A small plant. Pseudobulbs crowded, 0.5 - 1 Inflorescence much higher than the leaves. Flowers in a
cm, conical - ovoid, green and hard. Leaf 4-8 cm, oblong­ loose panicle, white, tinged with pink. June-July. D.J. 5 5 2
lanceolate, green - purplish green. Inflorescence arising ( K , UPS).

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 39

Fig. 45. Rangai/ris brachyceras.


falcate. Inflorescence paniculate. Flower minute, pale
yellow. July-Aug. D.J. 6 5 7 (UPS).
P. tessallata Lindl.
Guin., S.L., Lib., lv.C., Ghana, N.Nig., S.Nig., Cam.
Basal and middle parts of large branches. A tolerant
species. Full sun - open shade, bark - large humus
Fig. 44. Polystachya tessallata and Platycerium stemaria.
deposits. Common. Erect, up to 50 cm. Pseudobulb 1 cm,
conical brown-yellow. Leaves 1 0-25 cm, oblanceolate or
P. rhodoptera Rchb. f. oblong-elliptic, dark green, often somewhat purplish.
S.L., Lib., Iv.C., S.Nig. Inflorescences much longer than the leaves, somewhat
Almost exclusively found on the lowest parts of trunks. branched, branches secund. Flower small, greenish-yellow
Open shade, h umus deposits. Rare. Erect, 20-40 cm high. with purple markings. May-June. D .J. 522 (K, UPS). Fig.
Leaf 7- 1 2 x 1 cm, narrowly lanceolate. Inflorescence 44.
somewhat branched. Flower pale rose. Sept.-Oct. Leaves
held in a more or less vertical position. Empty capsules, Rangae.ris S u m m erh.
1 5 -20 mm long, persist on the inflorescence for a year. R. brachyceras (Summerh.) Summerh.
D.J. 447 (K), 462 (UPS). Guin., S.L., Lib. (new), lv.C., S.Nig.
P. saccata (Finet) Rolfe Middle or outer parts of large branches. Full sun, minor
Port.G., Guin., Lib. (new), N.Nig. humus deposits. At altitudes above 1 200 m. Rare. Erect,
This is one of the few species that frequently occur on the 10-20 cm. Stem woody, short. Leaves in a fan-like
upper parts of trunks. Full sun - open shade, bark. Rather arrangement, 6- 1 6 x 1 -2 cm, ligulate or slightly wider in
common. A minute plant. Pseudobulb 0.5 cm, conical, the upper part, v-shaped in cross-section. Inflorescence
green. Leaf 2-5 cm, oblong - lanceolate, green - purplish longer than the leaves, slender. Flower pale, yellowish
green. Inflorescence 3-8 cm, arising from the leafless white. June-July. D.J. 594 (K). Fig. 45.
pseudobulb. Flower white and rose. Feb.-March. In­ R. muscicola (Rchb. f.) Summerh.
conspicuous in a flowerless state. Leaves and pseudo­ Guin., S.L., Lib., lv.C., S.Nig., C am.
bulbs similar to those of P. pobeguinii and P. elastica. D.J. Middle parts of large branches. Full sun, bark. Common.
460 (K), 456 (UPS). Fig. 43. Erect, 8- 1 5 cm. Stem short, woody. Leaves 1 0-20 x 1 - 2
P. subulata Finet c m , v-shaped in cross-section. Inflorescence straight,
Guin., S.L., Lib. (new), lv.C., N.Nig. longer than the leaves. Flower white, star-shaped, with a
Middle or basal parts of large branches. Open shade, long spur. C apsules 3-5 cm long and narrow. June-July.
minor humus deposits. Rare. Erect, 1 2-25 cm high. This plant might be confused with A erangis laurentii,
Pseudobulb 5 -8 cm high but less than 0.5 cm in cross­ which, however, has a zigzag inflorescence. D.J. 540 (K,
section, narrowly cylindrical, green. Leaf 6- 1 0 cm, ligulate UPS). Fig. 59.
or oblanceolate, dark purplish-green. Inflorescence simple R. rhipsalisocia (Rchb. f.) Summerh.
or seldom somewhat branched. Flower pale yellow-green Sen., Guin., S.L. , Lib., Iv.C., Ghana, S.Nig., C am.
with purple dots. Capsules large. June-July. D.J. 859 (K, Middle or outer parts of large branches. Full sun, bark.
UPS). Rare. Erect, 8- 1 5 cm. Stem short, woody. Leaves
P. tenuissima Kraenzl. arranged in a fan-shaped manner, 5- 1 0 cm, fleshy,
Lib. (new), Iv.C., Ghana. flattened, much curved, waxy, bluish green. Inflorescences
Middle parts of large branches. Full sun - open shade, longer than the leaves, stout. Flower white, rather small.
bark - minor humus deposits. Rare. Erect, 1 0-20 cm. Jan.-Feb. May resemble Podangis dactyloceras or
Pseudobulbs 2-6 cm, terete, green. Leaves terete, slightly Bolusiella talbotii. D.J. 724 (K, UPS).

A cta phytogeogr. suec. 59


40 Dick Johansson

... · .•
· �.
9
·•·.. ·· . .··:. < .··. ·
F� . . ..
Fig. 4 7. Tridactyle bicaudata.

i !� / / , �
: •... ·, :

Tridactyle Sch ltr.


T. anthomaniaca (Rchb. f.) Summerh.

(j CJ D�
S.L., Lib., Iv.C., S.Nig., C am .
Middle o r outer parts o f large branches. Full sun, bark.
Common. Climbing or hanging. A rather large plant,
branching in many directions. Stem stout, woody. Leaf 2-
8 cm, oblong or elliptical oblong, broadly and obtusely
Fig. 46. R hipidog/ossum paucifolium. (A) Flowering bilobed at the apex, flat, somewhat fleshy but stiff,
plant, (B) Flower with lip removed, front view, (C) Dorsal leathery. Inflorescences less than 0.5 cm, carrying 1 -3
sepal, (D) Lateral sepal, (E) Petal . (F) Lip, (G) Column flowers. Flower small, yellow-brown. Oct.-Nov. D.J. 676
with anther cap and pollinia removed, side view, (H) (K, UPS).
Rostellum, front view, (I) Pollinium. (From Johansson T. armeniaca (Lindl.) Schltr.
1 974.) Guin., S.L., Lib., Ghana.
Basal parts of large branches. Open sh ade, bark - minor
humus deposits. Common. Erect, often branching and
creating dense 'stands'. Stem long, woody in the basal
part. Leaf 5- 1 2 x 1 .0- 1 .5 cm, oblong - lanceolate, un­
equally bilobed at apex, flat, rather thin. Inflorescence
Rhipidoglossum Sch ltr.
mostly below the leaves. Flowers small, yellow or apricot.
R. paucifolium Dick Johanss.
Feb.-March. T. fusifera Mansf. and T. tridactylites are
Lib. (New).
vegetatively similar to this species. The first species has
B asal parts of large trees and on small trees and shrubs.
not been observed although F.W.T.A. 1 968 3 : 1 , gives two
Open shade, bark. Rare. Stem very short, with a very
records from the Nimba area. The latter species only oc­
large root system. Leaves few, 4-6 x 1 .5-2.0 cm, oblong­
curs at altitudes above 1 000 m where T. armeniaca is
lanceolate. Inflorescences pendulous. Flowers crowded,
rare. D.J. 452 (K , UPS).
semi-transparent, pale green. Aug.-Sept. This species is
only known from the Nimba area (Johansson 1 974). D.J. T. bicaudata (Lindl.) Schltr.
5 72 (K, UPS). Fig. 46. S.L., Lib., Iv.C., Ghana, N.Nig.
Middle or basal parts of large branches. Open shade, bark
Stolzia Schltr. - minor humus deposits. Rather common. Erect. 30-50
S. repens (Rolfe) Summerh. cm. Stem long, woody. Leaf 1 0-20 x 1 -2 cm, narrowly
Lib. (new), S.Nig., C am. ligulate, unequally bilobed. Inflorescence from the leaf ax­
Middle or outer parts of large branches. Open shade - full ils or below the leaves. Flower pale yellow-orange. Oct.­
sun, minor humus deposits. R are. Prostrate. Pseudobulb Nov. Vegetatively similar to T. armeniaca. It is easy to
clubshaped, looking like a swollen part of the rhizome. identify by the very strong aromatic, somewhat vanilla­
Leaves elliptical or nearly orbicular, two at each pseudo­ like scent that is emitted from the roots, particularly when
bulb. Never seen flowering in the Nimba area. F.W.T.A. the plant is pulled off it's substrate. D.J. 684 (UPS). Fig.
1 968, 3 : 1 describes the penduncle as stout .. very short, 47.
flower small,orange, ochra-coloured or reddish. Its creep­ T. crassifolia Summerh.
ing habit makes it hard to detect. It can be mistaken for a Lib. (new), Ghana.
Peperomia rotundifolia, since both grow in similar On the trunk or anywhere on the large branches. Full sun,
h abitats. D.J. 604 (UPS). bark. Rare. Erect, or sometimes hanging, with a stout

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 41

woody stem. Leaf 3-8 x 0.6- 1 .0 cm, unequally bilobed, v­ Vanilla M i l l .


shaped in cross-section, the upper surface concave with a
V. crenulata Rolfe
narrow groove along the centre, the lower surface convex,
Guin., S.L., Lib., Iv.C., Ghana, S.Nig.
very fleshy, stiff. Surface waxy, bluish green.
Climbs the basal parts of the trunks. Open shade, large
Inflorescences very short. Flowers greenish yellow.
humus deposits. Rare. Climber. Sizes up to three meters in
March-April. D.J. 695 (K).
length have been observed. Stem slender. Leaf 5- 1 0 cm,
T. tridactylites (Rolfe) Schltr. oblong - elliptical, waxy, yellow-green. Inflorescence a
Guin., S.L., Lib. (new), Iv.C., N.Nig., S.Nig., Cam., F.Po. long hanging raceme. Flowers large, white or yellow with
Basal or middle parts of large branches. Open shade, purple markings on the lip. July-Aug. Jn a flowerless state
minor humus deposits. At altitudes above 1 000 m. Com­ this species is difficult to differentiate from Vanilla
mon. Erect, often branching, forming large masses of africana Lindl. D.J. 723 (K, UPS). Fig. 5 1 .
growths. Stem woody. Leaves strap-shaped, distichous,
deeply notched. Inflorescences short and few-flowered, in
-
the lower le afless part of the stem. Flower yellow. Jan.­
Feb. In vegetative parts similar to T. armeniaca, but has,
Other vascular epiphytes
however, never been observed at altitudes below 1 000 m.
Begonia L. ( B e gon iaceae)
The possible confusion of the two species is therefore
B. mannii Hook.
limited to high altitudes. In observations of distribution (>
S.L., Lib . . S.Nig., C am., F.Po.
1 000 m), T. tridactylites is considered the only species
Basal part of large branches. Open shade, humus deposits.
present. D.J. 726 (UPS). Fig. 86.
Rare. D.J. 5 5 1 (K, UPS).
T. tridentata (Harv.) Schltr.
B. polygonoides Hook. f.
Guin., S.L., Lib. (new), Ghana.
Guin., Lib., Iv.C., S.Nig., C am.
Middle or outer parts of large branches. Full sun, bark.
Basal parts of large branches. Heavy - open shade, humus
Rather common. Pendulous, 30-50 cm. Leaves 6- 1 0 x 0. 1 -
deposits. Rather common. D.J. 545 (K) 735 (UPS).
0.3 c m i n cross-section, terete, distichous, with a shallow
groove running along their entire length. Inflorescence B. rubro-marginata Gilg.
short, few-flowered. Flower pale yellowish. Aug.-Sept. Lib. (new), Cam.
This species can be confused with Nephrangis filiformis Basal or middle parts of large branches. Open shade - full
which, however, has leaves that are shorter and markedly sun, humus deposits. Common. D.J. 736 (K), 733 (UPS).
falcate. D.J. 59 1 (K, UPS). Fig. 5 9 . Fig. 65.

Fig. 48. Left : A strangling fig.


(Ficus anomani) established as an
epiphyte in the crown _ of a palm.
Later it will send roots to the
ground and continue as an in­
dependent plant. Nimba Range
600 m. Right : A strangling fig
when the roots have reached the
ground. The original phorophyte
is dead and has left a hollow
formed by the decayed trunk.

A cta phytogeogr. suec. 59


42 Dick Johansson

Ca/voa H ook. f. ( M elastomataceae) Table 7. Facultative epiphytes and filmy ferns in the Nimba
C. monticola A. C hev. ex Hutch. & Dalz. area.
Guin, S.L., Lib., Iv.C .
The lowest parts o f the trunks. Heavy shade, humus FAC ULTATI VE EPIPHYTES
deposits. Common. D .J. 785 (K, UPS).
Pteridophytes
C. trochainii Jac.-Fei. Asplenium variable Hook var. variable
Guin., Lib. Bolbitis fluviatilis (Hook.) C hing
Basal parts of trunks or middle parts of large branches. Bolbitis salicina (Hook.) Ching
Open shade, minor humus deposits. Common. D.J. 5 73 Lygodium smithianum Presl ex Kuhn
(UPS). Pityrogramma calomelanos (L.) Link
Selaginella blepharophylla Alston
Medinilla Gaud. ( M elasto m ataceae) Selaginella molleri Hieron.
M. mannii Hook. f. Selaginella myosurus (Sw.) Alston
Lib., F.Po. Selaginella versicolor Spring
Middle or basal parts of large branches. Open shade, Selagim:lla zechii Hieron.
humus deposits. At altitudes above 1 000 m, often solitary.
Other vascular epiphytes
At lower altitudes, often associated with Drynaria lauren­
Araceae
tii. Common. D.J. 725 (UPS).
Cersestis afzelii Schott
Culcasia angolensis Welw. ex Schott
Peperomia R u izd & Pav. ( P i pera cea e)
Culcasia liberica N .E.Br.
P. fernandopoiana C . DC.
Raphidophora africana N.E.Br.
Guin., Lib., Iv.C., S.Nig., Cam., F.Po.
In all sections of the phorophyte in a wide range of Begoniaceae
habitats. Common. D.J. 475 (UPS). Begonia oxyloba Welw. ex Hook.
Begonia quadrialata Warb.
P. molleri C. DC.
Lib., Ghana, Togo, S.Nig., C am., F.Po. Melastomataceae
Basal parts of trunks. In heavy shade, minor humus Tristemma incompletum Benth.
deposits. Rare. D.J. 882 (K).
Moraceae
P. rotundifolia (L) H.B. &K.
Ficus anomani Hutch. (Fig. 48)
S.L., Lib., Iv.C., Ghana, Togo, C am.
Ficus kamerunensis Warb. ex Mildbr. & Burret.
Basal or middle parts of large branches. Open shade,
Ficus leprieurii Miq.
humus deposits. Rather common. D.J. 498 (UPS). Fig.
Ficus sagittifolia Warb. ex Mildbr. & Burret.
59.
Urticaceae
Preussiel/a G i lg. ( M elasto m ataceae) Urera oblongifolia Benth.
P. chevalieri Jac.-Fel. Urera rigida (Benth.) Keay
Guin ., Lib., Iv.C. FILMY FERNS
Basal parts of large branches. Open shade, humus Hymenophyllum kuhnii C.Chr.
deposits. Rather common. D .J. 7 8 5 (K).
Trichomanes africanum Christ
P. kamerunensis Gilg. Trichomanes chamaedrys Tanton
Lib., C am . Trichomanes erosum Willd.
Basal parts o f large branches. Open shade, humus Trichomanes guineense Afzel. ex Sw.
deposits. At altitudes above 1 000 m. Rare. D.J. 824 (K). Trichomanes mannii Hook.

Remusatia Schott. (Araceae)


R. vivipara (Roxb.) Schott.
Guin., S.L., Lib. , Iv.C., C am.
Basal parts of large branches or trunks. Open shade, large
humus deposits. Rare. D.J. 49 1 (K).

Rhipsalis G ae rtn. {Cactaceae)


R. baccifera (J. Mill.) W.T. Steam
S.L., Lib., Iv.C., Ghana, S.Nig., C am., F.Po.
Middle or outer parts of branches. Full sun - open shade,
bark or minor humus deposits. Common. D.J. 793 (UPS).
Fig. 59.

Acta phytogeogr. suec. 59


Ill. Biology of vascular epiphytes

Reproduction biology following moisture that actually induces flowering."


(Cf. Arditti 1 966.)
Flowering periods of orchids The genera Bulbophyllum and Polystachya, ac­
The flowering periods are of particular interest since counting for about half of the number of species,
they may indicate an interbreeding barrier between show a marked difference in their blooming peaks
closely allied species. (Fig. 49). The Polystachyas have the most frequent
The general pattern in the flowering of the orchids blooming late in the dry season or early in the rainy
is presented in two-month-periods for the year 1 969 season, while the Bulbophyllums have a very
(Fig. 49). Two flowering peaks are easily noticed. In pronounced peak during the end of the rainy season.
his studies of the Nigerian rain forests Richards Most of the Bulbophyllum species flower on
( 1 939) found this bimodal culmination of flowering mature shoots, e.g. at the end of the rainy season.
to be a general rule for the vegetation. A similar Bulbophyllum winkleri and B. oreonastes are two ex­
pattern in flowering times of West African orchids is ceptions. They flower on new growths early in the
reported by Sanford ( 1 97 1 : 1 66) : "Casually, the two growing season.
blooming peaks seem to follow the long and short Among the Polystachyas that flower during the
dry seasons. Such coincidence has probably lead to dry season there are four species bearing their
the belief that dry seasons trigger flowering in trees flowers on leafless pseudobulbs (Curtis 1 943), viz. P.
(see Ashton 1 969) . . . , however, there is no evidence dalzielii, P. elastica, P. pobeguinii, and P. saccata.
for orchids that it is either the dry season or the At the end of the rainy season in Nov.-Dec. the

N o. of
species
20
O rc h i d s
R a i nfa l l Yekepa
5 1 0 m 1 969

, .,. .. ... . ... ..

_ ... � � "'
_ ..... - -

JA N. N O V.

FE B . DEC.

16 B u l bophyl l u m spp.
Polystachya spp.

Fig. 49. Flowering periods. Upper


10
figure : Total number of flowering
orchids in two-month-periods dur­
ing a year in relation to the rain
fall . Lower figure : Number of
flowering species of the genera ,., .,.
...... - - _ _ ,

..... .... .....

Bulbophyllum and Polystachya.


A cta phytogeogr. suec. 59
44 Dick Johansson

Fig. 5 1 . Vanilla crenulata.

genus Bulbophyllum which flower simultaneously


favour the possibilities of hybridization.
There are also examples of closely related species
with different flowering times e.g. Calyptrochilum
emarginatum and C. christyanum. These grow in the
same habitat under non-limiting conditions (as in the
Nimba area), but there are considerable intervals
between their flowering periods.

Observations of orchid pollination


Since in sexual flowering plants pollination is a pre­
requisite for the seed formation, any limitation in this
process will effect the total \ number of seeds.
However, among the angiospermous epiphytes of
Nimba all species seem to be pollinated with one ex­
ception. Vanilla crenulata has never been observed
with capsules. Two plants of this species which were
'
observed during three years produced a large
number of flowers each year (Fig. 5 1) but no signs of
Fig. 50. Polystachya pobeguin �i. pollination were noted. The flowers simply wilted
away after the anthesis. No actual observation of the
pollinators of any Vanilla species seems to exist,
leaves are shed, and the plant seemingly disappears even if various insects and hummingbirds are said to
since the pseudobulbs are very small (less than one participate in the process (Childers et al. 1 9 59 :483).
centimeter) and the root system is hidden in cracks The possibility of the Vanilla species being
in the bark or covered by debris, lichens and mosses. autogamous has also been considered (Van der Pijl
Thus when the inflorescence appears one gets the im­
pression that it is growing straight out of the bark Table 8. Insects caught on orchid flowers cultivated in Yekepa.
(Fig. 50).
Orchid Insects frequently visiting
It is evident that the species of some genera flower
the flower
during a limited period of the year, while others have
a more prolonged flowering time. The six species of Bulbophyllum buntingii Syntomis sp.
Bulbophyllum calamarium Trigona sp.
the genus A ngraecum flower in the middle of the
Bulbophyllum schinzianum Polybiodes tabida (Tab.),
rainy season. In July all of them flower Polistes marginalis (Tab.)
simultaneously. In August the four Diaphananthe Cyrtorchis aschersonii Bradylema sp.,
species have a similar synchronous blooming time. Gabonia nudeus (Weise)
The genus Tridactyle, on the other hand, includes Diaph ananthe pellucida Euchromia lethe (F.)
Polystachya pob(!guinii Trigona sp.
species that flower during every month of the year.
Tridactyle tridentata Gabonia nudeus (Weise)
The many closely allied species especially in the

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 45

Fig. 5 2. Euchromia lethe.

& Dodson 1 9 6 9 : 1 25). Whether insufficient pollina­


tion is a limiting factor to this species is hard to tell
from the few observations available.
A list of some insect species caught visiting orchid
flowers at Yekepa is given in Table 8. If all of these
insects are effective in the pollination process is
doubtful, but two of them, Euchromia lethe (Fig. 5 2)
and Polybiodes tabida were observed with pollinia
attached to them from the orchids visited.

Diaspore size and dispersal


Judged by the types of diaspores (Table 9) among
the epiphytes one may assume that the dispersal is
mostly done by wind (cf. Oliver 1 930, Curtis 1 9 3 3).
The seeds of the Peperomias and Rhipsalis bac­
cifera, which have fleshy fruits said to be edible Fig. 5 3 . C apsule, flower and bud of Bulbophyllum schin­
zianum. Several orchids open their flowers singularly or a
(Moreau 1 93 5 :20), are probably dispersed by
few at a time. This may favour the possibility of pollina­
animals (Fig. 5 5). tion by insects.
Orchid seeds are usually small, reduced, and very
between 1 00-250 �m in size, and among the
numerous. One seed weighs from 0.3 to 1 4 �g (Ar­
Polystachya species between 200-400 �m. Par­
ditti 1 96 7 :3). Ames ( 1 946) estim ated the number of
ticularly large seeds are noticed for L iparis nervosa
seeds to 3, 770,000 in one single capsule of Cynoches
(500-660 �m) a species of the lower parts of the
chlorochilon. The sizes of the orchid seeds of the
trunks. Fern spores are in general considerably
species in the Nimba area varies generally between
I 00-400 �m. Particularly small seeds were found in
smaller than orchid seeds, e.g. Drynaria laurentii (40
�m) and Asplenium barteri (40-50 �m).
A ngraecum classensii ( 1 5 -50 �m) and A .
podochiloides (20-50 �m). Bulbophyllum seeds are
Germination and establishment
After the diaspore has reached the phorophyte the
Table 9. Diaspores of the epiphytes in the Nimba area.
germination and further development may be in­
Diaspores No. of species C ategories hibited for several reasons. The dryness of the sub­
-
Total % strate can be assumed to be a usual reason for such a
Minute spores 39 25.5 Pteridophytes
failure. However, there are some reports that other
Minute seeds 1 09 7 1 .2 Orchidaceae, Melasto- properties of the bark may prevent germination. The
mataceae, Begoniaceae influence of the substrate on the epiphyte flora will
Seeds with viscid 0.7 C actaceae be discussed more specifically in chapter V. A list of
coating
factors that might influence the germination of
Fleshy fruits 4 2.6 Araceae, Piperaceae
orchid seeds has been presented by Arditti ( 1 967).

A cta phytogeogr. suec. 59


46 Dick Johansson

Fig. 54. Bulbophyllum phaeopogon. The hairy lip is mo­


bile in a vertical direction and the slightest breeze will in­
itiate a movement.
Fig. 5 5 . Rhipsalis baccifera with berries, each holding a
large number of small seeds.
The seeds of the orchids are dependent on certain
mycorrhizal fungi for their development (Heagarty hold on the substratum, the likelihood of dispersal
1 9 5 5 , Arditti et al. 1 9 72), although it is possible to after the entrance of the mycorrhizal fungi would be
grow them asymbiotically in vitro (Knudson 1 922, diminished and opportunities for successful dispersal
Burgeff 1 93 6, 1 95 9). Thus, the presence of these severely limited." (Ames 1 922: 23 2.)
fungi on the phorophyte is a prerequisite for the es­ The disadvantage for the seed in becoming a little
tablishment of orchids (Withner 1 9 5 9 : 3 1 7). heavier with their fungal freight is regarded to be of
All the fungi belong to the Hymenomycetes. The little importance. More critical i s desiccation, but ex­
majority are members of the imperfect genus Rhizoc­ periments with Goodyera pubescens R. Br. showed
tonia, and as such are probably haploid mycelia of that the seeds in their early stages of germination
Corticiae (Burgeff 1 95 9 : 3 76). easily could withstand a journey of twenty to thirty
Rhizoctonias are present in all green orchids. No miles without lethal desiccation.
specific relation between fungus species and orchid
host seems to exist (Curtis 1 9 3 9). The possibility for
orchid seeds to be inoculated with mycorrhizal fungi Life forms
in the early stages of germination at a temporary
resting place, e.g. at the root system of the mother The many strange life forms among epiphytes at­
plant, and then blown to a final resting place, has tracted attention early (Goebel 1 88 8 , Schimper
been suggested (Ames 1 922). Ames considers the 1 888, Karsten 1 89 5 , 1 925). A life form system has
slow germination of the orchid seeds to be of advan­ been created by Hosokawa ( I 943 , 1 949, 1 9 5 5). Ob­
tage, increasing their possibilities to get in contact jections towards this system have been raised by
with the fungi. "If orchid seeds germinated quickly Holtum ( 1 960). In a dense forest a general pattern
and gained weight rapidly, if the embryo immediate­ can be observed. The epiphytes of the lowest parts of
ly put forth roots and in a brief time secured a firm the trees are living under rather constant conditions

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 4 7

Fig. 5 6 . A splenium nidus i s an exceptionally large


epiphyte. The fronds may reach a length of two metres�
Amani 950 m, Usambara Mts, Tanzania.
Fig. 5 7. Xeromorphic epiphytes on the outer part of a
branch of Lophira alata. Nimba Range 600 m. (A)
of humidity. There is little need for adaptations Bulbophyllum schimperanum, (B) Nephrangis filiformis,
against desiccation (Fig. 5 6). (C) Chamaeangis vesicata, (D) Tridactyle anthomaniaca,
The filmy ferns, Trichomanes and Hymeno­ (E) Bulbophyllum linderi.
phyllum that are common on the lowest parts of
trunks, represent the most drought-sensitive epi­ mm ( 1 08) was observed for Plectrelminthus
phytes. Higher up on the trees the epiphytes with caudatus, which grows in exposed habitats, while the
somewhat fleshy leaves establish themselves. A lowest number was found in Habenaria leonensis
reduction of transpiration surface compared to the from a more protected habitat. However, Ae'rangis
volume of the plant can also be observed. In more laurentii, which occurs in the same habitat as Plec­
exposed habitats, the size and total number of leaves trelminthus caudatus, has only about half the
are reduced (Fig. 5 7). number of stomata per sq. mm of the latter. From
The common Bulbophyllum species have only one this scanty material it is hard to draw any conclu­
or two leaves with the basal part of the leaf sion, but the number of stomata may well prove to
developed into a pseudobulb (Fig. 5 8). Xeromorphic be of minor importance as an environmental adapta­
structures e.g. succulent and terete leaves are com­ tion.
mon in the most exposed habitats (Fig. 5 9).
The number of stomata and their distribution on Drought tolerant and drought avoidant epiphytes
the leaves m ay also be of adaptative importance. A One m ay divide the epiphytes into two major groups,
study of several orchids (Table 1 2) revealed that the the drought tolerant and the drought avoidant (cf.
stomata in general are located on the lower side of J arvis & J arvis 1 96 3). The shape and texture of the
the leaves. The highest number of stomata per sq. leaves often indicate to which group a certain species

A cta phytogeogr. suec. 59


48 Dick Johansson

Sympodial and monopodial growth


The two largest orchid genera, Bulbophyllum and
Polystachya, have sympodial growth. This means
that the plants develop a shoot which is limited in
apical growth. Thus the development of pseudobulbs
and leaves takes place during the wet season, each
shoot terminating in an inflorescence. The following
wet season a new shoot will appear. This growth
form is generally considered to be favourable for
epiphytes (Holtum 1 960) apart from giving them an
indefinite length of life. (The Botanical Gardens of
Copenhagen have a living plant of Oncidium
sphacelatum from Mexico, which has been in
cultivation for 1 27 years, Anonymus : 1 968.)
One should expect that among the orchids with
sympodial growth there would be no need of ad­
vanced adaptations towards desiccation. It is
therefore somewhat surprising to find advanced
adaptation against drought among e.g. the
Bulbophyllum species. The Polystachyas give ex­
amples of leaves that have no adaptations and are
deciduous, e.g . P. puberula, as well as succulent
.

leaves that remain on the pseudobulb during the dry


season, e.g. P. tenuissima. Whether or not these suc­
culent leaves remaining on the plant during the dry
season are an asset or liability in the· survival of the
new shoot is hard to tell. They naturally lose water
Fig. 5 8 . Bulbophyllum congolanum. through transpiration, but their photosynthetic ac­
tivity is difficult to estimate. Thus the adaptations
against drought among the sympodials could be con­
belongs. The leaves of the drought avoidant
sidered as much an adjustment to drought periods of
epiphytes, being deciduous, are not adapted to sur­
varying length during the wet season as an adjust­
vive the dry season. Examples of such plants are the
ment to the dry season proper.
ferns Davallia chaerophylloides, Drynaria laurentii,
The orchids with monopodia! growth, where each
and Nephrolepis undulata. Among the orchids the
stem has an indefinite apical growth, are subject to a
species belonging to the genera Brachycorythis,
more serious threat to desiccation, since they keep
Graphorchis, Habenaria, and Liparis are all
their leaves during the dry season. Holtum ( 1 960)
deciduous, as well as many Polystachyas. Begonia
states that monopodia! orchids are therefore found
rubro-marginata and Remusatia vivipara also ex­
mainly in regions which have no long dry season.
hibit this adaptation.
In the Nimba area there are 60 species of sym­
The drought tolerant epiphytes often have a
podial orchids compared to 4 1 monopodia} species.
reduced number of leaves, e.g. Eurychone
rothschildiana; · succulent leaves, e.g. Microsorium
punctatum, A ngraecum distichum, Peperomia rotun­
Water e:conomy
difolia; or terete leaves, e.g. A ngraecum subulatum,
Nephrangis filiformis, Tridactyle tridentata. Water uptake
The combination of few and succulent leaves are Most of the substrates used by epiphytes have a
exemplified by Bolusiella talbotii and Rangaeris limited storage capacity for water. This means that
rhipsalisocia. Rhipsalis baccifera represents the stem the epiphytes themselves must be able to effect a
succulents. rapid water uptake during the limited period when
A cta phytogeogr. suec. 59
Ecology of vascular epiphytes in West African rain forest 49

Fig. 59. Life forms of some epiphytes. (A) 1 . Rangae"ris Rhipsalis baccifera, (E) Peperomia rotundifolia, (F)
muscicola, 2. Nephrangis filiformis, 3. Tridactyle triden­ Polystachya a/finis.
tata, (B) A ngraecum podochiloides, (C) A . distichum, (D)

A cta phytogeogr. suec. 59


50 Dick Johansson

Fig. 6 1 . Graphorchis lurida, aerial roots in the fore


ground. Yekepa 500 m.

velamen. Went ( 1 940 :9 1) claims that the velamen


h as an important role in the uptake of water and
Fig. 60. Roots of Ae'rangis laurentii. mineral nutrients : "Es ist nun klar, dass das Velamen
sich direkt mit dem ersten mineralreichen
the substrate is wet, or collect the water in small Sickerwasser fiillt . . . " Dycus and Knudson ( 1 9 5 7)
pools between the leaves (tank or cistern epiphytes found that the uptake of water was approximately
among many bromeliads of the Americas). Water equal to the rate of water lost by evaporation in the
absorbed in humus may also serve as a water supply aerial roots of orchids. The absorption was
(Fig. 1 1 3). equivalent to that required to saturate the velamen
The osmotic potential of epiphytes seem to be high (cf. Wallach 1 93 9). Condensation of water vapor by
(Harris 1 9 1 8, 1 934, Blum 1 93 3). W aiter ( 1 97 1 : 1 3 6- living roots could not be proved. If condensation did
1 3 7) found that orchids with tubers had a value of occur, it was so limited in comparison to the
-2.0 bar. Somewhat · lower values (-4 . 7 to -6.7 evaporation that it could not be detected. The prin­
bar), were found in orchids without tubers and the cipal roles of the velamen in free aerial roots were
lowest values were found among epiphytic ferns con sidered to be a mechanical protection and the
(-9. 1 to -33.7 bar). prevention of an excess loss of water from the cortex.
Non-vascular epiphytes such as mosses and Goebel ( 1 922) reports that roots of Epidendron noc­
lichens are well. known for their rapid water uptake. turnum that had their velamen carefully removed,
Indirectly their moisture holding capacity may be of lost 20 % of their water content in 24 hrs as com­
importance for the vascular epiphytes. The filmy pared to a 7 % loss by roots with their velamen un­
ferns have a water uptake similar to the mosses. touched.
The roots of epiphytic orchids exhibit several However, when an aerial root comes in contact
specialized features in their morphology and with a solid surface the velamen cells at the point of
anatomy. The aerial roots are surrounded with a attachment become markedly modified. Certain salts
white, papery or spongy layer of dead cells called the are absorbed by the modified part of the root and

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 51

(Fig. 62). These erect roots which are covered with


velamen form a dense 'tussock'. No function has
been shown for these erect roots, although Sanford
(quoted from Sanford & Adanlawo 1 973) suggested
that they might be advantageous in catching and
holding organic debris.
Very recently an anatomical study of the erect
roots of A nsellia africana (Sanford & Adanlawo
1 973 :3 1 2) showed that the exodermal cells have very
little thickening of their walls. The authors consider it
very probable that these roots are more permeable
than the others and so may have a special absorptive
function. The geographical distribution of these two
orchids (Fig. 6 3), (A nsellia africana occurs in
deciduous forest or wooded grasslands and
Graphorchis lurida in exposed habitats in dry and
wet forest types), might also indicate that such roots
may participate in the water uptake. The existence of
th�se roots greatly increases the surface area of the
root system towards the air. Water vapour might
condensate on the roots during the night and then be
transferred to the living tissues of the plants. This
could be one explanation to the occurrence of these
two species in dry habitats. Incidentally, A nsellia
gigantea (Fig. 1 02), which is closely related to A .
africana, occurs i n (for epiphytes) extremely dry
Fig. 62. Root system of A nsellia africana (D.J. 4 1 2 habitats in East Africa (Piers 1 96 8 : 1 5 8).
UPS).
may enter the cortex. (Dycus & Knudson 1 95 7.) Storage of water
In the Nimba area the velamen is particularly well By comparing the total weight of various organs and
developed among the species that grow in exposed their water content the distribution of water in the
positions on substrates with insignificant amounts of plant may be obtained. In most cases it is very dif­
humus depostis, e.g. Ae·rangis laurentii (Fig. 60), ficult to remove the total root system, which natural­
Plectrelminthus caudatus, and Rangae·ris rhip­ ly limits the use and exactness of this method.
salisocia. A conspicuous habit of the aerial roots is Another problem is that the variation in water con­
exhibited by Ansellia africana and Graphorchis tent during the year may be unproportional between
lurida (Fig. 6 1 )
. various parts of the plants. This method, however,
From the "normal" aerial roots attached to the will give an idea of what part of the plant holds the
substrate numerous negatively geotropic roots arise bulk of the water (cf. Gessner 1 95 6). The water con-

Fig. 63. Records of


A nsellia africana (0), and
Graphorchis lurida (e) in
West Africa.

Acta phytogeogr. suec. 59


52 Dick Johansson

Fig. 64. Medinilla mannii. Collected from a Drynaria


laurentii - A splenium megalura epiphyte community.
Nimba Range 700 m. Fig. 65. Begonia rubro-marginata exemplifies stem suc­
culence among the epiphytes. Grassfield, december 1 964.
tent is determined as the difference in weight between
the fresh weight and the dry weight (24 hrs in 1 05°C)
Bulb-like tubers occur on Medinilla mannii and
expressed as percentage of the fresh weight.
Remusatia vivipara. Small bulbs are frequently
The figures presented refer to the mean of five
found on the roots of Nephrolepis undulata; their
separate investigations of the species concerned.
function is unknown.
In the root system
In the stem
Some epiphytes with a large root system compared
As a rule the stem and rhizomes of most epiphytic
to the stem and leaves, e.g. Medinilla mannii (Fig.
orchids are woody and hold only minor amounts of
64) naturally hold a major part of the water in the
the water supply. A nsellia africana, with its spindle­
roots. Orchids with a proportionally large volume of
shaped stem is an exception. Begonia rubro­
the water supply in the roots are e.g. Plectrelminthus
marginata (Fig. 65) and Rhipsalis baccifera are
caudatus and Tridactyle armeniaca (Table 1 0).
good examples of stem succulence. The rhizome of
certain ferns is utilized for water storage, e.g.
Table 1 0. Water content in different parts of some orchids as a
percentage of the total water content in the plant. Mean of five Drynaria laurentii which holds 8 5.4 % of the total
samples. water content in the roots + rhizome (7.4 % and
7.2 % respectively in the fertile and sterile leaves).
Root Stem Leaves

C alyptrochilum christyanum 29.9 1 0. 2 5 9.9 In the leaves


Plectrelminthus caudatus 72.0 1 .2 26.8 The Bulbophyllum orchids have a part of the leaf
Tridactyle armeniaca 70. 7 6.5 22.8
developed into a pseudobulb which is built up by

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 53

Table 1 1 . Water content in different parts of some observation that a leaf will temporarily continue its
Bulbophyllum species as a percentage of the total water content transpiration after it has been detached. This
in the plant. Mean of five samples. transpiration is noticed as a decrease in weight.
Loss in weight of plant material is often regarded
Root + Leaf Pseudo-
rhizome proper bulb
as a measure of water loss, since changes in water
content involve greater weight changes in the plant
Bulbophyllum cocoinum 7.9 29.3 62.8
Bulbophyllum congolanum 1 3 .9 34. 1 5 2.0
Table 1 2. Transpiration rates and some records of stomata
Bulbophyllum winkleri 3.5 3 1.9 64.6
numbers in the detached leaves of some epiphytes. The
transpiration rates are given as the loss of weight per hour in
percentage of the fresh weight. All figures refer to the mean of
water holding tissues. The proportions between the five samples.
water in the anatomical leaf (leaf proper + pseudo­
bulb) and the rest of the plant were roughly 9 : 1 for Species No. of stomata per sq. mm Transpi-
three species examined (Table 1 1 ). Upper Lower ration
epidermis epidermis rates
The fleshy leaves of Calyptrochilum christyanum
hold approximately 60 % of the total water content Polystachya dalzielii 0 70 20.0
of this spec.ies (Table I 0). The Peperomias have Graphorchis lurida 0 72 1 1 .2
fleshy leaves, particularly Peperomia rotundifolia Polystachya puberula 0 58 1 1 .2

(Fig. 5 9). Due to a poorly developed root system and Asplenium geppii 1 0. 7
Polystachya leonensis 0 82 8.6
a thin trailing stem the plant appears to consist main­
Begonia rubro-marginata 6.9
ly of the discoid leaves. Medinilla mannii 6.7
Preussiella chevalieri 6.3
Expenditure of water Arthropteris orientalis 5.6

The transpiration of epiphytic orchids was found to Ancistrochilus rothschildianus 0 52 5.5


Brachycorythis kalbreyeri 0 52 6.3
be low by Kamerling ( 1 9 1 2). It was for some time
Polystachya affinis 0 66 4.9
thereafter assumed that this was a typical feature for Drynaria laurentii (fertile leaO 4.5
epiphytic plants. Later, Spanner ( 1 9 3 9) showed that Nephrolepis undulata 4.2
certain epiphytes, e.g. Myrmecodia, have high Tridactyle tridactylites 12 52 3.7

transpiration rates, and the opinion was established Elaphoglossum isabelense 3.5
Phymatodes scolopendria 2.9
that the epiphytes, as most other ecological groups,
Plectrelminthus caudatus 0 1 04 2.9
provide examples of great variation in transpiration Polystachya rhodoptera 0 62 2.8
rates. Short period transpiration tests by Coutinho in Oleandra distenta 2.4
a virgin forest in Brazil, quoted from W alter Angraecum subulatum 2.3

( 1 97 1 : 1 3 5 - 1 3 6), gave very low transpiration rates Microsorium punctatum 2.0


Aerangis laurentii 0 58 1.7
for orchids and bromeliads.
Angraecum distichum 1.5
Coutinho ( 1 964) also observed the 'De S aussure Polystachya laxiflora 1 .4
effect', i.e. the absorption of atmospheric C02 dpring Habenaria leonensis 0 34 1 .0
the night and its assimilation in daylight, in epiphytic Rhipsalis baccifera (stem) 1 .0

bromeliads and orchids. Similar observations in Listrostachys pertusa 0.9


Cyrtorchis arcuata 0.8
fleshy orchid leaves have been reported by Arditti &
Angraecum podochiloides 0 66 0.7
Dueker ( 1 968 :866). Diaphananthe pellucida 0 42 0.7
The 'true' transpiration value is rather difficult to Diaphananthe bidens 0 56 0.6
obtain and this requires extensive instrumentation. Tridactyle anthomaniaca 10 60 0.6

When the aim is to establish transpiration differences Bulbophyllum linderi 0 46 0.5


Bulbophyllum winkleri 0.4
between various species simpler methods can be
Bulbophyllum saltatorium 12 48 0.4
used. Bulbophyllum bufo 8 70 0.4
In this study the detached leaf method, or quick Bolusiella talbotii 0.3
weigh technique, was used (Iwanoff 1 928, Stocker Chamaeangis vesicata 0.3

1 929, Willis & Jefferies 1 963, Swanson & Lee 1 966, Nephrangis filiformis 0.3
Calyptrochilum christyanum 0.3
Rutter et al. 1 96 3). This method is based upon the

A cta phytogeogr. suec. 59


54 Dick Johansson

than do changes in the content of any other sub­ Table 1 3 . Transpiration rates in the detached leaves of some

stances present. The transpiration rate at the time of orchids. Given as the loss of weight per hour in percentage of
the fresh weight, and in absolute figures (g/dm2/hr).
cutting the leaf m ay be obtained by extrapolation
from the values in weight decrease during the first Species Transpiration (water loss)
three minutes. The effects of abscission and placing % of fresh g/dm2/hr
the leaf in a different environment for the weighing weight/hr
Graphorchis lurida 1 1.21 0. 1 4
make the accuracy of the result questionable
Bulbophyllum linderi 0.72 0.08
(Iwanoff 1 928, Rufelt 1 963, Willis et al. 1 96 3). But
Bulbophyllum winkleri 0.4 1 0.05
to the extent that these effects are similar for C alyptrochilum emarginatum 0.3 6 0.04
different species, the method is valuable as an index­ Diaphananthe pellucida 0.66 0.03
ing technique (Kochenderfer & Lee 1 97 3 : 1 77). Cyrtorchis arcuata 0.56 0.02

Since the transpiration rate is influenced by the


temperature, relative humidity and even time of day
'
all results given refer to a vapour pressure deficit
(V.P.D.) of 20-25 mb. The experiments were carried The high transpiration rates of Polystachya
out in daylight in the afternoon ( 1 4°0 - 1 6°0). puberula that occurs on more humid substrates is
After detachment the scar at the base of the leaf less surprising. The leaves are very thin and lack visi­
was sealed with nail-varnish. The leaf was then ble adaptations towards desiccation.
placed on a torsion balance (sensitive to 0. 1 mg) and Ferns from humid environments could naturally
the decrease in weight during the first three minutes be expected to have high transpiration rates, as
was recorded. The values given refer to ex­ found for Asplenium geppii.
trapolations of these recordings. Three orchid species showed rates of 5 - 1 0 %.
The water loss may be given as the rate of Polystachya /eonensis occurs at the same altitudes as
transpiration compared to the weight of the leaf (Ta­ P. dalzielii, while the two others, Brachycorythis
ble 1 2). Three species of orchids (Polystachya kalbreyeri and A ncistrochilus rothschildianus,
dalzie/ii, Graphorchis lurida and Polystachya belong to moist, drought-protected habitat. Begonia
puberula) showed a loss of more than 1 0 % of their rubro-marginata, Preussiella cheva/ieri and
fresh weight per hour. During the dry season with Medinilla mannii also belong to this group. All these
non-existent rainfall and the substrate drying up, it is species are deciduous. Transpiration rates below
impossible to maintain such a rate of transpiration 1 .0 % have only been found among the orchids, even
and these three species are all deciduous. if R hipsa/is baccifera is close ( 1 .0 %).
Polystachya da/zielii which showed the highest Particularly the orchids with a monopodia}
rate of transpiration grow on the highest, cool and growth, i.e. that keep their leaves during the dry
mist-swept parts of the range, an environment that season, exhibit low transpiration values, e.g.

It is more dif!t cult to understand how Graphorchis


helps to keep the transpiration low. Bo/usiella ta/botii, Calyptrochilum christyanum,
Chamaeangis vesicata and Nephrangis filiformis.
lurida is able to replace the heavy water losses, since Transpiration can also be given in absolute
it occurs in very exposed habitats, mostly on bark, figures, e.g. as g/dm2/hr (Table 1 3). This method
which naturally can only supply a very limited also results in the lowest transpiration rates for the
amount of water. The special root system, previously xeromorphic types of monopodia! orchids, e.g. Cyr­
discussed, may well be the answer to the ability of torchis arcuata, Diaphananthe pellucida, and Calyp­
thi s species to �aintain its water balance. trochilum emarginatum.

A cta phytogeogr. suec. 59


IV. The occurrence of epiphytes

Epiphytes on forest trees


typica' (Table 1 ) and the number of species of all
The occurrence of epiphytes in a particular region vascular plants. A similar quotient called 'l'indice
has been expressed in several ways : epiphytique' has been used by Schnell ( 1 9 5 2 : 3 3 8). In
1 . Total number of epiphytic species in the region the Nimba Mts outside Liberia the figures from the
as documented from floras or extensive collections. mountain forest ( foret montagnardes) are given as
2. The quotient between the number of epiphytes 0. 3 1 -0.3 3 compared to 0.04-0. 1 6 for the forests at
and the number of other vascular plants. the base of the mountains ( forets inferieurs). The
3. Percentage of trees of a certain minimum height results received with this method are dependent on
carrying epiphytes. knowledge of the entire flora in a certain area. As
4. Highest total number of epiphytic species on stated earlier the epiphytic flora tends to be un­
one tree. derestimated, which limits the credibility of this
5. Number of individuals ('stands') of epiphytic method.
species on one tree.
Combinations between these methods have often 3. Percentage of tall trees carrying epiphytes
been used.
High forest
I. Total number of epiphytic species known In the Nimba area the percentage of trees in high
In the area investigated a total of 1 0 1 epiphytic forest that carried epiphytes was investigated in three
orchid species were recorded. From Nimba Range sample plots (clear-felling plots according to
outside Liberia Schnell ( 1 95 2 : 5 30-5 34) recorded 27 Richards' terminology, 1 93 9 : 1 9). Each plot was 50
species of epiphytic orchids which probably repre­ m long and 15 m wide. To simplify the recording,
sent less than 25 % of the total number. It is signifi­ trees smaller than ten meters were excluded since
cant that Schnell only observed four Bulbophyllum they seldom carry epiphytes. The plots were sur­
species (versus 32 in this study). Most members of veyed in advance, involving the limitation, marking
this genus are concealed in the crowns of the tallest and numbering of the trees. Each of the trees 10 m or
trees and are not likely to be found, except when higher was drawn on a sketch figure at its proper
specifically sought for. One of the few reports of this place along the profile (Figs. 66, 67, 68). The crowns
kind from Africa refers to the Budongo rain forest in have been drawn as if they were compact. This is
Uganda (Eggeling 1 947 :5 5-56) : "Excluding hemi­ seldom the case. After the trees had been felled the
epiphytic figs nearly one hundred species of vascular size of the trunk and the crown were recorded with a
epiphytes have been collected from Budongo." This measuring tape.
method is reliable only in regions where the flora is In these three plots the felling of the trees was
very well-known. With rare exceptions it is not useful done for other purposes than vegetation studies. This
in the tropics. meant that once the felling operation started, it con­
tinued with high speed and therefore limited the time
2. Quotient epiphytes/other vascular plants available for the examination of the phorophytes.
The presence of epiphytes may also be expressed as The marking of the trees is very important since in
a comparison to the non-epiphytic plants. Hosokawa all three cases trees outside the sample strips were
( 1 94 3 , 1 950) has introduced this method, called the also felled. The trees fell in various directions, knock­
Epiphyte quotient (Ep.Q.), which is simply the ing down parts of branches from one another and
quotient between the number of species of 'Epiphyta often ending up on top of each other on the ground.

A cta phytogeogr. suec. 59


56 Dick Johansson

50

Fig. 66. Sample plot I. (See text.)


Profile diagram representing a 5 0
m long and 1 5 m wide strip of the
forest. Trees < 1 0 m are excluded.
....--�.
... _,__�-'--LIIDI._.!-'-
!....L --'-'-.I.'--
..l-..-' --»LL...L.\-L-..>.L...-l.-'---1.''"----'-"-
-' L.-'--'-
..Ll.. ---'-'-'
- '-'..L----'WlL...L....�
._, ...J.J...
...J .J.l...
.:.L. ...L so m The figures refer to Table 1 4.

Table 1 4. Occurence of epiphytes on the trees ( 1 0 m or taller) in sample plot 1. The number mark the position of the tree in the pro-
file diagram (Fig. 66).

No. Height (m) No. of species


.
Species of tree Pteridophytes Orchids Others Totai

Piptadeniastrum africanum 1 38 2
Chlorophora regia 2 17
Chlorophora regia 3 17
Lophira alata 4 37 3 4 8
Heritiera utilis 5 23 6 1 7
Piptadeniastrum africanum 6 13
Lophira alata 7 39 4 4 2 10
Heritiera utilis 8 26 2 1 3
Mitragyna ciliata 9 29 5 3 2 10
Piptadeniastrum africanum 10 50 4 7 12
Parinari excelsa 11 11
Parinari excelsa 12 19 3 5
Nauclea diderrichii 13 38 6 4 11
Calpocalyx aubrevillei 14 18 1
Erythrophleum ivorense 15 28 3 3 7
Piptadeniastrum africanum 16 19
Lophira alata 17 44 2 3 6
Calpocalyx aubrevillei 18 16
Calpocalyx aubrevillei 19 25 2 2
Erythrophleum ivorense 20 15 2 4
Chidlowia sanguinea 21 13
Parinari excelsa 22 37 9 4 4 17
U apaca guineensis 23 20 2 2 2 6
Heritiera utilis 24 10
Chlorophora regia 25 26 1 2
Parinari excelsa 26 43 9 8 5 22
Chidlowia sanguinea 27 10
Calpocalyx aubrevillei 28 21 2 2 4
Parkia bicolor 29 29 2 2

Total 29 11 22 4 37

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 5 7

50

Fig. 6 7 . S ample plot 11. (See text.)


Profile diagram representing a 50
m long and 1 5 m wide strip of the
forest. Trees < 1 0 m are excluded.
50 m
The figures refer to Table 1 5 .

Table 1 5 . Occurrence of epiphytes on the trees ( 1 0 m or taller) in sample plot ll. The number mark the position of the tree in the
profile diagram (Fig. 6 7).

Species of tree No. Height (m) No. of species

Pteridophytes Orchids Others Total


Anthocleista nobilis 12
Pycnanthus angolensis 2 19
Anthocleista nobilis 3 12
Parinari excelsa 4 39 6 7 2 15
Parinari excelsa 5 16
Chidlowia sanguinea 6 12
Heritiera utilis 7 35 2 3
Chidlowia sanguinea 8 13 2 2
Heritiera utilis 9 21 2 2
Lophira alata 13 38 6 5 12
Terminalia ivorensis 11 20
Fagara tessmannii 12 26
Klainedoxa gabonensis 13 14
Terminalia ivorensis 14 37
Lophira alata 15 17
Calpocalyx aubrevillei 16 20 2
Piptadeniastrum africanum 17 27
Lophira alata 18 43 3 9 3 15
Parinari glabra 19 13 1
Heritiera utilis 20 31
Alstonia boonei 21 23
Calpocalyx aubrevillei 22 11
Parkia bicolor 23 42 2 2
Fagara tessmannii 24 45
Elaeis guineensis 25 25 4 4
Lophira alata 26 11
Lophira alata 27 10
Elaeis guineensis 28 18 3 3
Anthocleista nobilis 29 12
Chlorophora regia 30 46 5 6 11
Antiaris toxicaria 31 11
Anthonotha fragrans 32 15 2 2 4
Elaeis guineensis 33 25 3 3
Chlorophora regia 34 14
Chlorophora regia 35 14

Total 35 14 19 4 37

A cta phytogeogr. suec. 5 9


58 Dick Johansson

Fig. 68. Sample plot Ill. (See


text.) Profile diagram represent­
ing a 50 m long and 1 5 m wide
strip of the forest. Trees < 10 m
are excluded. The figures refer to
Table 1 6.

Table 1 6. Occurrence of epiphytes on the trees ( 1 0 m or more) in sample plot lll. The number mark the position of the tree in the
profile diagram (Fig. 6 8).

Species of tree No. Height (m) No. of species


Pteridophytes Orchids Others Total

Cryptosepalum tetraphyllum 1 33 2 3 5
Albizia zygia 2 11
Cryptosepalum tetraphyllum 3 32 2 3 5
Terminalia ivorensis 4 15
Triplochiton scleroxylon 5 23 1
Piptadeniastrum africanum 6 45 6 8 14
Anthocleista nobilis 7 16
Elaeis guineensis 8 23 5 5
Anthocleista nobilis 9 12
Elaeis guineensis 10 23 4 1 5
Lophira alata 11 42 3 7 3 13
Lophira alata 12 14
Terminalia ivorensis 13 19
Pentadesma butyracea 14 13
Lophira alata 15 12
Parinari excelsa 16 32 9 4 5 18
Calpocalyx aubrevillei 17 12 z 2
Terminalia ivorensis 18 42
Piptadeniastrum africanum 19 22
Lophira alata 20 16
Cryptosepalum tetraphyllum 21 36 3 3 7
Coula edulis 22 13 1 2
Heritiera utilis 23 18 2 2
Lophira alata 24 45 7 7 3 17
Canarium schweinfurthii 25 24
Piptadeniastrum africanum 26 38 2 3
Cryptosepalum tetraphyllum 27 15
Cryptosepalum tetraphyllum 28 13
Cryptosepalum tetraphyllum 29 14
Cryptosepalum tetraphyllum 30 19
Parinari excelsa 31 33 6 8 4 18
Parinari excelsa 32 12

Total 32 12 26 6 44

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 59

The plots were situated on level ground and on Table 1 7. Occurrence of epiphytes on phorophytes (1 0 m or
well-drained soils. It is doubtful if any of these taller) in high and secondary forest. (High forest figures from

forests are so called primary, i.e. untouched by Tables 1 4, 1 5 and 1 6.)

human activities. However, they represent the


Forest No.of Trees with No. of epiphyte species
highest and oldest sections of forest at this level. In type trees epiph . % Pterid. Orch. Others Total
any case the dominance of Piptadeniastrum High
africanum, Lophira alata, Parinari excelsa and 1 29 62. 1 11 22 4 37
Heritiera utilis in the plots corresponds very well 11 35 40.0 14 19 4 37

with their presence in well-drained forest in the whole 111 32 50.0 12 26 6 44


Total 96 5 2. 1 19 38 8 65
area investigated. Almost no terrestrial herbs oc­
Secondary 27 1 4. 8 9 3 13
curred in any of the plots. The trees and shrubs
below 10 m height consisted of straight poles with
small, thinly foliated crowns. an older secondary forest ('mixed') 1 6. 3 % o f the
The first plot (Fig. 66) was located 5 . 5 km SW of trees. carried epiphytes of 25 species, while in . a
Grassfield at approximately 500 m altitude. This plot primary or climatic climax 'Iron wood' forest, 22.7 %
is close to the areas surveyed by Ad am ( I 969) and of the trees were colonized by epiphytes of 3 5
Gorgla ( 1 969) (see chapter I). The plot examined species. I n this steady rise o f both the occurrence
went through a section of high and dense forest that and the number of species from the colonizing forest
was surrounded by patches of forest of lesser height. to the climatic climax forest, some interesting details
The second plot (Fig. 6 7) was situated 1 km east can be noticed. Nearly the whole increase in the
of the water purification plant in Yekepa, not far number of epiphytic species between the 'mixed' and
from the river Y ah, at approximately 600 m altitude. the 'lronwood' forests can be traced back to the in­
This section of the forest was rather rich in oil palms crease of the orchid species in the latter. In the mixed
(Elaeis guineensis) and Terminalias. The presence of forest there were 1 5 species of orchids constituting
several trees common to secondary forest, e.g. Pyc­ 60 % of the epiphytic flora, while in the climax forest
nanthus angolensis and Fagara tessmannii, might there were 24 species (68 %). There were even fewer
suggest recent human influence. species of ferns in the Ironwood forest than in the
The third plot (Fig. 68) was located 3 km NW of mixed forest, seven compared to eight.
Zulowie in the foothills of Mt Tokadeh at roughly Results from the use of this method, either as an
500 m altitude. This forest was not as dense as the indication of the occurrence of epiphytes or as a
two others. More light penetrated through the characterization of the vegetation, are available from
canopy and small-sized shrubs and trees were more several places (Table 1 8).
numerous. The method is an approximate one that preferably
On the average every second tree ten m or higher can be used when the determination of species is dif­
carried epiphytes (Tables 14-1 7). ficult, or there is a shortage of time. It may refer to
investigations made on felled trees, or in more open
Secondary forest and low-growing forest types to trees standing. The
A survey by telescope (distance observation) was un­ accuracy is in the latter case naturally somewhat
dertaken in a secondary forest, east of the Grassfield lower. It must be observed that the minimum of
airstrip in Nimba. A total of twenty-seven trees in height of the trees that are included in the study is of
sizes from 1 0-29 m were studied. The presence of a utmost importance. The smaller the trees included in
high number of pteridophytes compared to the the study the lower will be the percentage of trees
number of orchid species was observed. The quotient with epiphytes observed. The minimum height must
between pteridophytes and orchids was 3 : 1 as com­ be adapted to the structure of the forest investigated,
pared to 1 : 2 in the high forest (Table 1 7). a fact that makes comparisons between different
There are few remarks in the literature on areas hard to achieve (Grub b et al. 1 96 3 : 5 9 2).
epiphytes in secondary or degraded forests. Eggeling
( 1 947 : 5 6) gives some interesting figures from Ugan­ 4. Highest number of epipbytic species on one
da. In a colonizing, secondary forest (Maeopsis) tree
1 1 . 7 % of the trees bore epiphytes of 1 4 species. In The highest number of epiphytes on one tree in the

A cta phytogeogr. suec. 59


60 Dick Johansson

Table 1 8. Aspects on the presence and abundance of epiphytes as reported from various parts of the world.

Country and Min. height of Total no. of % of trees with Max. no. of ep. Total no. of
author trees being trees being vase. epiphytes on any one tree ep. in the area
considered considered (est.)

Nigeria. (Richards, 1 93 9) 1 5 ft 67 15 13 35

Nigeria. (Richards, 1 9 39) 1 5 ft 75 24 13 35

Uganda. (Eggeling, 1 94 7) 1 5 ft 44 22.7 26 1 00

Sarawak. (Richards, 1 9 3 2) 25 ft 91 13

Sarawak. (Richards, 1 93 2) 25 ft 44 11

British Guiana. (Davies


& Richards, 1 93 3 , 1 934) 5 m 1 93 16 11 200

British Guiana. (Davies &


Richards, 1 93 3 , 1 934) 14 m 55 38 11 200

Ecuador, Lowland.
(Grubb et al., 1 963) 20 ft 42 60

Ecuador, Mountain
(Grubb et al., 1 96 3) 20 ft 52 96

Liberia, Rain forest.


(This publication.) to m 29 62. 1 22 153

Liberia, Rain forest.


(This publication) to m 35 40.0 15 153

Liberia, Rain forest.


(This publication) to m 32 50.0 18 . 153

Tanzania, Riparian forest.


(This publication) 5m 20 60.0 2 7

Tanzania, Wooded grass-


land. (This publication) 5m 24 8.3 7

Tanzania. Wooded grass-


land. (This publication) 4m 50 1 00.0 2 2

three (sample) plots was 22 (9 ferns, 8 orchids, 5


other epiphytes) observed on a 43 m high Parinari
excelsa, but a higher number of epiphytic species on
certain trees has been �bserved (Table 1 9). During a
brief visit to the Usambara Mts in Tanzania, I was
able to count 25 species of epiphytes on a Parinari Table 1 9 . Highest number of epiphytic species recorded (close
excelsa (Table 27). ohs.) on single individuals of some phorophytes common in the

Eggeling ( 1 94 7 : 5 5 ) reports from the Budongo rain Nimba area.

forest in Uganda: " . . . In these plots the maximum


Phorophyte Height Number of epiphyte species
number of species of epiphytes on a single tree is (m)
twenty-six but elsewhere in the forest I have collected Pterid. Ore h. Others Total
between forty and forty-five species from an es­
Heritiera utilis 36 9 20 I 30
pecially suitable host." Richards ( 1 964 : 1 1 3) 33 10 4 28
Mitragyna ciliata 14
probably underestimates the richness of epiphytes on Chlorophora regia 42 6 20 I 27
certain trees : "The number of species of epiphytes on Parinari excelsa 38 7 14 6 27
a single tree is seldom very large even in rich dis­ Lophira alata 42 6 12 4 22
Piptadeniastrum africanum 49 7 13 21
tricts. The largest number of species recorded by the

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 61

No of e p i p hytic author is about fifteen in British Guiana and about


species
1 3 in Nigeria." From Brazil 53 species of epiphytes
22
on one tree has been reported (Focke 1 893).
20
In the clear-felling plots the number of species of
18
epiphytes on each phorophyte correlated to the
16
height of the p�orophyte is shown in Fig. 69. The
age, species and environment of the . trees in­
12 vestigated naturally exert a strong influence on the
10
number of species of epiphytes that will be present.
Thorold ( 1 952), however, found no appreciable
change in epiphyte rating (see enumeration methods)
between cocoa trees, 1 5 respectively 60 years of age,
while Richards ( 1 93 9 :30) states "epiphytes occur
chiefly on the largest individuals of each species of
S ize of
--....-
� C!XD <OOXi> <iXD <D <D <D
-- .. --..----....--
... ---4 Phorophyte
tree".
10 20 30 40 so m The usefulness of this method as an instrument for
comparison of the epiphyte flora in various regions is
Fig. 69. Correlation between the size of the phorophyte
limited. Comparison must be performed on the same
and the occurrence of epiphytes in sample plots 1-111
(Tables 1 4- 1 6). Figures in circles indicate the number of species and sizes of phorophytes due to the strong
trees in each size class that were devoid of epiphytes. correlation between epiphyte flora and the
phorophyte. For instance in the first plot a Parinari
Table 20. Number of epiphyte 'stands' (a) and their distribution excelsa (no. 26) carried 62 % of the total number of
(b) on a 43 m tall Heritiera utilis. Seka Valley 700 m. The distri­ epiphytic species (that also occurred on other
bution is given as a percentage of the total number of 'stands'.
phorophytes) in that plot. Similar results are found in
the t.wo other strips. A Lophira alata (no. 1 8) in the
(a) Species 'Stands '
second plot and Parinari excelsa (no. 3 1) in the third
Bulbophyllum oreonastes 443
carried 40.5 % and 40. 9 % respectively of the species
Bulbophyllum linderi 361
Listrostachys pertusa 274
in their plots.
Tridactyle armeniaca 153
Elaphoglossum salicifolium 1 05 5. Number of epiphytic specimens or 'stands' on one
Bulbophyllum saltatorium 1 09 tree
Elaphoglossum isabelense 88
In the Nimba area it was observed that certain
Tridactyle anthomaniaca 78
species of trees may harbour a very high number of
Rangaeris muscicola 49
Cyrtorchis arcuata 49 epiphytic individuals. A total of 1 85 7 individuals
Polystachya polychaete 37 ('stands') were counted on a Heritiera utilis (Table
Bulbophyllum schinzianum 27 20), and on a Parinari excelsa.with one large branch
Polystachya saccata 21
excluded , 1 1 7 1 individuals ('stands') were recorded.
Nephrolepis undulata 20
Enumeration of the total number of individuals
Asplenium megalura 18
Bulbophyllum barbigerum 16
('stands') of epiphytic species that occur on a tree is
Begonia rubro-marginata 8 made difficult due to the inaccessibility and the
Raphidophora africana 6 growth habits of many epiphytes. Richards
Polystachya tessallata 5
( 1 964 : 1 1 3) states : "No information is available as to
Drynaria laurentii 4
the number of individuals of epiphytes for the
Begonia mannii 3
Graphorchis lurida 2 tropics, but in New South Wales, Turner quoted by
Remusatia vivipara Longman & White ( 1 9 1 7 :64) counted over 200 in­
dividuals of epiphytic orchids on one tree."
Total 1 857
The number of epiphytes in this sense is not very
(b) Section ll Ill IV V useful in determining their occurrence in a particular
Distribution 5. 7 50.7 39.3 4.3 geographical region because of the same reasons as
of the 'stands' presented in the previous method. Grubb et al.
·

A cta phytogeogr. suec. 59


62 Dick Johansson

( 1 963 :592) notices this in Ecuador : "Amongst


twenty-five trees bearing vascular epiphytes on the
lowland plot three trees bore 40 % of the total
number."
However, records of this nature are valuable when
analyzing the correlation between the phorophyte
and its epiphytic flora.

Epiphytes on other substrates

On cultivated trees
Trees, many of them of exotic origin, that are kept in
cultivation in and around the villages seem to be of
varying suitability as phorophytes. The common
mango tree, Mangifera indica, has a very poor or
non-existent epiphyte flora. Citrus trees (grapefruit,
orange and tangerine) seem to carry epiphytes only
occasionally, their bark, however, is covered with
white crustaceous lichens to such a degree that it
often gives a whitish impression.
Coffee trees may be observed with a substantial
number of epiphytes.

Table 2 1 . Epiphytes recorded on Hevea brasiliensis in Firesto­


ne plantations at Barbel, Liberia. Fig. 70. Epiphytes on a rubber tree (Hevea brasiliensis).
Firestone Rubber Plantation, H arbel, Liberia. (A)
Pteridophytes Bulbophyllum cochleatum Microgramma owariensis, (B) Graphorchis lurida, (C)
Arthropteris monocarpa Bulbophyllum cocoinum Bulbophyllum falcatum, (D) Drynaria laurentii.
Arthropteris orientalis Bulbophyllum congolanum
Asplenium megalura Bulbophyllum falcatum The cocoa trees in the Nimba area seem to offer a
Davallia chaerophylloides Bulbophyllum intertextum rather good environment for epiphytes. A total of 1 3
Drynaria laurentii Bulbophyllum linderi
species of epiphytes have been observed on this
Elaphoglossum barteri Bulbophyllum nigritianum
phorophyte. Thorold ( 1 9 5 2) reported 32 species on
Elaphoglossum salicifolium Bulbophyllum oreonastes
Lomariopsis guineensis Bulbophyllum pavimentatum
cocoa trees in Nigeria.
Loxogramme lanceolata Bulbophyllum recurvum The cola tree that is found outside almost every
Lycopodium warneckei Calyptrochilum emarginatum village shows a striking richness in epiphytes. Not
Microgramma owariensis Cyrtorchis arcuata less than 2 1 species of ferns, 39 species of orchids
Microsorium punctatum Cyrtorchis monteiroae
and 5 species of other vascular epiphytes have been
Nephrolepis biserrata Diaphananthe bidens
Nephrolepis undulata Graphorchis lurida
recorded on this small tree.
Oleandra distenta Listrostachys pertusa Ornamental trees, e.g. Delonix regia (the flam­
Phymatodes scolopendria Polystachya tessallata boyant tree), Jacaranda sp. , Terminalia catappa,
Platycerium stemaria Tridactyle anthomaniaca and Sphatodea campanulata P. Beauv., all seem to
Vittaria guineensis Tridactyle armeniaca
offer a poor milieu for epiphytes. This may be a local
Orchids Other vascular epiphytes
effect since, in Douala, C ameroon, the flamboyant
Aerangis biloba Begonia polygonoides trees that line the streets are overgrown with
Ancistrorhynchus cephalotes Peperomia sp. epiphytes, Microgramma owaroides and Poly­
Angraecum distichum Preussiella chevalieri stachya odorata Lindl. among others.
Bulbophyllum bufo Rhipsalis baccifera
The rather young rubber trees in the plantations in

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 63

Fig. 7 1 . Epiphytes on an oil palm (Elaeis guineensis).


Yekepa 600 m. (A) Phymatodes scolopendria, (B) Vittaria
guineensis, (C) A splenium megalura, (D) Lycopodium Fig. 72. Cyathea manniana. Iti river valley. Nimba Range
warneckei, (E) 0/eandra distenta, (F) Nephrolepis un­ 800 m.
dulata, (G) Microsorium punctatum.
On tree-ferns
Two species of tree-ferns are present in the Liberian
the Nimba area hold very few epiphytes, but old Nimba, Cyathea camerooniana Hook. and C. man­
trees in the Firestone rubber plantation at Harbel are niana Hook. Cyathea camerooniana is found in deep
very rich in eiphytes (Table 2 1 ) (Fig. 70). A total of shade, in gullies often near streams, or close to small
45 species were observed. swamps from 500 m to 1 200 m altitude. It has a
When the forest is cut for farming the oil palms short stem usually between 1 -2 m high. The fronds
are spared for their economic value and the_y are reach up to four meters in length. No epiphytic
generally able to survive the farm-burning. Thus in vascular plants have been noticed growing on these
clearings and secondary bush the palms are con­ stems.
spicuously numerous. As the older fronds die, the Cyathea manniana forms veritable 'forests' in
persistent woody bases remain adherent to the trunk, suitable places along the streams (Fig. 72). This tree­
covering its length with horny stubs sharply angled fern is rather common, particularly at higher
upwards, so that large quantities of dead plant altitudes. It is like a slender oil palm in shape. The
material are collected. In these humus deposits many stem may reach a length of 10 m with a diameter of
epiphytic pteridophytes are growing (Fig. 7 1 ). Only only 5 -6 cm. The stem is built up by persistent leaf­
one epiphytic orchid, Habenaria procera, was bases and matted adventitious roots which form an
observed on the oil palm. Several species of other excellent substrate for the roots of the epiphytic
vascular plants, both "true" epiphytes and oc­ plants. As a matter of fact the stem from tree-ferns is
casional or ephemeral (facultative) ones, utilize this considered by professional orchid growers to be one
habitat (cf. Van Oye 1 92 1 , Vanderyst 1 922). of the best potting media for epiphytic orchids.

A cta phytogeogr. suec. 59


64 Dick Johansson

Table 2 3 . The epiphytic flora on eleven dead trees. Yekepa 500


Table 22. Epiphytes recorded on 25 Cyathea manniana, (2 m
m.
or taller). Nimba Range 700 m. The figures refer to the percen­
tage of the total number of tree ferns that carried each species of
Group Species 'Stands' .
epiphyte.

Total no. % Total no. %


Species % of tree ferns
carrying each sp. Pteridophytes 8 1 6. 7 43 4.2
of epiphyte Orchids 38 79.2 966 95.2
Other vase. plants 2 4. 1 6 0.6
Arthropteris monocarpa 64
Asplenium dregeanum 52 Total 48 1 00.0 1015 1 00.0
Asplenium barteri 32
Raphidophora africana 32
lmpatiens sp. 32
Both these perennial herbs occur among rocks in
Begonia oxyloba 28 areas that for one reason or another are more or less
Peperomia sp. 24 devoid of trees. In this milieu they simply act as a
Asplenium variable var. paucij. 16 substitute for trees.
Nephrolepis biserrata 16
Nephrolepis undulata 12
Tectaria angelicifolia 12
On dead trees
The epiphytic flora of eleven dead trees was studied
(close or distance observation) (Table 23). The trees
The translucent crowns of these gigantic ferns per­
had no humus deposits, and in some cases even the
mit a certain amount of light to penetrate to the
bark had slipped off forcing the epiphytes to grow
ground creating an almost uniform light intensity.
directly on the wood. Since there was no foliage the
On a sunny day the light intensity at ground level is
plants were subject to full sunlight.
around three times greater in a tree fern forest than
The ratio between the species of pteridophytes and
in a high forest.
orchids was nearly 1 :5 (8 : 3 8). The number of in­
Twenty-five tree ferns (C. manniana), 2 m or
dividuals shows more clearly the effect of the ex­
higher growing in a stream-bed at 700 m alt., were
treme environmental conditions on the epiphytic fern
examined as to the presence of epiphytes on their
flora. Only 4.2 % of the total number of individuals
stems (Table 22). No orchids were recorded or ever
('stands') of the epiphytes were pteridophytes. Some
observed on Cyathea manniana, but the presence of
species occur in large numbers. On a 1 4 m high tree
pteridophytes and other vascular epiphytes was
4 1 6 epiphytic individuals were found, with the
striking. 84 % of the tree ferns examined carried at
dominance of two species, Polystachya puberula
least one species of epiphyte.
( 1 8 1) and P. paniculata ( 1 2 1 ) (Table 24).

On 'herbs'
No epiphytes have been observed growing on herbs Table 24. Number of epiphyte 'stands' on a 1 4 m tall dead tree.
in the Nimba area. Guillaumet ( 1 967 :5 7) reports Yekepa 500 m.

that the curious looking Afrotrilepis pi/osa (Bock.) J.


Species 'Stands'
Raynal., (Cyperaceae), which has stems covered
with closely packed persistent old leaf sheets, may Bolusiella talbotii 7

support an epiphyte flora. Eleven species of Bulbophyllum maximum 19


Bulbophyllum oreonastes 17
epiphytic orchids are reported to occur on old
Chamaeangis vesicata 31
Trilepis pilosa 'trunks' on the slopes of C arter Peak
Diaphananthe rutila 4
in Nigeria (RiChards 1 95 7 :568). One of these is Plectrelminthus caudatus 16
Polystachya odorata Lindl. var. trilepidis Summerh., Polystachya adansoniae 7
which is known only as an epiphyte on Trilepis Polystachya obanensis 3
Polystachya paniculata 121
(Richards 1 95 7 :568). The stems of Vellozia
Polystachya puberula 181
splendens Rend., (Velloziaceae), are covered in the Polystachya tessallata 5
same manner as described for Afrotrilepis. Epiphytic Rangaeris rhipsalisocia 5
orchids very frequently occur on Vellozia splendens
Total 416
in Malawi (Morris 1 968, 1 970 : 1 3).
A cta phytogeogr. suec. 59
Ecology of vascular epiphytes in West African rain forest 65

Some species of phorophytes that have a sparse


epiphytic flora such as Terminalia ivorensis have
frequently been observed to carry a larger number of
epiphytes when dead. Similar observations are also
reported from Uganda: "Smooth barked species e.g.
Cynometra rarely carry epiphytes until they are so
old that the branches have begun to die b ack, when
they too may be fairly heavily colonized." (Eggeling
1 94 7 :56.)

On rocks and soil


In the Nimba area several species of epiphytes occur
on sloping rock surfaces and b are soil patches at
1 1 00- 1 200 m altitude on Mt Gbahm. This habitat . is
created by human activities (road construction) and
sooner or later the ground will be colonized by
terrestrial plants with Dissotis spp. (Melasto­
mataceae) as pioneers. Presently ( 1 9 70) a large
number of epiphytes are thriving there. The most
common are Polystachya dalzielii, P. leonensis, P.
laxiflora and Bulbophyllum oreonastes, which can
all be seen by the hundreds. Less frequent are
Lycopodium warneckei, A ngraecum birrimense, A .
distichum, Bulbophyllum magnibracteatum, B.
scariosum, Graphorchis lurida, Polystachya pan­
iculata, P. pobeguinii, P. rhodoptera, P. saccata, P.
Fig. 73. Epiphytes on a small dead tree. A total of 1 7 tessallata, Rangae'ris muscicola, Tridactyle
species were recorded (2 pteridophytes and 1 5 orchids).
armeniana, and T. anthomaniaca.
(A) Drynaria laurentii, (B) Graphorchis lurida, (C)
Chamaeangis vesicata, (D) Polystachya tessallata and It is well known that epiphytic species under
Phymatodes scolopendria, (E) Polystachya puberula, (F) special circumstances will grow on rocks or on the
Rangae'ris rhipsalisocia, (G) A ncistrorhynchus ground (Oliver 1 930:2, Polack 1 93 3 : 26, Richards
cephalotes. 1 93 6 : 3 2, 1 964 : 1 22, Jaeger & Adam 1 972). This
behavior is natural in areas devoid of suitable phoro­
The rich epiphyte flora that frequently can be phytes and where the microclimate fits the re­
found on recently dead trees has mostly established quirements of the epiphytes. It is in relatively dry
itself several years earlier when the tree was alive. areas that one will find epiphytes on the ground due
During the dying process of trees many epiphytes to lack of trees and weak competition from other
take advantage of the increased light intensities, terrestrial plants. In humid areas the opposite effect
(caused by the falling of the leaves) but probably also appears, i.e. habitually terrestrial plants grow as
by a beginning decomposition of the bark (Went facultative epiphytes.
1 940:9 1 ). When the trees are dead the epiphytes will From the Loma Mts in Northern Sierra Leone,
flourish during the first rainy season. J aeger et al. ( 1 968) report no fewer than 34 species
The very exposed habitat will rapidly exert a of epiphytes of which 1 3 also were found growing on
strong influence on the species composition of the rocks, e.g. Polystachya dalzielii, Bulbophyllum
epiphyte flora favouring the drought resistant scariosum and B. bifarium. Four species of
orchids (Fig. 73). In a few years the altered en­ epiphytes are reported occurring on granite boulders
vironmental conditions combined with an exhaustion in the southern Ivory Coast (Guillaumet 1 96 7 :5 7).
of the substrate will result in an impoverished From Kenya1 Piers ( 1 968 a : 1 5 8) gives an example
epiphyte flora, even if certain species will occur in a of a healthy establishment of an Ansellia gigantea
very large number of individuals. (Rchb. f.) var. nilotica (B aker) Summerh. : "An ex-
A cta phytogeogr. suec. 59
66 Dick Johansson

ceptional specimen plant, a gigant covering some 20 Such a pattern has been observed by several
square metres and growing intermingled with Aloes, authors. Schimper ( 1 903) states that the epiphytic
Euphorbias and other xerophytic terrestrials, grows plants on a tree in a virgin forest are not the same
near the shores of Lake Elmenteita in the Rift V alley from its base to its topmost branches but exhibit a
in soil which is almost pure volcanic ash." well-marked differentiation. Moreau ( 1 943 :8): "A
Rangaeris muscicola which in dry areas of East few small species e.g. A ngraecum viride are found
Africa occurs on rocks is often connected to the always on twigs rather than branches." Holtum
presence of lichens as judged by notes on herbarium ( 1 960) observed that some small epiphytes grow
sheets : "On lic�ens on the huge granite boulders at only on the smaller branches of trees or shrubs.
the bottom of Njombe river cascade, Tanzania" Richards ( 1 964 : 1 1 8) : "Large epiphytic ferns such as
Lynes 1 93 1 (K), "Covering rocks in a mass. C ling­ Asplenium nidus and A. africanum tend to prefer the
ing to rock surface with lichens" Zambia, Abercorn trunks to the branches, but many epiphytes show the
Distr. H.M. Richards 8404 (K), "on lichen-covered opposite preference." Morris ( 1 9 70:5): "On all trees
rocks" Malawi, Mlanje, Morris 9 1 (K), "Epiphytic there is a pronounced stratification of the epiphytic
or lithophytic orchid, always in association with grey flora." But where an existence of some distribution
lichens" N. Rhodesia, Ika Hills, Morze 1 96 1 (K). patterns has been documented no actual investiga­
One might speculate that the lichens with their tion has previously been done to describe or analyze
water-absorbing and water-holding capaCity could them.
create a suitable microclimate for the germination of
orchid seeds. The microclimate could also be of ma­ Ecological subdivision of the phorophyte
jor importance in the possibilities for the young Previous schemes
orchid plants to survive dry periods. To make it possible in a short time to record ac­
The epiphytic orchids, with their slow growth and curately the epiphytes it is necessary to use some
small leaf surface, combined with a root system kind of subdivision of the tree.
adapted to superficial growth, appear suitable for Van Oye ( 1 924 b) divides the oil palm (Elaeis
this kind of habitat. guineensis) into five zones. This zonation is useful
for the oil palms and other palms whose crowns are
built up by large leaves. For the often extensive
Distribution of epiphytes on the phoro­ crowns of the large rain forest trees this zonation is
phytes less useful. In Europe Ochsner ( 1 928) uses the same
The first impression one gets, when looking at the zonation as Van Oye, excluding the submedian zone.
distribution of epiphytic plants within a tropical rain Richards ( 1 9 3 9) in certain cases used the height
forest, is the lack of order. It seems that the above the ground, in others a more relative descrip­
epiphytes are able to grow at any place on the trees. tion, e.g. base of the first branch, in the records of
There are many reasons that help to create this im­ the growing site of the epiphyte. Sometimes these
pression. One of these is the limited sector of the two pieces of information are combined. The use of a
trees that can be obser.ved from the ground, another large number of relative terms, without strict defini­
that many observations are done in areas where tion, is rather confusing. In 28 felled trees, recorded
human activities have disturbed the original as bearing epiphytes, Richards used no less than
stratification, i.e. bright light may penetrate to the eleven such relative 'localities'.
lowest level of the forest. In such cases one often Hosokawa ( 1 954 a) divides the phorophyte into
finds species normally belonging to the topmost four sections. An arbitrary system of 1 5 ft (4.5 m)
layer of the forest growing close to the ground. zones from the ground upwards has been used by
Many other influences may contribute towards an Grubb et al. ( 1 96 3 :592-59 3).
impression of randomness in the distribution of the Tixier ( 1 966) works with three large sections, but
epiphytes. A regular pattern emerges first when one uses 5 subunits within the 'base of the trunk section'.
has the possibility of examining a substantial number The study of epiphytic lichens and bryophytes in
of trees in an undisturbed environment from the out­ temperate climates has caused many subdivisions of
ermost branches in the crown down to the base of the phorophytes. Barkman ( 1 95 8 :3 1 -3 2) has made a
the stem. valuable compilation of that terminology. The lower

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 67

Fig. 74. Simplified drawing showing the distribution of epiphytes. Above : on a 1 2


m long branch of a Mitragyna ciliata. Yekepa 500 m . (A) Raphidophora africana,
(B) Asplenium geppii, (C) Preussiel/a chevalieri, (D) 0/eandra distenta, (E)
A ngraecum birrimense, (F) Polystachya tessal/ata, (G) Bulbophyllum cocoinum,
(H) A splenium megalura, (I) Drynaria laurentii, (J) Rhipsalis baccifera, (K)
Polystachya paniculata, (L) Tridactyle anthomaniaca, (M) Bulbophyllum
oreonastes, (N) Bulbophyllum linderi, (0) Polystachya adansoniae, (P) Bolusiel/a
talbotii.
Right: on a branch of a 4 1 m high Lophira alata. Seka V alley 600 m. (A)
Elaphoglossum isabelense, (B) A ngraecum distichum, (C) Listrostachys pertusa,
(D) Cyrtorchis arcuata, (E) Polystachya po/ychaete, (F ) A ngraecum birrimense,
(G) Drynaria laurentii, (H) Rhipsalis baccifera, (I) Tridactyle anthomaniaca, (J)
Bulbophyllum schinzianum, (K) Bulbophyllum linderi, (L) Nephrangis filiformis,
(M) Bulbophyllum oreonastes.

parts of the phorophytes have generally been subject epiphytes appear. This stratification of the rain forest
to many subdivisions, while as a rule the whole is by no means uniform, but rather large variations
crown has been regarded as one section. This can in can be found (Foggie 1 947, Newman 1 95 4, Robbins
many cases be traced back to poor knowledge of the 1 95 9, Holdridge 1 970). As a rule the crowns are
epiphytes of the crown rather than being based upon rather small compared to the length and size of the
real observations. A number of subdivisions used in trunks, but a few species of trees with the ramifica­
studies of vascular epiphytes is presented in Table tion close to the ground give rise to crowns of large
25. dimensions.
Usually the emergent (superstory ; Odum 1 970 a)
Biotopes on the phorophytes in the Nimba area trees have more flattened and expanding crowns
In the tropical rain forest there is a pronounced than the trees in the lower ' strata' of the forest.
stratification of the vegetation. The crowns of the The richness in biotopes which can be found in the
trees form one or more 'strata' and it is in the top­ crowns of the large trees is in sharp contrast to the
most of these 'strata' that the main part of the more uniform environment in the lower 'strata' of the

A cta phytogeogr. suec. 59


68 Dick Johansson

Fig. 75. Plank buttresses of Pip­


tadeniastrum africanum (above)
and stiltroots of Uapaca guineen­
sis (below). Yekepa.

Table 25. Subdivision of the phorophyte by some authors


forest. This fact is reflected in the zonation used in
Author 'Trunk' 'Crown' this study.

Van Oye 1 9 24 region basale, r. sous- region eo-


Subdivision used in this study
mediane, r. mediane, r. ronaire
sous-coronaire A subdivision must conform to the varying size and
shape of the trees to make comparisons between
Ochsner 1 928 basis-teil, mittelteil, kronenteil
various phorophytes meaningful. Any simple system
subkronenteil.
of height zones will inevitably fail to correspond with
Schnell 1 950 strate epiphytique strate epi- natural 'zones' of environmental conditions, because
inferieure phytique
of the uneven nature of the canopy as pointed out by
superieure
Richards ( 1 964 : 1 1 6) and Grubb et al. ( 1 963 :592).
Hosokawa 1 9 5 4a Trunk-bases, trunks Crown-bases,
Hosokawa ( 1 95 4 a) noticed two zones on the
crowns
branches, crown-bases and crowns. One also finds
Tixier 1 96 6 Base de troncs, strate Cime these two zones on the phorophytes in the Nimba
fructescente area. However, there is very frequently in the middle

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 69

Recording and registration of epiphytes


The very limited knowledge of the ecology of the
epiphytic plants in the rain forest is, as previously
mentioned, connected with the difficulties in studying
them in the field (Fig. 7 7). As a matter of fact a large
part of the epiphytes described from the rain forest
are species naturally occurring on the lowest part of
the trunks.
Under ordinary circumstances it is from the
ground level difficult to even get a limited view of the
crowns of the highest trees. The shorter trees,
together with lianas and other climbers, form a layer
of varying thickness that can be looked through only
11
here and there. This means that even if it is possible
to study a certain part of the crown, the overall pic­
ture is not discernible.
The possibility of a definite host specificity also
greatly complicates any attempt of random sampl­
ing, as has been pointed out by C urtis ( 1 95 2 : 5 5 1 ).
One may of course study the epiphytes that occur on
the trees along roads, at the edge of the high forest
etc. In these cases one will find an epiphytic flora
Fig. 76. Subdivision o f the phorophyte. that is by no means representative for the high forest
Section : proper.
I The basal parts of the trunk (0-3 m) To be able to compare the epiphytes from the high
11 The trunk from three m up to the first ramification
forest, observations have also been performed on
Ill The basal part of the large branches ( 1 /3 of the total
length of the branch) phorophytes in other environments (e.g. around
IV The middle part of the large branches ( 1/3 of the total villages and in plantations).
length of the branch) To enable the use of all possible information three
V The outer part of the large branches ( 1 /3 of the total different observation methods have been applied in
length of the branch)
the recording of the epiphytic flora. They may simp­
ly be called the close observation, distance observa­
tion and occasional observation methods. One
separate record was produced for each phorophyte
part of a branch an epiphytic flora of a different that was examined.
composition than the one occurring at the base of the
branch, or in the outermost parts (Fig. 74). This Close observation
irregular distribution of the epiphytes can be found This method means that the investigation has been
on trees of varying sizes. The branches have performed on a tree lying on the ground. The access
therefore been divided into three sections of equal to felled trees was limited, but occasionally road con­
size. struction, landscaping or forestry operations
The trunk has been divided into a short section produced useful material. Trees knocked down by
close to the ground, and a considerably longer sec­ natural causes are very seldom found.
tion that includes the rest of the trunk up to the first The investigation of a felled tree must be carried
ramification. This division is based upon the easily out almost instantly, since many plants will either die
observed difference between the epiphytic flora at the or be favoured by the change in the environment.
base of the tree and the one occurring on the upper Commonly the orchids suffer from too low light in­
part of the trunk (Fig. 7 5). The division of the tensities, while certain species of pteridophytes have
phorophytes into sections used in this study are sum­ an advantage with the higher relative humidity close
marized in Fig. 76. to the ground. The practice of shifting cultivation

A cta phytogeogr. suec. 59


70 Dick Johansson

Fig. 77. The concealment of the epiphytes by the foliage observed. Note the distribution of the large Platycerium
as illustrated by a Triplochiton scleroxylon. This tree is ferns.
deciduous for a short period when the epiphytes are easily

might seem to offer an immense material. Unfor­ the monopodia] orchids grow out from the branches,
tunately these farming activities are to the largest ex­ and thus a very minute part of the plant, aside from
tent carried out in so-called secondary forests. Here the extensive root system, covers the sampling area.
the vegetation is after repeated destruction kept in The scale constructed is therefore based on the ac­
various young secondary growth stages. tual number of plants rather than the area covered.
When the tree is lying on the ground it is relatively The transects can preferably be drawn along the
simple, at least when one knows well enough the branches in the crown, since it is here that the most
vegetative parts of the epiphytes to make deter­ substantial changes occur in substrate, light and
minations, to map the· distribution and to study the microclimate. From the point of ramification, where
ecology of the various species. Certain large the branch is relatively broad, the transect as it runs
branches can be used to illustrate the continuous aiong the branch, covers a smaller and smaller area.
changes in the epiphytic flora from the base to the This fact should be considered when reading the
outermost parts, so-called 'transects'. transect results (Fig. 79).
In the description of vegetative change along an
environmental gradient the 'transects' are of a con­ Distance observation
siderable importance. But since neither the trunks Such investigations have been performed on a stan­
nor the branches represent straight lines a transect of ding phorophyte on which it has been possible to in­
traditional type is not easy to work with. Therefore a vestigate the trunk and at least one complete branch
modified type of transect has been used. The method (or more seldom the whole phorophyte). This
is simply to let a measuring tape follow the twistings method, described by Went ( 1 940), allows the recor­
and bends of the stretch to be examined. This stretch ding of the epiphytic flora from the ground. With the
is later presented as if it was a straight line. Most of aid of field glasses, or rather a telescope on a tripod,

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 71

it is possible to identify the majority of epiphytes. An Occasional observation


advanced knowledge of the vegetative parts of the It is normally possible to study the epiphytic flora
plants is a prerequisite since you seldom find the only on a limited part of the phorophyte. Such
plants in a flowering state. To acquire that records that may refer to close as well as distance
knowledge a sample of almost every epiphyte was observations, are called occasional observations.
collected and kept alive. They thrived growing on From the lm.yest parts of the phorophytes a large
substrate similar to the one they were collected from, number of observations may easily be obtained and
(in small wooden boxes, 20 x 20 x 5 cm, with broad parts of branches that have fallen down from the
slots in the bottom). The boxes were kept on shelves crowns also offer abundant material.
in a specially constructed 'frame house'. Species dif­ For most species these records add little or
ficult to classify were also kept in this manner until a nothing to the knowledge achieved from close and
safe determination could be made. distance observations, but for certain species this is
From this collection of living epiphytes informa­ ± the only information about their distribution and
tion was extracted which proved useful in the field general ecology that has been possible to obtain. The
work. Through a telescope at a magnification of 40x records from this latter group of species are listed
even small details can be observed, and epiphytes a under Occasional observations (Table 29).
few centimeters in size can be identified on the
branches of the highest trees. As a rule Registration
characteristics, such as remaining floral parts The traditional way of determining the abundance or
pseudobulbs etc., are of great help in the determina­ presence is not particularly useful for epiphyte
tion process. (A description of the vegetative parts of rating. In most cases, the limited presence, and the
the orchids has been presented in chapter 11.) stout and rather large size of many epiphytes, make
However, for certain orchids, there are sometimes it possible to count the individual plants, or groups of
only small details in the flowers that separate two plants.
species from one another. In these cases the generic The sympodial plants with their vegetative mul­
name only has been used in telescopic surveys. tiplication present a special problem. These plants
The pteridophytes were, for the most part, easier often form dense 'mats' of rhizomes and stems mixed
to separate from each other than the orchids. There with immature plants. This makes it very difficult to
are very few morphologically similar species that determine the number of individual plants that are
could cause problems in the identification. present.
Among the other vascular epiphytes the An arbitrary system of epiphyte rating, primarily
Peperomia species are difficult to determine. qualitative and only quantitative in so far as impor­
There are certain disadvantages with this method. tance is attached to the presence of one or more
It is very time consuming. The plants growing in the colonies has been used by Thorold ( 1 95 2).
most exposed places will easier be discovered than Grubb et al. ( 1 963 :59 1 ) works with the term 'an
those in more obscure habitats. Large growing individual' for each clearly separated clump of
species will be spotted more often than minute plants. Sanford ( 1 968b : 5 6) has given an accurate
plants. Many more objections can be raised. It is description of the problems concerning the enumera­
therefore important to separate the observations with tion of epiphytic orchids : "Actual counting of orchid
the distance method from those performed in other plants on a tree presents special difficulty. Many
ways. plants such as Bulbophyllum spp. forms mats of
The distance recording served two purposes. C er­ pseudobulbs connected by relatively long rhizomes.
tain species of epiphytes were never observed in the It is impossible to determine where one plant ends
studies of felled trees. The distribution on the and another begins. Other species, such as those of
phorophytes and habitat conditions for these species the genus Calyptrochilum, form great masses of pen­
must therefore be based upon information from this dulous stems, and again determination of individuals
type of investigation. Furthermore, the records ob­ is arbitrary. For this reason a special usage of 'stand'
tained from the distance recordings, can be used to is used in these enumerations. A stand is taken to
determine the accuracy of this method in com­ mean a collection of individual stems or/and plants
parison to the close investigation method. spatially separated from another group of the same

A cta phytogeogr. suec. 59


72 Dick Johansson

species either by an area on the tree devoid of strate at its growing site were recorded in a three­
orchids or occupied by another species. In the case graded scale.
where the same area is occupied by an intermingling
of more than one species, one stand is counted for L ight
each species present." Light intensity was divided into three classes : heavy
During the investigation the same unit (called shade, open shade and full sun. The amount of light
'group'), as described by Sanford, was used. In order available varies of course with the time of the day
not to cause confusion by adding one more name for and cloudiness of the sky, and also somewhat with
the same thing, the term 'stand' sensu Sanford is the time of the year. Since the light intensity was
adopted. only estimated, mistakes cannot be excluded. How­
ever, the main bulk of the observations are easy to
Ecological observations group.
In order to get a rough idea of the importance of cer­ Light intensities regarded as heavy shade normal­
tain environmental influences on the distribution of ly appear in the lower strata of the rain forest. In ex­
the epiphytes within the forest, attention has been ceptional cases this intensity can be found higher up
paid to light and substrate. For each epiphyte that in the trees due to dense foliage. In most cases there
was recorded, the light intensity and type of sub- is no problem to separate heavy shade from open

Table i6 _ The trees ( 1 0 m or taller) o f the high forest included in the s tu d y . Trees with epiphytes are marked

C lose observation =
D= ta
iso..=
=-on=
ce ob
"---
"--'
-"' s e"---
'--"
rv:. a
:-=:
.:: Lic:
on
.: _
:..:._ ___________

No . % with No . % wi th
Phorophytes epiphytes Sizes epiphytes Sizes

Albizia glaberrima 50 . 0 36, 38, 38;40'


A. zygia lOO. 0 1 7 ; 21 ; 2 6 ; 60 . 0 1 2 , 1 3 , 1 8 ; 21 ;24 '
A Is tonia boonei 50 . 0 1 2 , 14, 1 7 ; 2 1 ; 2 5 , 27 ; 2 9 ; 3 o 6 33 . 3 1 3 , 1 5 , 1 8 , 1 9 , 22 ;3 1 '
Amphimas pterocarpoides 50 . 0 1 1 , 1 4 , 1 4 ; 34 . 4 50, 0 12, 1 6 , 29;32 .
Anthocleista nobi l i s 28 . 5 1 1 , 1 4 , 1 4 , 1 5 , 1 6 ; 17 ; 1 8 12 16.7 1 0 , 1 0 , 1 1 , 1 1 , 12 , 1 2 , 1 3 , 1 3 ' 1 5 , 1 5, 1 6 ; 1 9 '
Anthonotha fragans 1 00 . 0 17. 85. 8 1 1 , 1 5; 1 6;2o;21;25;32·
Antiaris toxicaria 1 00 . 0 34;35' 66. 7 22, 24;28, 29;30 ;34 .
Berlinia confusa 66. 7 1 4 , 33 ; 41 · 50 . 0 13, 17'
Bombax buonOP07.ense 66 . 7 24, 30;41 · 0. 0 26
Bussea occidentalis 66 . 7 21 ; 2 2 , 2 6 '
C a lpocalyx aubrevillei 13 46.2 1 1 . 1 1 , 12 , 1 2 ; 14 , 1 6 , 1 s ; 2o ; 2 3 , 2 3 , 2 5 ; 2 6 ; 2 9 · 2 50 . 0 14, 1 8 '
Canarium schweinfurthii 40 . 0 1 3 , 14 , 2 1 , 2 9 ; 3 8 ' 5 60. 0 1 7 . 2 0 , 2 3 ; 26 ; 3 4 '
Ceiba pentandra 1 00 . 0 19;31 . 12 83 . 3 20, 2 5 , 2 6 ; 2 9 ; 3 0 ; 34 ; 3 6 ; 3 7 : 4 0 ; 4 0 ; 4 0 ; 42 .
Chidlowia sanguinea 14 42 . 9 1 1 , 1 2 , 1 2 ' 1 2 ; 1 3 , 1 3 , 14 , 14 ; 14 ; 1 5; 1 5; 1 6 , 12 7 5. o 1 2 , 1 2 ; 1 3 ; 1 4 , 1 6 ; 17 , 17 ; 1 s ; z o ; 2 0 ; 21 ; 24 •
18, 1 8 .
C hlorophora regia 60 . 0 17, 1 8, 28;39;42' 75. 0 1 8 , 2 3 ; 2 5 ; 27 ; 2 9 , 2 9 ; 3 0 ; 3 9 '
Chrysophyl lum perpulchrum 75. 0 23;2 6 ; 3 1 , 3 1 ' 1 00 . 0 28'
Combretodendron macrocarpum 75.0 1 4 , 1 4 ; 27 ; 31 ' 71. 4 12 , 1 2 ; 1 4 , 1 7 ; 1 9 ; 1 9 ; 3 4 '
Cou\a edulls 66.7 16, 26;30' 75, 0 23 ; 25, 3 0 ; 3 4 .
C ryptosepalum tetraphyllum 66.7 1 0 , 2 1 , 2 3 ; 24 ; z s ; 27 • 66 . 7 14, 26;28, 28;2n; 34 •
Daniellia ogea 42 . 9 14, 1 6 , 1 9 , 2 0 , 2 6 ; 3 4 ;37 . 75. 0 2 1 , 2 3 ; 2 6 ; 34 .
D i stemonanthus benthamianus 33 . 3 26 ;27 . 30
.
E n tandrophragma utile 1 00 . 0 29;38;60' 66 . 7 2 1 , 4 0 ; 42
E rythrophleum ivorense 75. 0 15, 28;32;36. 75. 0 12;14, 18;41 '
Fagara tessmannii 33.3 2 9 , 3 1 ; 3 2 , 3 6 , 3 8 ;4 2 0. 0 1 4 , 2 1 , 2 1 , 34 , 4 0 , 42
Guarea cedrata 66 . 7 2 1 , 32 ; 3 6 ' 1 00 . 0 1 9; 2 5 '
Heritiera u ti lis 85.7 1 3 , 1 6 ; 1 8 ; 2 6 ; 3 2 ; 3 5 ; 43 . 14 92 . 3 1 7 ; 2 2 , 2 4 ; 2 7 ;2 9 ; 3 o ; 3 o ; 3 1 ; 3 1 ; 3 1 ; 3 3 ; 3 4 ; 3 4 ; 3 7 •
Khaya anthotheca 1 00 . 0 34;3 6 ' 33. 3 14;16, 19
Lophira alata 71.4 1 1 , 1 6 , 3 9 ; 3 9 ; 4 2 ; 43 ; 44 . 50 . 0 1 2 , 1 3 , 24 ; 3 8 '
Lovoa trichilioides 60 . 0 1 U , 24 , 27 ; 3 3 ; 3 8 ' 1 00 . 0 36'
Mammea africana 66 . 7 14;16, 33. 6G. 7 1 3 , 2 0 ; 27 .
Mitragyna ciliata 1 00 . 0 1 1 ; � 2 ;3 2 ; 3 3 · 1 00 . 0 12 ; 1 4 ; 2 6 ; z s ; 3 o •
Nauc\ea diderrichi i 1 00 . 0 2 8 ; 3 2 ;3 8 . 0.0 24
Parinari aubrevillei 1 00 . 0 2 1 ; 23 .
P. excelsa 80 . 0 1 2 , 2 1 ; 2 2 ; 3 s ;42 • ]8 83 . 3 1 2 , 1 4 , 1 4 ; 1 s ; z o , 21 ; z s ; z 5 ; 23 ; 3 3 ; 34 ; 3 5; 3 5 ; 36;
3 9 ; 3 9 ; 4 0 ; 44 .
P . glabra 75. 0 1 9 ; 20 , 2 3 ; 23 .
Parkia bicolor 28 . 5 2 1 ' 2 3 , 2 6 , 29, 2 9 ; 3 0 , 32 . 33 . 3 1 1 ' 2 3 , 2 5 , 2 6 , 27 . 3 0 ; 3 1 ; 3 2 , 3 3 .
Pentaclethra macrophylla 75. 0 2 0 , 2 2 ;2 7 ; 3 4 . lOO. 0 27 ; 3 6 '
Pentadesma butyracea 1 00 . 0 3 1 ;40 ' 50 . 0 1H;23
Piptadeniastrum africanum 50. 0 1 2 , 1 8 , 3 4 ; 4 3 ; 4 9 , 54 . lD 52. 6 1 3 , 1 6 , 2 0 , 2 0 ;2 1 ; 24 , 2 6 , 2 6 ; 2 8 , 28; 3 2 ; 3 4 ; 3 5 ; 3 8,
42; 4 3 . 44 . 5 1 ; 53 .
Pycnanthus angolensis 33 . 3 1 6 , 20, 2 1 , 2 3 ; 24, 27 . 16.7 1 4 , 1 � . 2 4 , 26, 3 0 , 3 4 .
Rhodognaphalon brevicuspe 75. 0 2 1 ;27 , 3 2 ; 3 5 ' 1 00 . 0 33 '
Sacoglottis gabonensis 50 . 0 14,39' 1 00 . 0 2 9 ; 33 ; 3 4 ; 40 .
Terminalia ivorensis 22 . 2 14, 2 1 , 3 1 , 3 2 , 3 � . 4 0 , 4 0 ; 42 , 43 . 11 . 1 1n, 1 H , 23, 30, 33, 40, 4 1 , 4 1 , 41 '
T. superba 28 . 5 28, 3 2 , 33, 35; 4 1 , 42, 42. 20. 0 17 , 2 8 , 42 ;43 , 44
Triplochiton scle roxylon 57 . 1 1 2 , 1 5, 2 2 , 3 0 ; 3 2 ; 4 1 ; 44 . 57 . 1 l O , l l , l !i , z o ; 3 !J ; 4 2 ; 4 5 ·
Uapaca guineensis 1 00 . 0 17 ; z o ; 2 1 • HO. 0 1 2 , 1 2 ; 1 H ; z 5 ; 27 •
Vitex micrantha :;o . o 1 3 , 24 . 1 00 . 0 1 0 ; 1 2 ; 17 ; 23 '

Total 222 24 1

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 73

shade intensities. The light in a tropical rain forest is ( 1 940) used a relative scale to judge the light intensi­
not gradually increasing from the ground level up to ty at the growing sites of the epiphytes. He estimated
the crowns of the emergent trees. It is rather discon­ the light intensity at the growing site in per cent of
tinuous, as many author have pointed out, e.g. the light outside the forest.
Richards ( 1 939 : 30). In the crowns of the trees of the
topmost stratum high light intensities are formed by Substrate
open shade. The limit between full sun and open The substrate was divided into three classes : bark,
shade is also easy to observe, since only plants in e.g. bark without humus attached, minor humus
very exposed posttlons receive light intensities deposits, e.g. bark with thin and scattered humus
regarded as full sun. layer, and large humus deposits. Humus is here
In a study of the epiphytic societies in Java, Went regarded as dead organic material in various stages

Table 27a . Total number of records, according to .the close observation method, o f pteridophytes and other vascular epiphytes in the high fore s t .

N o of Section o f the Heavy Open Full


observ . pho rophytes shade shade sun

II Ill IV V A B c A B c A B c

Pteridophytes

A n t rophyum immersum 33 . 3 66 . 7 33 . 3 66. 7


A. mannianum 1 00 . 0 1 00 . 0
A r thropteris monocarpa 57 . 1 28. 6 14 . 3 14 . 3 42. 9 14.3 2�. 6
A. orientalis 27 70. 4 29. 6 14. 8 48. 1 29 . 6 7.4
A. palisoti
Asplenium aethiopicum 1 00 . 0 1 00 . 0
A. africanum 7 14.3 57 . 1 28. 6 14. 3 42 . 9 28. 6 14 . 3
A. barteri 16 56. 2 6. 2 37 . 5 37. 5 25. 0 37 . 5
A. dregeanum 93 4.3 25. 8 38. 7 28.0 2. 1 2. 1 16. 1 2. 1 38. 7 26. 9 10.7 3.2
A. geppii 22 9. 1 !! 6 . 4 4.5 4.5 27 . 3 6 !! . 2
A . hemitomum 1 00 . 0 1 00 . 0
A. megalura 172 1.2 34. 9 51 . 2 12. B 1.7 12. 2 21 . 5 2 . !J 2 3 , 1! 37 . �
A. variable va r . paucijugum
Daval\ia chaerophylloides 45 8. 9 28. 9 51 . 1 11.1 17 . 8 33. 3 4.4 22 . 2 17 . 8 4.4
Drynaria laurentii 499 3.8 48. 3 41.9 6. 0 31 . 3 6.2 53 . 9 S, 6
Elaphoglossum barteri 28. 6 28. 6 42 . 8 28. 6 14 . 3 57 . 1
E. chevalieri 20 20. 0 50 , 0 25. 0 5. 0 25. 0 50 . 0 15. 0 10. 0
E. isabelense 1 60 16.2 30. 0 53 . 7 26. 2 20 . 0 4.4 2S. l 21 . 2
E. kuhnii
E . salicifolium 208 45. 2 22 . 1 10. 6 26. 4 17 . 8 47 . 1 8. 6
22. 1
Lomariopsis guincensis 1 1 00 . 0 1 00 , 0
Loxogramme lanceolata 32 15. 6 37 . 5 12 . 5 25. 0 9.4 9. 4 15. 6 31 . 2 43. 7
Lycopodium mildbraedii 11 18. 2 81 . 8 63 . 6 9. 1 9. 1 18. 2
L . warneckei 13 92 . 3 7.7 7.7 23 . 1 15.4 38. 5 15.4
Microgramma owa riensis 80 16 . 2 50. 0 27 . 5 6. 2 40 . 0 36. 2 6.2 13.7 3.8
Microsorium punctatum 143 8.4 44 . 1 41.2 6.3 14. 0 37 . 1 9.8 16. 1 22 . 4 0.7
Nephrolepis biserrata 16 87 . 5 12. 5 18.7 37 . 5 31 . 2 12. G
N. undu l a ta 300 2.7 56. 3 39. 7 1.3 3.3 31 . 0 26. 7 3.7 20 . 7 14 . 7
0\eandra distenta 1 84 4.3 59. 8 35. 9 10. 9 35.3 33 . 7 8.1 1 0 . !) 1.1
Phymatodcs sco lopcndria 226 5.7 52 , 6 41 . 6 30. 1 40.7 4.9 16.4 8. 0
Platycerium angolense 2 22 3.1 19.4 60 . 8 16. 7 15.8 9. 5 0. 9 47 . 7 24 . 3 1.8
P . stemaria 142 25.3 60. 6 14. 1 8.4 16. 9 5. 6 39. 4 29. 6
Pyrrosia mechowii 40 30. 0 70.0 5. 0 40. 0 5. 0 15.0 30 . 0 5. 0
Tectaria angelicifolia
T . fernandensis
V i t kuia guineensis 247 6. 1 12. 6 44. 1 29. 9 7.3 1.6 3.6 20, 2 38. 9 8. 1 9. 7 1 5. 8 2.0
Xiphopteris oosora 46 52 . 2 39. 1 8. 7 21 . 7 5 4. 3 23. 9
X . serru lata
X. villosissima 60. 0 40 . 0 lOO. 0

Total no. of observations 3000

Other vas e . epiphytes 11 Ill IV V A B c A B c A B c


Begonia mannii 10 1 00 . 0 1 00 . 0
B. polygonoides 3 1 00 . 0 1 00 . 0
B. rubro-marginata 72 1.4 45. 8 48 . 6 4.2 8.3 27 . 8 27 . 8 1.4 18. 1 16. 7
Calvoa monticola
C. trochainii 50 . 0 16. 7 33 . 3 16. 7 33 . 3 16.7 33 . 3
Medinilla mannii 67 4.5 31 . 3 64 . 2 14. 9 38. 8 17 , 9 6.0 . 17.3 4 .5
Peperomia fernandopoiana 148 16. 9 19. 6 30 . 4 30. 4 2. 7 11. 5 12, 2 3.4 15. 5 22,3 16,2 6.8 6.8 5. 4
P . molleri
P . rotundifolia 11 . 1 66 . 7 22. 2 22 . 2 55. 6 22 . 2
Preussiella cheva \ i e ri 39 2.6 53 - 8 35. 9 7.7 28. 2 41 . 0 12 . 8 17 . 9
P . kamerunensis 34 11.8 50. 0 35. 3 2.9 29. 4 47 . 1 23 . 5
Remusatia vivipara lOO. 0 1 00 . 0
Rhipsalis baccifera 109 0.9 14. 7 54 , 1 30 . 3 19.3 15. 6 3.7 27 . 5 27 . 5 6 .4

T o ta l n o . of observations 503

The dis tribution o f the records between ( 1 ) the di fferent sections of the phorophytes and ( 2) the different light and substrate classes each refer to
a percentage of the total number of records. A = Bark, B ; Minor humus depos i ts , C ; Large humus deposits.

Acta phytogeogr. suec. 59 _


Table 27 b. Total num be r o f records, according to the close observation method, of orchius in the high forest. (Explanation in Table 27 a . )

No. o f Section o f llie Heavy Open Full


observ. pho rophyte s shade shade

II Ill IV V A B c A B A n

Orchids

Al!rangis bi loba 27 51. 8 48. 1 ?? , 7.4 44.4 2o. 9


A . laurentii 23 39. 1 30.4 30 . 4 1 00 . 0
Ancistrochilus rothschildianus 53 56 . 6 35. 8 7.5 11.3 4 !1 . 1 9. 4 9.4 20. 7
Ancis trorhynchus capitatus 22 13. 6 45.4 31. 8 9. 1 4o . 9 21 . 3 4. ,; 21 . 3
A . cepha lotes 57 . I 42. n 100. 0
A. clandestinus
A. recurvus
Angraecopsi s elliplica 25 48. 0 52.0 4.0 2� . 0 6�.0
Angraec um bi rrimcnse 29 48. 3 44 . 8 G.9 17 . 2 G2. 1 13. � G . !l
A. che va lieri 19 73.7 26.3 57 . 8 10. 5 1 5 . tl 1 5 . tl
A. classensii
A . di s ti chum 54 29. 6 53 . 7 16.7 H. 5 11.1 3.7 4� . 1 14 . � 3.7
A. podochi loides
A. subulatum 50. 0 50. 0 16.7 66 . 7 11 ; . 7
Ansellia africana
Bo\usiella ta lbolii 73 4.1 24 . 7 71. 2 4. 1 !)] . ti 4. 1
Brachycorylis kalbre_veri
Bulbophyllum barbigerum 21 4. 8 1 9. 0 61 . 9 14.3 33 . 3 !J . 5 47 . 6 !J, !j
B. bifarium
n . bufo 67 71.6 28.4 1 o. 4 52. 2 1!1. 4 :1 . o 11 . 9 3. o
B. buntingii 58 10. 3 7 5 . !1 13. 8 20 . 7 6. 9 6 !1 . 0 3.4
B. calamarium 2 1 00 . 0 lOO. 0
B. cochleatum 94 54 . 3 45.7 37 . 2 36. 2 !l . 6 13. � 3.2
B . cocoinum 34 52 . 9 44 . 1 2. 9 8. 8 47 . 0 2 !1 . 4 2 . !1 11.8
B . congolanum 19 31 . 6 68 . 4 57 . !) 4 ::! . 1
B. distans 33 42 . 4 4 ti . :. 9. 1 3. 0 42. 4 24 . � 30. 3
B. fa l catum 1 00 . 0 66 . 7 33 .3
B . fla vidum 33 . 3 66 . 7 16.7 50 . 0 33 . 3
B . imbricalum
B . inflatum
58. 3 ?? ?
B . inlerlexlum 36 19.4 22. 2 G !J . <I �-3
B . josephii 21 4.8 61 . 9 33 . 3 D0 . 5 9.5
B. lindert 810 0.5 44 . 3 45. 5 9. 6 34 . 7 6. 7 4 �. 5 !l. 9 0.1
B . lucifugum 76 44.7 47 . 4 7. 9 40 . � 47 . 4 :.. 3 6. 6
B . lupulinum 11 18.2 63 . 6 1 8. 2 18.2 54 . ;; 27 . 3
B . magnibracteatum 1 00 . 0 1 00 . 0
B . maximum 31 1 9. 3 67 . 7 1 2 . !1 61 . 3 3K. 7
B. melanorrachis lOO. 0 1 00 . 0
B. nigritianum
B . oreonastes 1 1 85 2. 1 23 . 7 54 . 3 19. 9 33. 6 G.7 l.J 44 . 4 1 � . !1 1.0
B . pavimentatum 3 33 . 3 66.7 l OO . 0
B. phaeopogon 11 36. 4 63 . H 1 ::1: . 2 l "'i . � 1 1; . � 4 :) , 4
B. recurvum
B. rhizophorac
B . sallatorium 28 7.1 64 . 3 28. 6 42 . H 50 . 0 7.1
B . scariosum 1 99 G. 5 45. 2 4�. 2 !1 . 5 12.1 J.n 32.2 42 . � :J . o
B. schimperanum 45 13. 3 55. 5 31 . 1 2 4. 4 71 . 1 4. 4
B . schinzianum 86 31 . 4 59.3 9.3 40.7 5. � 3�. 4 1 :. . 1
B . w�hleri H 16.7 56. 7 16.7 53 . 3 20 . 0 20 . 0 !l . 7
Calyptrochilum chris lyanum 33 60. 6 24 . � 15. 1 24 . 2 75. .-,
C. cmarginatum 50 . 0 50 . 0 ;,o . o so . o
Chamaeangi s vesicata 18 5.3 94 . 7 5.3 52 . 6 42 . 1
Cyrtorchis arcuala 176 50 . G 44 . D 4. 5 27. 3 34 . 7 4.5 I !I . U Io . 2 3. 4
C . aschcrsonii 25 32.0 52 . 0 16. 0 12.0 �- 0 l li . 0 li4 . 0
C. monleiroac
Diaphananthc bidcns 35 2. 9 28. 6 57 . 1 11.4 31.4 11 . 1 o.7 31 . 4 14.3
D. densiflora 1 00 . 0 50 . 1) 511 . 0
D. pellucida 66 . 7 33 . 3 )• ? ? 66. 7 11.1
D. rutila 1 00 . 0 1 00 . 0
Eurychone rothschi ld i ana
Genyorchis pum ila
Graphorchis lu rida 13!J 4.3 24. 5 57 . 5 13.7 l ti . 7 2. 9 H7 . 6 1 0. 1 0.7
Habenaria leonensis
H . p roc e ra
Liparis caillei 1 00. 0 1 00 . 0
L . nervosa
Lis trostachys pertusa 389 5.9 26. 0 46. 8 21 . 3 2 4 . !> 9,2 55 . 5 10. 3
Nephra ngi s filiformis 1 93 19. 1 60. 6 20. 2 34.7 19.7 38 . 3 7.3
Pleclrelminthus caudatus 66. 7 33. 3 100 . 0
Podangis dactyloceras
Po ly s ta chya adansoniae 268 1 5. 3 44 . 0 40 . 7 17 . 9 1 6. 8 1. 9 25. 0 34 . 3 4. 1
P. affinis
P. dalzielii 93 7.5 49. 5 43. 0 3.2 25. 8 2 6 . !) 44 . 1
P . elastica
P. galeata 56 33.9 55. 4 10.7 7.1 25 . 0 25.0 3. 6 28. G 10. 7
P. laxiflora 21 9. 5 57 . 1 33 . 3 9. 5 38. 1 47 . 6 4.8
P. leonensis 48 2. 1 10.4 39. 6 43.7 4.2 14. 6 37 . 5 14 . 6 16.7 1 2 . !i 4.2
P . microbambusa
P . obanensis 62. 5 37 . 5 75.0 25. 0
P . pani cu la ta 1 61 18.6 6 0 . !) 20 . 5 24. 2 6. 8 14.3 37 . 3 17.4
P . pobeguinii 1 00 . 0 1 00 . 0
P . polychaete 222 3.6 39. 2 42 . 3 14 . 9 6. 3 26. 6 21 . 6 !1 . 0 25. 2 11.3
P. puberula 232 18.4 51 .3 30 . 3 7.3 23 . 1 1.7 13.7 49. 1 :. . 1
P. ramulosa
P. rhodopte ra
P. saccata 85 49.� 30. 6 17 . 6 2.3 42 . 3 3.5 50. fi 3. 5
Polystachya subulala
P . tenuissima 1 00 . 0 50. 0 :5 0 , 0
P . tessal la ta 158 2.5 42. 4 39. 9 15. 1 15. 2 31 . 6 1 !1 . 0 9. 5 1 0> . 8 H . !J
R a ngal! ris brachycerns 100 . 0 lOll. 0
R . muscicola 143 2.1 20.3 58. 7 1H. 9 14.0 4.2
R . rhipsa lisocia
Rhipidoglossum paucifolium
Stolzia repcns 1 00 . 0 50. 0 !'iO . O
Tridactyle anthomaniaca 650 0. 6 24. 5 46. 6 28.3 16.2 4.8 0.3 73 . 7 4.6 O . !i
T. armeniaca 363 6. 1 51 . 8 33. 6 �. 5 41.3 31. 9 7.2 11.0 li . 6 1 . !J
T. bicaudata 25. 0 75. 0 75.0 25.0

T. crassifolia 16 37 . 5 25.0 1 ti . 7 18. 7 25 . 0 ?G.0


T. tridacty\ites 62 4.8 37 . 1 35. 5 22 . 6 24 . 2 43 . 5 G. 4 3.1 21 . 0 1 . li
T. tridentata 53 17. 0 54 . 7 28. 3 7. 5 !JO . 1 . !)
Vanilla c renu l a ta

Total no . of observations 6940


Ecology of vascular epiphytes in West African rain forest 75

of decay. This subdivision of substrates was easy to representing 4 7 species of 2 1 genera. 222 trees were
work with during the recording. investigated according to the close observation
Each epiphyte or group of epiphytes has thus been method and 24 1 by distance observation (Table 26).
recorded in one of the nine possible light-substrate The quantitative figures from this investigation for
classes. The idea was to provide a base for a quan­ each species of epiphyte is presented in Tables 27, 28
titative judgment of each species' dependence of light and 29.
and substrate. However, it should be emphasized It became evident that most species of epiphytes
that many other influences effect the presence and are ± restricted to one or two particular section(s) of
distribution of epiphytes. the phorophyte, while a few occur more evenly in a
number of sections (Fig. 78). An analysis of the
General distribution pattern results from this investigation will later be presented,
The distribution pattern was recorded on 463 in­ correlated to certain environmental influences
dividual trees C> 10 m in height) of the high forest, (Chapter VI).
Table 2 8 a. Total number o f records, according t.o the distance observation m e thod, o f pteridophytes and other vascular epiphytes i n the

high fore s t . �Explanation in Table 27 a. ) .

No . of Section of the Heavy Open Full


observ. phorophytes shade shade sun

II III IV V A B c A B c · A B c

Pteridophytes

Antrophyum immersum 10 30. 0 70. 0 70.0 30 . 0


A. mannianum 75. 0 25. 0 25. 0 50 . 0 25. 0
A rthropteris monoca rpa
A . orientalis
A. palisoli
Asplenium aelhiopicum
A . africanum 19 10. 5 73. 7 15.8 10. 5 42 . 1 47 . 4
A. barteri 26 57 . 7 '! 3 . 1 11. 5 7.7 19.2 30. 8 11. 5 3. 8 30 . 8 3. 8
A. dregeanum
A. geppii 22 4. 5 77.3 18. 2 9.1 90 . 9
A. hemitomum 1 100. 0 1 00 . 0
A. megalura 1 83 20. 2 62 . 3 17 . 5 1.1 22. 9 10. 9 3.3 32 . 8 29. 0
A. variable v . paucijugum 71.4 28. 6 1 00 . 0
Davallia chaerophylloides 63 20. 6 41 . 3 31 . 7 6.3 22 . 2 31 . 7 3.2 15.9 22. 2 4 . t>
Drynaria laurenlii 929 2.8 42. 6 42 . 1 12 . 5 0. 2 28.3 14 . 8 2.3 45. 5 8. 8
Elaphoglossum barteri 6 16. 7 83 . 3 16.7 16.7 66. 7
E. chevalieri
E. isabelense 76 7.9 2 7. 6 57 . 9 6. 6 14 . 5 46. 1 5. 3 15.8 1 5. 8 2. 6
E. kuhnii
E. salicifolium 144 9.7 14. 6 59. 0 16.7 34 . 0 20. 8 31 . 9 13 . 2
Lomariopsis guineensis 1 00 . 0 100. 0
Loxogramme lanceolata 14 7.1 42 . 9 35. 7 14.3 28. 6 71.4
Lycopodium mildbraedii
L. warneckei 39 5. 1 76. 9 17 . 9 7.7 10. 3 5. 1 10.3 59 . 0 7.7
Microgramma owariensis 47 t> . 5 51 . 1 40.4 36 . 2 42 . 6 10 . 6 2.1 6.3 2.1
Microsorium punctatum 178 7.3 50 . 6 39. 3 2. 8 14. 0 44.3 12 . 4 5. 6 19.7 3.9
Nephrolepis biscrrala 42 64 . 3 35.7 7.1 21 . 4 4 . l! 45. 2 21 . 4
N. undulata 342 1. 5 46. 5 47 . 7 4. 4 6.4 36. 3 30 . 7 5. 6 14 . 6 6.4
Oleandra distenta 352 4. 8 64 . 8 30 . 1 0. 3 0. 6 17 . 0 39. 2 35. 8 2. � 3.4 1.1
Phymatodes scolopendria 174 9. 2 57 . 5 28 . 7 4.6 21 . 8 53 . 4 5. 7 6. 9 12 . 1
Platycerium angolense 264 3. 8 24. 6 53 . 4 1 8 .2 24 . 2 17 . 0 1. 1 44 . 7 12. 9
P . stemaria 126 0.8 11. 9 61 . 9 25. 4 15.9 25. 4 0. 8 48 . 4 9. 5
Pyrrosia mechowii 16 62 . 5 37 5 37. 5 62. 5
Tectaria angelicifolia 75. 0 25. 0 50 . 0 50 . 0
T. fernandensis
Vittaria guineensis 265 2. 6 8.7 28. 7 49. 8 10.2 3. 0 16. 2 45. 3 4.9 13. 2 16. 2 1.1
Xiphopteris oosora
X. serrulata
X. vilosissima

Total n o . o f observations 3 3 54

Other vase . epiphytes II Ill IV V A B c A B c A B c

Begonia mannii 1 00 . 0 16.7 83 . 3


B. polygonoides 16 68.7 31 . 2 6.2 62 . 5 31. 2
B. rubro-marginata 70 35. 7 58. 6 5.7 2.9 27 . 1 35. 7 1.4 11 . 4 21 . 4
C alvoa monlicola
C . trochainii 50 . 0 50. 0 50 . 0 50 . 0
Medinilla mannii 144 1.4 32. 6 55. 6 10. 4 2. 1 15. 3 19.4 11 . 8 22. 2 29. 2
Pcperomia fernandopoiana 121 27 . 3 33 . 9 26. 4 12 . 4 �. 1 32 . 2 2. 5 5. 0 38. 0 9.9 0. 8 2. 5
P. molleri
P . rotundifolia 10 20. 0 70. 0 10. 0 50. 0 30 . 0 20. 0
Prcussiella cheva l i e ri 27 74. 1 25. 9 11 . 1 51 . 9 7.4 29. 6
P. kamerunensis 32 78. 1 21 . 9 12 . 5 50 . 0 9.4 28. 1
Remusata vivipa ra
Rhipsalis baccifera 170 24 . 7 58 . 8 16. 5 4.1 21 . 8 10. 0 12 . 9 32 . 4 18.8

Total n o . of observations 598

A cta phytogeogr. suec. 59


" Table 28 b. Total number o f records, according to the distance observation method, of orchids in the hi�h fores t . (Explanation in Table 27 a . \

No. o f Section o f the Heavy Open Full


observ. phorophytes shade shade

I! lil IV V A B A B A B c

Orchids

A�rangis bi loba 64 25.0 32 . 8 42 . 2 1 5. 6 21 . 9 9.4 34.4 15. 6 3.1

A. laurentii 19 57 . 9 42. I 21. 1 73 . 7 5.3


Ancistrochilus rothschildianus 21 42. 9 57 . 1 4.8 85.7 9. 6
Ancis trorhynchus capitatus 20 20 . 0 25. 0 50 . 0 5.0 30 . 0 35. 0 5. 0 20 . 0 10 . 0
A. cephalotes 54 29. 6 24. 1 40.7 5. 6 35. 2 1.9 63 . 0
A. clandestinus 1 00 . 0 66 . 7 33 . 3
A. recurvus
Angraecopsis e lliptica 8 100 . 0 12. 5 87 . 5
Angraecum birrimense 26 11. 5 84 . 6 3. 8 11.5 38. 5 46. 2 3. 8
A. chevalier\ 11 63 . 6 36.4 27 . 3 54 . 5 1M. 2
A. c\assensii 2 50 . 0 50. 0 l OO . 0
A. distichum 59 10.2 ·2 0 . 3 64 . 4 5. 1 20. 3 15.3 45. � 1�.6
A. podochiloides
A. subulatum 1 00 . 0 25. 0 50 . 0 25. 0
Ansellia africana
Bolusiella Lalbotii 76 43 . 4 56 . 6 11. 8 72.4 l!' i. :-1

Brachycorytis kalbreyeri lOO. 0 1 00 . 0


Bulbophyllum barbigerum 1 00 . 0 33 . 3 66 . 7
B. bi fa rium
B . bufo 189 41. 8 49.7 8. 5 13 . 2 38. 1 20 . 7 5.3 20. 1 � . (;
B. buntingii 56 8. 9 57 . I 34 . 0 8. 9 76. 8 14.3
B . calamarium
B. cochleatum 72 2.8 45. 8 51 . 4 20. 8 48. 6 15.3 2.8 11 . 1 1.4
B . cocoinum 1 62 58. 1 38.7 3.2 8. 1 1.6 3_3 . 9 45 . 2 4.8 6 . 0,
B . congolanum 19 52 . 6 47 . 4 84 . 2 l !'i . 8
B. distan s 3 66.7 33 . 3 33 . 3 33.3 33 . 3
B. fa\ca tum
B. flavidum 20 60. 0 40 . 0 �0. 0 �o.o Gu . o
B. imbricatum 100. 0 1 00 . 0
B . inflatum
B. intertextum 74 58. I 41 . 9 21. 6 10. 8 U
! .4 16.2
B . josephii 5 20 . 0 80 . 0 40. 0 40 . 0 �0.0
B. linderi 430 14. 2 61 . 6 24. 2 17.0 5. 3 63 . 0 14.4 0.�
B . \ucifugum 48 37 . 5 60. 4 2. 1 31 . 2 43 . 8 25. 0
B. \upu linum
B . magnibracteatum
B. maximum 1 04 I.0 13. 5 66. 3 I 9. 2 26. 0 3. 8 57 . 7 1 2 . !}
B. melanorrachis
B. nigritianum
,., ·) 1.7
B . oreonastes 828 1.4 11.7 52 . 4 34.4 16. 5 ;) . l:S 0. 2 :)3 . ;)
B. pavimentatum
B . phaeopogon
B . recurvum 62. 5 37. 5 37 . 5 62 . 5
B . rhizophorae
B . saltatorium 14 7.1 7.1 85. 7 64 . 2 7. I 28. (i

B . scariosum
B . schimperanum
B . schinzianum 19 73 . 7 26. 3 15. 8 5.3 63. 2 1 5. �
B . winkleri 15 20 . 0 40 , 0 40. 0 60. 0 6.7 33 . 3
C a lyptrochi \urn christyanum 70 11.4 43 , 0 41 , 4 4.3 23. 0 H.3 . �.G 37 . 1 17.1
C . em a rginatum �5.0 75. 0 1 00 . 0
Chamaeangis vcsica t...1. 45 6. 6 4.4 80,0 8. �) 17 . 8 11.1 60 . 0 11.1
Cyrtorchis arcuata 186 1.1 52 . 7 45. 2 1.1 24 . 7 32 . 3 1 5. 6 IG. 1 10. 8 1.6
C . aschersonii 17 11. 8 23 . 6 47 . 1 17. 6 23 . G 5 , !! 5 8 . :S l l . ti
C. monleiroac
D iapha nanlhe bidens 24 8.3 37 . 5 45. 8 8.3 8.3 29. 2 1� . 5 :. n . :J 1 � . :)
D . densiflora
D. pellucida 27 14. 8 55 . 6 2 !J . G 7.4 66 . 7 7. 4 3.7 11. I 3.7
D . rutila
Eurychone rothschildiana
Genyorchis pumila
Graphorchis lurida 2 57 0.8 17. 9 60.7 20 . 6 16.0 3. 5 71.2 8. D 0.4
Habenaria leonensis 2 50 . 0 50 . 0 100. 0
H . procera
Liparis caillei
L. nervosa
Llstrostachys pertusa 455 5.3 24.6 52 . 7 17.4 21. 1 14. 9 0.7 45. 1 17. 6 0.7
Nephrangis filiformis 306 7.5 69 . 3 23 , 2 18. 0 2.0 74 . 2 5. 9
Plectrelminthus c audatus 20 75. 0 25. 0 5.0 5.0 80 . 0 10. 0
Podangis dactyloc eras
Polystachya adansoniae 229 17.0 50 . 7 32 . 3 10.5 23 . 6 1.7 23 . 1 39.3 1.7
P . affinis 1 100 . 0 100 . 0
P. dalzielii
P . elastica
P. galeata 87 58 . 6 37 , 9 3.4 3.4 48. 3 20 . 7 1.1 18. 4 8. 0
P. lru<iflora 10 80 . 0 20 . 0 50. 0 50 . 0
P . leonensis
P. microbambusa
P . obanensis 50. 0 50 . 0 1 00 . 0
P. paniculata 259 37 . 1 50 ' 6 12. 4 32 . 8 13. 9 0.4 37 . 8 14.7 0.4
P. pobeguinii
P. polychaete 1 85 31. 9 47 . 0 21 . 1 9.2 18.4 22. 2 10.3 28. 1 11.9
P. puberula 240 22. 5 54 . 2 23 . 3 13.7 33 . 3 5.0 22.1 23 . 8 2.1
P . ramulosa
P . rhodoptera
P. saccata 13 15.4 76. 9 7.7 69. 2 30 . 8
Polystachya subulata
P . tenuissima 33, 3 66 . 7 1 00 . 0
P. tessallata 244 0. 8 38. 1 51 . 2 9. 8 12.7 29. 9 23 . 8 9. 0 13. 5 11.1
Ranga�ris brachyceras
R . musctcola 1 80 25. 0 63 . 3 11. 7 11.7 8. 9 13 . 3 49.4 16.7
R . rhipsalisocia 19 63 . 2 36. 8 94 . 7 5.3
Rhipidoglossum paucifolium
Sto lz ia repens
Tridactyle anthomaniaca 766 9. 0 50 . 0 41. 0 14 . 6 G . !l 1.3 64 . ;, 11.1 o.n
T. armeniaca 352 2.8 53 . 4 41. 8 2.0 20. 7 4 3 .8 17.G 7.7 9.9 0. �
T. bicaudata
T. crassifolia 12. 5 75.0 12.5 12. 5 62 . 5 25.0
T. tridactylites 50 8.0 52, 0 30. 0 10, 0 10. 0 40. 0 4.0 16. 0 24. 0 6.0
T. tridentata 43 7. 0 7 0. 0 23 . 0 11.6 2.3 79.1 7.0
Vanilla crenulata

Total no. of observations 6551


Ecology of vascular epiphytes in West African rain forest 77

The distribution on the phorophytes of the various previously described. A rather clear distribution
epiphyte species showed somewhat different results pattern for most species of epiphytes was obtained
depending on observation method. Compared to the (Fig. 79), no matter if there were few or many
results from the close observation method many species present. The distribution pattern also seemed
species showed a ± marked displacement towards to be similar between various species of phorophytes.
the outer parts of the crown in the distance observa­ Thus, the transects along the large branches seem
tion. This could be anticipated since the epiphytes in to represent a rather fast and informative way to get
the distance _ observation are more easily noticed a good insight into the distribution pattern for the
when growing on the outer parts of the branches. various species. This method may preferably be used
Large branches from several phorophytes were when there is a limited number of phorophytes to be
examined according to the transect technique examined.

Table 2 9 ta l number of
. To records, according to the occasional observation m e thod, o f some epiphytes i n e
th h igh fore s t . (Explanation in Table 2 7 a . )

No . of Section of the Heavy Open Full


observ . pho r ophytes shade shade sun

I! III IV V A B c A B c A B c

t
P e ridophyt es
1
A rthropteris monocnrpn 46 21 . 7 54 . 3 15. 2 �.7 26. 1 -H . 8 21 . 7 4.3
00 . 0

en u c t
A . pa li so ti 1 00 . 0
A spl i m a h i opic u m 23 4.3 39. 1 52 . 2 4.3 8.7 43 . 5 26. 1 8. 7 13.0

u
A . hemitomum 10 20. 0 80. 0 60. 0 20 . 0 20. 0
A. variable v . pa u c ij gum 54 59 . 3 40. 7 66. 7 25. 9 7.4
Elaphoglossum kuh n i i 26 88. 5 11.5 11 . 5 76. 9 11. 5

Lomnriopsis guineensis 19 52 . 6 47 . 4 73. 6 15. 8 5.3 5. 3


Tectarin nngclicifoli:� 36 GD. 4 30. 6 30 . 6 47 . 2 22.2
T. fcrnandcnsis 35 �0. 0 20 . 0 14. 3 28.6 57 . 1
Xiphoptc ris s c r ru l n t a 1 00 . 0 1 00 . 0
X . \'i losi s s i ma 16 6.2 G� . G· 25. 0 G. 2 6. 2 31.3 43 . 7 18. �

Orchids

Ancis t ro rh y nc hus c landes tinus 7 5. 0 25. 0 7 5. 0 25. 0

Angraecum c l n s s e n s i i 46 30. 4 69. 6 54 . 3 45.7


A . podochi ! a i d e s 37 1 6. 2 78.4 5. 4 10. 8 81 . 1 8.1
A n sellia africana 32 9. 4 53 . 1 37 . 5 15. 6 18.7 6.2 12. 5 46. 9
Brachycorythis kalbreyeri 65 69. 2 15. 4 10. 8 4.6 1.5 35.4 50 . S 3. 1 9. 2
Bu lbophyll u m bifarium 17 58 . 8 · 29. 4 11.s 70. 6 11. 8 5. 9 11.8
B. calamarium 25.0 50. 0 25. 0 50 . 0 37 . 5 12. 5
B . falcatum 28 7. 1 50. 0 35 . 7 7- 1 17 . 9 53 . 6 14 . 3 14 . 3
B. imbricatum 23 4.3 17 . 4 73 . 9 4.3 21 . 7 52 . 2 8.7 17 . 4
B . i nfla tum 44 9. 1 56 . 8 27 . 3 6.8 20 . 4 47 . 7 4.5 27 . 3
B. magn i brac tca tum 11 36. 4 45.4 18. 2 18. 2 18. 2 54 . 5 9. 1
B. melanorrachis 16 62 . 5 37 . 5 62 . 5 37 . 5
B. nigrilianum 17 23 . 5 17. 6 58 . 8 11.8 58 . 8 17 . 6 11.8
B. pavimcntntum 33. 3 66 . 7 83 . 3 16.7
B. recurvum � 62 . 5 37 . 5 37 . 5 62 . 5
B. rhizophorae 10 20. 0 20. 0 60 . 0 20 . 0 10.0 70 . 0

c
Calyptrochi turn emarginntum 40 30. 0 57 . 5 7.5 5. 0 15. 0 2. 5 77 . 5 5. 0
Cyrtorchis mo n t i rone 14. 3 71.4 14.3 28. 6 71.4
D iaphananthc densinora 43 48. 8 46. 5 4.7 83 . 7 7. 0 7. 0 2.3
D . rutila 120 61 . 7 33 . 3 5.0 23 . 3 0. 8 73. 3 2. 5
E u rychone rothschi ldiana 27 66. 7 33 . 3 59 . 3 40. 7
Genyorchis pum i l a 16 25. 0 25. 0 43. 7 6.2 12. 5 18.7 25. 0 37 . 5 6.2
Habcnaria leonensis 13 15. 4 84 . 6 7. 7 61 . 5 23 . 1 7.7
H . procera 35 11.4 68. 6 17 . 1 2 . !) 22. 9 54 . 3 5. 7 17 . 1
Lipa ris cni l lc i 26 76. 9 11. 5 11.5 7. 7 69. 2 23 . 1
L . nervosa 29 62 . 1 31 . 0 3.4 3.4 6. 9 48. 3 24 . 1 6. 9 13 . 8
Podangis dnctyloccras 12 83 . 3 16.7 33 . 3 8. 3 41 . 7 16. 7

.0
Polys tachyn nffinis 76 27 . 6 63 . 1 6. 6 2. 6 51 . 3 13 . 2 30.3 5. 3
P . e l astica 1 00 . 0 1 00
P . micmbnmbusa 21 38. 1 57 . 1 4 . 8 14, 3 19.0 66 . 7
P . pobeguinii 33 3. 0 24 . 2 42 . 4 15. 2 15. 2 9. 1 21 . 2 18.2 48. 5 3 . 0
P. ramulosa 16 62. 5 37 . 5 12. 5 18.7 50 . 0 18.7

55
P . rhodoptcra 31 74 . 2 25. 8 3.2 22 . 6 35. 5 3.2 9.7 25. 8
P. s ubu l a ta 22. 2 .6 22 . 2 22 . 2 33. 3 44 . 4
P. Lt=1iui s sim a 14 35. 7 57 . 1 7.1 28. 6 21 . 4 50 . 0
Rangallris brachycerns 18 5.6 22.2 66.7 5. 6 16.7 5.6 50. 0 27 . 8
"
Rhipidoglo s s u m paucifolium 27 51 . !J 37 . 0 11.1 37. 0 11. 1 37 . 0 14. 8
Stolzia repens 15 6.7 13. 3 60 . 0 20 . 0 80. 0 20. 0
Tridactyle bicaudaW 57 26 . 3 21. 1 47 . 4 5.3 22 . 8 54 . 4 3. 5 3.5 15.8
Va n i l la c rcnu l a tn 83 . 3 16.7 33 . 3 66 . 7

Other vas e . epiphytes

Calvoa monticola 28 85.7 14.3 35. 7 50 . 0 7.1 7. 1


C. L ro c ha i n i i 46 34 . ti 26. 1 39. 1 6.5 54 . 3 32 . 6 2 . 2 4.3
Peperomia m o l lcri 25 92 . 0 8. 0 8.0 32. 0 44 . 0 8. 0 8. 0
'
Rcmusali:t \' i v i pa r a 32 34. 4 62. 5 3. 1 18. 7 53 . 1 9. 4 18.7

Total n o . o f observa tions 147�

A cta phytogeogr. suec. 59


78 Dick Johansson

Asp l e n i u m Asp l e n i u m Asp l e n i u m Loxogra m m e Epiphyte communities


ba rteri geppii m ega l u ra l a nceol ata

Terminology
The frequently observed aggregation of epiphytes
('arboreal gardens' according to Oliver 1 930 :2) in
limited areas on the phorophyte is of interest not
only because of the information they provide as to
Polystachya B u l bophyl l u m B o l u s i e l l a Polystachya common environmental requirements for various
ra m u losa b u fo t a l bot i i leonensis
species but also as a source in the analyses of the
colonization and development of the epiphytic flora
in general. The terminology that has been used in the
classification of epiphyte communities is confusing.
The terms 'epiphyte societies' (Oliver 1 93 0 : 1 6)
and ' association' � Went 1 940 :86) have been used.
C a lvoa B eg o n i a R h i psa l i s Pepero m i a Epiphytic association and epiphytic sociation were
monticola rubro- m a rg i nata ba ccifera· sp.
proposed by Omura ( 1 95 3). Hosokawa ( 1 95 3 ,

vl
IV
1 954 a , b , c , 1 968) has created a new terminology
Ill
for the community units of epiphytes. From the
11 viewpoint of the ecosystem concept he does not
I . agree with the use of phytocoenosial units for
____,

50 100'1, classifying epiphyte communities. Instead he has


suggested the terms epies, epilia and epido, for
Fig. 78. Occurrence of some epiphytes on the various types of epiphyte communities (Hosokawa
phorophytes distributed among the different sections. 1 968 :483).
(From Table 27, Polystachya ramulosa from Table 29.)

R a p h i dophora afri c a n a - • �10 'stands'


E l a p h o g l o ss u m i s a be l e nse -- - 4-9 'sta nds'
Vittaria g u i neense --- - 1 -3 ' st a n d s '
Angraec u m b i rr i m e n s e -----­

Ca lvoa troc h a i n i i
M ed i n i l l a m a n n i i
Pepero m i a sp.

O l e a ndra d istenta -
B u l bo p h yl l u m coc o i n u m -

B u l bophyl l u m i n tertext u m - -

A n g ra ecopsis e l l iptica
P h y m a todes scolopendria -
Ang ra e c u m disti c h u m
Po lystachya tessa l l ata
Aspl e n i u m meg a l u ra
D ry n a ri a l a u re n t i i
P l a t y<:e r i u m a n golense -
Asplen i u m m eg a l u ra
Tridactyle a rm e n i a c a -
1/i tt a r i a g u i ne e n s i s
Polyst a chya g a leata
A n g r a e c u m d i st ic h u m
Cyrtor c h i s a rc u a t a -
Polystachya tessa l l at a
Polystac hya p u b e r u l a -
R h i ps a l i s baccifera
R h i p s a l i s baccifera - =
Dryn aria l a u re n t i i
G ra ph orc h i s l u ri d a
G ra phorc h i s l u rida
B ol u si e l l a ta l bo t i i
O l e a nd ra diste nta - -
16 10 15
������ m �------�--����- m

Fig. 79. Number of epiphytes and their distribution along tadeniastrun africanum, illustrated according to the
a branch of: Left : Parinaria excelsa, illustrated according 'transect' technique. Yekepa 500 m.
to the 'transect' technique. Seka Valley 600 m. Right : Pip-

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 79

To avoid adding to the already confusing ter­ when a group of three or more species form a unit,
minology of epiphyte communities, they will here and when the distance from two of them does not ex­
simply be called epiphyte communities (e.c.). To ceed 0.5 m to the third. The root systems and the
separate them from each other they are named by stems are often heavily intermixed, to such a degree
the dominant species ('cover') and the most frequent­ that the plants often grow on top of each other.
ly occurring species in the community. This unit Epiphytes that colonize the vegetative parts of
�hould be homologous to the term 'epilia' proposed other epiphytes are called hyper-epiphytes by Tixier
by Hosokawa. ( 1 966) and 'secondary' epiphytes by Hosokawa. The
delimitation of an epiphyte community in most cases
Methods is easy to do, since the grouping of plants is rather
The technique used in the investigation and descrip­ distinct (Fig. 79). There are some spectacular excep­
tion of terrestrial plant communities is impossible to tions from this rule. Drynaria laurentii may often
use under the circumstances of growth and oc­ cover stretches of more than five meters length. This
currence which the epiphytic plants exhibit. The extended distribution will of course connect a larger
knowledge of the epiphytic plant communities is number of plants to this particular epiphyte.
based on records from close and distance ex­ However, most epiphyte communities occur in a
aminations. The interest has been concentrated on limited area with no other plants nearby. This is the
the presence of the various species rather than on case particularly for certain communities formed by
their number. epiphytes that are attracted to humus deposits.
An epiphyte community is considered to exist Seven different epiphyte communities are recognized

Table 30. The number, distribution on the phorophytes and species composition of the epiphyte communities examined. (The distri-
bution is given as a percentage of the total number of communities and the species composition according to close and distance ob-
servations.)

Epiphyte community No of Distribution on the phorophyte No. of species


commun. Sections Pterid. Orchids Others Total Grand
total
ll Ill IV V Cl Di Cl Di Cl Di Cl Di

Angraecum birrimense-
-
Raphidophora· africana 65 43. 1 5 6.9 13 10 19 15 6 5 38 30 39

Asplenium africanum-
Peperomia sp. 13 3 8. 5 1 5 .3 46.2 3 3 0 3 3 6 7 9

Oleandra distenta-
Tridactyle armeniaca 1 04 6.8 70.2 23. 1 13 17 19 20 12 10 44 47 51

Microsorium punctatum-
Vittaria guineensis 63 6.8 70.2 23. J 6 5 13 11 7 5 26 21 29

Drynaria laurentii-
Asplenium megalura 1 90 5 . 3 34.7 48.9 1 1 . 1 13 8 37 25 8 8 58 41 62

Platycerium stemaria-
Nephrolepis undulata 54 7.4 5 1 .8 40.7 7 5 9 7 4 3 20 15 22

Platycerium angolense-
Nephrolepis undulata 50 30.0 1 2.0 58.0 6 4 7 5 3 3 16 12 17

Asplenium dregeanum-
-
Peperomia sp. 42 88. 1 1 1 .9 9 5 5 19 19

Tridactyle tridactylites-
-
Medinilla mannii 60 3 . 3 7 6 . 7 20.0 11 8 8 7 5 3 24 18 26

Bulbophyllum scariosum-
Polystachya dalzielii 25 1 2.0 64.0 24.0 11 8 6 5 3 4 20 17 21

A cta phytogeogr. suec. 59


80 Dick Johansson

below 1 000 m alt. and three more above that used (Table 3 1 ). There are, however, some excep­
altitude. tions. The 0/eandra distenta - Tridactyle armeniaca
The investigation includes a total of 666 'com­ and A ngraecum biDrimense - Raphidophora africana
munity aggregations'. Their distribution on the communities, built up by more than five species
phorophyte and the total number of species recorded each, were more common than those with a lesser
in the various communities are given in Table 30. number of species. The possible explanation to this
The ten most frequently occurring species in each could be that a more or less simultaneous coloniza­
community are presented in Table 3 2 . tion by several species is taking place when the com­
munity forming species has created a suitable en­
Abundance vironment. Both these communities are found in the
Like epiphytes. in general the epiphyte communities lower parts of the phorophyte, and could possibly
are more abundant in older trees. A total of 25 'com­ catch seeds or seedlings washed down by heavy rain­
munity aggregations' were found on the 1 1 8 trees fall.
that were examined in the three sample plots (Tables
1 4, 1 5 and 1 6). In relation to the abundance of the Epiphyte communities at altitudes below 1000 m
dominant species most communities were scarce. Angraecum birrimense-Raphidophora africana e.c.
0/eandra distenta, however, was found just as fre­ This community is most common around the first
quent in a community formation as in a single state. ramification on trees which have a rich supply of
As would be expected the most common number of light in that section. This means that the community
species in the individual communities is three, which will appear mostly in secondary forests or on trees
are the lowest possible according to the definition bordering rivers or creeks, or along tracks or roads.

Table 3 1 . The number of species in the different epiphyte communities, expressed as a percentage of the total number of records.

Community Method of No No. of species


examination examined
3 4 5 > 5

Angraecum birrimense - Close 36 8.3 9.0 22.0 55.5


Raphidophora africana Distance 29 1 3 .9 1 0. 3 1 7. 2 58.6

Asplenium africanum - Close 5 40.0 40.0 20.0


Peperomia sp. Distance 8 50.0 3 7.5 1 2. 5

Oleandra distenta - Close 26 26.9 1 9. 2 7. 7 46.2


Tridactyle armeniaca Distance 78 25.6 23. 1 1 1.5 39.8

Microsorium punctatum - Close 27 63.0 1 8. 5 7.4 1 1.1


Vittaria guineensis Distance 36 50.0 5.6 22.2 22.2

Drynaria laurentii - Close 62 45. 1 9.7 1 1 .3 33 .9


Asplenium megalura Distance 1 28 50.8 1 5 .6 1 4. 1 1 9. 5

Platycerium stemaria - Close 36 55.6 8.3 8.3 27.8


Nephrolepis undulata Distance 18 50.0 1 1. 1 5.6 33.3

Platycerium angolense - Close 40 72.5 2 5 .0 2.5


Nephrolepis undulata Distance 10 80.0 20.0

Asplenium dregeanum - Close 42 66.6 23.8 4.8 4.8


Peperomia sp. Distance

Tridactyle tridactylites - Close 13 46. 1 30.8 23. 1


Medinilla mannii Distance 43 60.5 27.9 2.3 9.3

Bulbophyllum scariosum - Close 7 5 7. 1 28.6 14.3


Polystachya dalzielii Distance 18 50.0 27.8 16.7 5.5

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 81

Fig. 80. Oleandra distenta - Tri­


dactyle armeniaca epiphyte com­
munity. Grassfield 5 5 0 m.

It gives an impression of an irregular jumble of her­ development of this community certain orchids may
baceous climbers, lianas and trailing epiphytes. grow between the rhizomes, but later as debris is ac­
The community is delicately balanced. An in­ cumulated they become fewer and fewer.
crease in light and evaporation will change this com­ This community is not subject to the extreme
munity leaving only the Diaphanthe orchids, a variation in microclimate that exists further out on
decreasing light supply will lead to an overgrowth by the branches, and the light intensity is naturally
the vigorous Raphidophora africana. lower. These conditions are also reflected in the com­
position of the species in the community. A high
number of pteridophytes and other vascular
Asplenium africanum - Peperomia sp. e.c.
epiphytes (non-orchids) are typical. Note that no
Asplenium africanum that is rather uncommon,
orchid species except Tridactyle armeniaca is found
forms an accumulation of humus around its attach­
among the 10 most common species (Table 3 2). This
ment to the substrate in a similar way as
community is composed of a mixture of humus­
Microsorium punctatum. Only 1 3 communities have
preferring species and species tolerating different
been found and examined, 8 with telescope and 5 on
kinds of substrate. Orchids with preference for rich
felled trees. With so few observations it is of little use
humus and open shade found in this community are,
to make an analysis of the occurrence of the various
e.g. Brachycorythis kalbreyeri and Habenaria
species. Worth mentioning is the presence of
procera. The majority of orchid species that oc­
Peperomia sp. in 69.2 % of the communities ex­
casionally occur here normally belong to the central
amined.
or outer parts of the branches. Also the
pteridophytes in this community are a mixture of
Oleandra distenta - Tridactyle armeniaca e.c. species of humid habitats such as filmy ferns and
The fern Oleandra distenta, that forms the base of Nephrolepis undulata, as well as Drynaria laurentii
this community, is common and easily recognized. It and Microsorium punctatum of more sunny and
often starts to grow in or around the ramifications of drier sections of the crown. Begonia polygonoides,
the larger branches in the central part of the crown. Remusatia vivipara, Peperomia sp. as well as more
It expands in all directions but is mainly found at the drought resistent species such as Rhipsalis baccifera
basal parts of the branches that are often completely and Begonia rubro-marginata are also frequent in
overgrown by thick cushions of this fern and species this community.
associated with it. The slender rhizomes often hang
down giving this community a characteristic Microsorium punctatum - Vittaria guineensis e.c.
appearance (Fig. 80). In the earlier stages of the This community has a very limited distribution and

A cta phytogeogr. suec. 59


82 Dick Johansson

Table 32. The ten most common species in the different epiphy­
Close observation Distance observation
te communities. The occurrence is given as a percentage of the
total number of communities examined (Table 30). Plathycerium stemaria - Nephrolepis undulata e.c.
Nephrolepis undulata 58.3 Nephrolepis undu1ata 66.7
Close observation Distance observation Polystachya tessallata 38.9 Begonia rubro-marginata 50.0
Peperomia sp. 27.8 Polystachya tessallata 38.8
A ngraecum birrimense - Raphidophora africana e.c.
Begonia rubro-marginata 27.8 Davallia chaerophylloides 38.8
Raphidophora africana 79.3 Raphidophora africana 83.3
Vittaria guineensis 22.2 Diaphananthe bidens 33.3
Vittaria guineensis 48.2 Elaphoglossum barteri 36. 1
Lycopodium warneckei 1 9.4 Rhipsalis baccifera 33.3
Elaphoglossum barteri 34.5 Vittaria guineensis 33.3
Rhipsalis baccifera 1 6. 7 Vittaria guineensis 27.8
Calvoa trochainii 34.5 Diaphananthe bidens 27.8
Polystachya galeata 1 3 .9 Tridactyle anthomaniaca 22.2
Diaphananthe bidens 24. 1 Ancistrorhynchus capitatus 2 5 .0
Asplenium megalura 1 1.1 Peperomida sp. 1 6. 7
Microgramma owariensis 20.7 C alvoa trochainii 25.0
Polystachya polychaete 1 1. 1 Asplenium megalura 1 1. 1
Nephrolepis undulata 20. 7 Peperomia sp. 1 9. 4
Habenaria procera 1 3.8 Microgramma owariensis: 1 3. 9 Platycerium angolense - Nephrolepis undulata e.c.
Ancistrorhynchus capitatus 6.9 Nephrolepis undulata 1 1. 1 Nephrolepis undulata 47.5 Nephrolepis undulata 60.0
Rhipidoglossum paucifolium 6.9 Diaphananthe rutila 5.6 Polystachya tessallata 30.0 Polystachya tessallata 40.0
Peperomia sp. 25.0 Davallia chaerophylloides 30.0
A splenium africanum - Peperomia sp. e.c.
Davallia chaerophylloides 1 7 .5 Begonia rubro-marginata 30.0
Peperomia sp. 80.0 Peperomia sp. 62.5
Begonia rubro-marginata 1 5 .0 Diaphananthe bidens 30.0
Asplenium megalura 40.0 Preussiella sp. 37.5
Vittaria guineensis 1 5 .0 Rhipsalis baccifera 30.0
Calvoa trochainii 40.0 Begonia mannii 25.0
Polystachya galeata 1 2. 5 Tridactyle anthomaniaca 30.0
Elaphoglossum barteri 20.0 Arthropteris orientalis 25.0
Lycopodium warneckei 1 2. 5 Vittaria guineensis 20.0
N ephrolepis undulata 20.0 Nephrolepis undulata 1 2. 5
Nephrolepis biserrata 7.5 Chamaeangis vesicata 1 0.0
Begonia mannii 20.0 Asplenium megalura 1 2. 5
Diaphananthe bidens 7.5 Lycopodium warneckei 1 0.0
Polystachya tessallata 1 2. 5
A splenium dregeanum - Peperomia sp. e.c.
Oleandra distenta - Tridactyle armeniaca e.c.
Peperomia sp. 69.0
Tridactyle armeniaca 53.8 Tridactyle armeniaca 4 1 .7
Polystachya leonensis 28.6
Peperomia s p . 42.3 Drynaria laurentii 30. 6
Elaphoglossum chevalieri 2 1 . 4
Nephrolepis undulata 34.6 Begonia polygonoides 27.8
Asplenium barteri 1 9.0
Phymatodes scolopendria 30.8 Rhipsalis baccifera 27.8
Medinilla mannii 1 6. 7
Begonia polygonoides 26.9 Phymatodes scolopendria 2 5 .0
Arthropteris monocarpa 1 4. 3
Calvoa trochainii 23. 1 Nephrolepis undulata 1 9.4
Polystachya laxiflora 1 4. 3
Rhipsalis baccifera 23. 1 N ephrolepis biserrata 1 6. 7
Asplenium aethiopicum 1 1.9
Preussiella chevalieri 1 9. 2 Elaphoglossum isabelense 1 6. 7
Begonia rubro-marginata 9 . 5
Asplenium megalura 1 5 .4 Peperomia sp. 1 3 .9
Bulbophyllum inflatum 7. 1
Microsorium punctatum 1 1 . 5 Asplenium megalura 1 2. 5
Tridactyle tridactylites - Medinilla mannii e.c.
Microsorium punctatum - Vittaria guineensis e.c.
Medinilla mannii 58.8 Medinilla mannii 53.5
Vittaria guineensis 5 1 .8 Vittaria guineensis 36. 1
Asplenium dregeanum 4 1 .2 Polystachya laxiflora 27.9
Peperomia sp. 44.4 Davallia chaerophylloides 3 3 . 3
Polystachya leonensis 35.3 Lycopodium mildbraedii 2 5 . 6
Davallia chaerophylloides 3 7.0 Peperomia sp. 30.6
Vittaria guineensis 35.3 Peperomia s p . 23.3
Polystachya tessallata 29.6 Polystachya tessallata 27.8
Asplenium barteri 29.4 Polystachya leonensis 23.3
Diaphananthe bidens 25.9 Asplenium megalura 25.0
Polystachya laxiflora 23.5 Asplenium dregeanum 20.9
Begonia rubro-marginata 2:2. 2 Medinilla mannii 25.0
Polystachya obanensis 23.5 Vittaria guineensis 18.6
Nephrolepis undulata 1 8. 5 Diaphananthe bidens 1 9.4
Rangaeris muscicola 23.5 Cyrtorchis arcuata 1 3.9
Oleandra distenta 1 8. 5 Graphorchis lurida 1 3 .9
Peperomia sp. 23.5 Bulbophyllum scariosum 1 1 .6
Medinilla mannii 14.8 Angraecum birrimense 1 3 .9
Elaphoglossum barteri 1 7. 6 Elaphoglossum salicifolium 1 1 . 6
Bulbophyllum cocoinum 1 4 . 8 Nephrolepis biserrata 1 1. 1
Bulbophyllum scariosum - Polystachya dalzielii e.c.
Drynaria laurentii- A splenium megalura e.c.
Xiphopteris oosora 42.9 Polystachya dalzielii 44.4
Asplenium megalura 53.2 Asplenium megalura 43.7
Polystachya dalzielii 42. 9 Lycopodium mildbraedii 27.8
Tridactyle armeniaca 46.8 Rhipsalis baccifera 29.7
Elaphoglossum salicifolium 28.6 Asplenium megalura 22.2
Rhipsalis baccifera 45. 1 Tridactyle armeniaca 2 5 .0
Bulbophyllum bifarium 28.6 Vittaria guineensis 22.2
Polystachya tessallata 32.3 Graphorchis lurida 25.0
Xiphopteris vilosissima 28.6 Tridactyle tridactylites 22.2
Bulbophyllum linderi 25.8 Bulbophyllum oreonastes 1 8 . 7
Arthropteris monocarpa 1 4. 3 Elaphoglossum salicif. 1 6.7
Ficus kamerunensis 22. 6 Angraecum distichum 1 7.2
Asplenium megalura 14.3 Asplenium geppii 1 1.1
Graphorchis lurida 22.6 Cyrtorchis arcuata 1 4. 1
Lycopodium mildbraedii 1 4. 3 Arthropteris sp. 1 1.1
Polystachya galeata 16.1 Polystachya tessallata 1 4. 1
Polystachya pobeguinii 1 4. 3 Podangis dactyloceras 1 1.1
Medinilla mannii 1 1 .3 Phymatodes scolopendria 1 2. 5
Vittaria guineensis 1 4. 3 Asplenium aethiopicum 1 1. 1
Nephrolepis undulata 9.6 Bulbophyllum linderi 1 1 .7

A cta phytogeogr� suec. 59


Ecology of vascular epiphytes in West African rain forest 83

Fig. 8 1 . Simplified drawing of a


Microsorium punctatum - Vittaria
guineensis epiphyte community.
(A) Microsorium punctatum,
(B) Davallia chaerophylloides, (C)
Asplenium megalura, (D) Poly­
stachya tessallata, (E) Tridactyle
anthomaniaca, (F) Peperomia
rotundifolia, (G) Vittaria guineen­
sis.

serves as a small refuge for humus-demanding


species, particularly if it grows on a smooth barked
phorophytes (Fig. 8 1 ). Two plants that are often
associated with this particular fern are Vittaria
guineensis and Davallia chaerophylloides.

Drynaria laurentii - Asplenium megalura e.c.


The Drynaria fern is probably the most common
species of epiphyte in the investigated area. It is
found in sunny habitats mostly in the crowns. It easi­
ly gets a foothold on smooth surfaces ; it can even be
seen growing on vertical concrete walls. This highly
specialized fern has two kinds of fronds (Fig. 26).
The fertile fronds have long narrow finger-like pin­
nately arranged lobes. These lobes or segments are
articulated from the midrib after maturity.
Afterwards the midribs remain in an erect position
for many years.
The sterile fronds are thin and papery, somewhat
resembling oak leaves, and are held upright and
more or less parallel to the substratum. Between
these sterile, mostly dead leaves and the substratum
a niche is formed in which materials of various kinds
are collected. The sterile leaves are persistent and re­
main in function as 'humus collectors'. When they
get older and start to decay, the veins persist forming Fig. 82. Simplified drawing of a Drynaria laurentii -
a delicate net. This species is a very hardy plant that Asplenium megalura epiphyte community, 29 m above
is able to survive severe conditions. A colony that the ground, on a Piptadeniastrum africanum. Yekepa 530
m. (A) Drynaria laurentii, (B) Oleandra distenta, (C)
was placed on top of a flat aluminium roof, with no
Bulbophyllum lupulinum, (D) Bulbophyllum bufo, (E)
protection whatsoever, showed no sign of discomfort Graphorchis lurida, (F) Bulbophyllum buntingii, (G)
but continued to grow year after year. Asplenium megalura.

Acta phytogeogr. suec. 59


84 Dick Johansson

Fig. 84. Platycerium stemaria - Nephrolepis undulata


epiphyte community. Nimba Range 700 m. Note the large
Begonia rubro-marginata in the centre.

Asplenium megalura will be common. In roughly


Fig. 83. Drynaria volkensii and Rhipsa/is baccifera. every second community examined this species was
Usambara Mts 900 m, Tanzania.
present.
The rhizome is thick and hard, covered with a In the final stage a wide array of plants that are
dense insulating layer of scales. Sometimes it favoured by a rich supply of humus invade the thick
branches, particularly when it grows in a vertical densely overgrown accumulation. Now even
direction. terrestrial plants appear as facultative epiphytes.
The Drynaria community has a wide distribution During this s�age the Drynaria fern itself plays an in­
and passes through several stages in its development. ferior role, and is in many cases dead or dying. As a
During the colonization stages (Fig. 82) epiphytic matter of fact the presence of rhizomes and old ster­
orchids, e.g. Bulbophyllum oreonastes grow on the ile leaves from Drynaria laurentii in the bottom
open bark surfaces between the rhizomes of the fern. parts of many rich accumulations indicate this
The high total number of orchids, with not less than process.
38 species registered from this community, are most­ From the Usambara Mts in Tanzania Moreau
ly connected to this stage. As the fern grows and ( 1 943) reports : "Somewhat contrary to the expecta­
takes larger parts of the substrate in possession the tion orchids do not occur in association with the
orchids become less numerous. coarse epiphytic fern Asplenium nidus and
The humus is utilized by several plants. Rhipsalis Dryopteris laurentii (Drynaria laurentii) which
baccifera and Peperomia sp. are among the first to would seem to offer favourable foothold, but orchids
settle in thi s environment (Fig. 83). Among the often have their roots closely interwoven with those
orchids A ngraecum distichum and even Graphorchis of smaller ferns and naturally are often more or less
lurida are still present during this stage in develop­ embedded in moss and lichens." This is partly true,
ment. When more humus has been formed even if I myself on a brief visit in the same area dis-

Acta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 85

Fig. 8 5 . Chamaeangis vesicata in


a Platycerium angolense
Nephrolepis undulata epiphyte
community. Y ekepa 500 m.

covered several ferns and one orchid, Cirrhopetalum clearly can be seen through the lower parts of the
umbellatum (Forst. f.) Hook. & Arn., associated newly developed sterile leaves. The fertile leaves are
with Drynaria volkensii. Moreau probably refers to long and pendulous.
the first stages when the Drynaria fern seldom has The rhizome is short. It is not creeping as in most
any other epiphytes associated with it. other epiphytic ferns but is always growing at a 90
degree angle towards its point of attachment.
Platycerium stemaria - Nephrolepis undulata e.c. The occurrence of the Platycerium communities in
and Platycerium angolense - Nephrolepis undulata relation to the total presence of these ferns, is low.
e.c. The low number of epiphytes associated with the
Two species of the genus Platycerium are common Platycerium ferns is also surprising. Platycerium
in the Nimba area, Platycerium angolense and P. angolense had 1 7 species and P. stemaria 22 species
stemaria can often be found together in non-limiting totally. A look at the species compositions shows
environments, but in open or more exposed habitats that the plants associated with the Platycerium ferns
Platycerium angolense dominates. In other parts of commonly appear in other epiphytic communities as
Africa they also grow on rocks, e.g. in the Matombo well. Nephrolepis undulata is the species that most
region east of the Uluguru Mts in Tanzania. Both frequent takes advantage of the substrate or perhaps
species are widespread in tropical Africa. They in­ a favourable microclimate caused by the large
habit the outermost branches of trees in the primary Platycerium leaves. Boy er ( 1 964) reports that
forest, but in secondary forest or on scattered trees Microgramma owariensis and Phymatodes scolopen­
around villages and in farmland they can be found dria sometimes are associated with Platycerium
on any part of the tree. ferns. She also mentions that small plants of cocoa
The fertile and sterile leaves are of a different trees and oil palms were found in accumulations of
shape (Figs 84, 85). This dimorphism is of a par­ these ferns. Holtum ( 1 960) states that: " Cymbidiella
ticular function in the life of the plant. The sterile rhodochila in Madagascar is invariably found grow­
leaves can be divided into two sections. The upper ing in association with a Platycerium. " The same
part is open towards the air on both sides, it is thin observation appears in Encyclopedia of C ultivated
and pergameneous. Their function is to collect the Orchids (Hawkes 1 965 : 1 3 9) with no reference to the
falling debris to be used in the building of humus. source.
The lower part of the sterile leaves is considerably
thicker. This part of the leaf is a deposition area for Epiphyte communities at altitudes above l 000 m
the debris that has been trapped between the upper Asplenium dregeanum-Peperomia sp. e.c.
part of the leaf and the branch or trunk. The space At the base of trunks (from the ground to roughly
between the sterile leaves is utilized by roots that 4-5 m height) Asplenium dregeanum and Peperomia

Acta phytogeogr. suec. 59


86 Dick Johansson

General observations
Ferns form the base in seven of the ten recognized
communities. In all these communities it is the
substrate-building capacity of the ferns that seem at­
tractiv·e to the other species. Interaction and an­
tagonism between ferns and orchids, as well as
between mosses and liverworths, have been reported
by Tixier. He found that the 'stations' rich in orchids
has generally no ferns and reciprocally. "Les
Fougeres presentent une axemie assez forte vis-a-vis
des Orchidees" (Tixier 1 966 : 1 24). However,
Fig. 86. Tridactyle tridactylites - Medinilla mannii
epiphyte community. Mt Gbahm 1 300 m. Drynaria rigidula was found to hold numerous
epiphytes. This statement may seem to contradict
the observations in the Nimba area, but the state­
ment is more correct for the lower parts of the trunks
sp. dominate. They are so common that nearly every (a section which was particularly carefully examined
tree has at least one of them on its trunk. At these by Tixier). It is, however, hard to asess the amount
altitudes the filmy ferns are also common and par­ of competition between ferns and orchids. In many
ticularly on the trunks, and thus are often associated cases it is probably the low light intensities close to
with the species of this community . . the ground that limit the presence of orchids, and
The dominating fern, Asplenium dregeanum, mul­ higher up in the trees the evaporating power of the
tiplies by sending out modified stolon-like fronds air often limits the presence of ferns.
which take root and produce a colony of daughter
plants, expanding the community in many directions.
Since this community occurs on the trunks Epiphyte succession
relatively close to the ground no distance obser­ As the phorophyte grows it will pass· through several
vations had to be performed. 'environmental stages', e.g. it's place in the forest
alters, the size and the shape of the crown and the
Tridactyle tridactylites - Medinilla mannii e.c. morphology of the bark changes. These are primary
In the central part of the crown and especially changes. Secondary changes, e.g. establishment of
around the larger ramifications the orchid Tridactyle new species of epiphytes will effect the competition
tridactylites forms huge interwoven masses (Fig. 86). between the epiphytes. The epiphytic flora itself also
Here several plants occur, such as Polystachya lax­ participates in the successional processes through
iflora, P. leonensis and Asplenium dregeanum. active or passive contributions to the formation of
Scattered groups of filmy ferns may occur but very humus.
dominant are the mosses that occupy the major part Direct or indirect methods may be employed to
of the branches in aHd around this community. estimate the changes. The best method would
naturally be to map the epiphytic flora on a number
Bulbophyllum scariosum - Polystachya dalzielii e.c. of branches in the crowns of living trees. These
This somewhat less clearly defined community is branches could then be re-examined after certain
found in the central or outermost parts of the time intervals, thus giving a correct picture of the
branches. It includes a number of small sized development and eventually changes of the flora.
epiphytes, e.g. Xiphopteris villosissima, that are easi­ This method would be similar to the 'permanent
ly overlooked in a distance examination. quadrate' technique used in terrestrial ecology, ex­
Two species of orchids dominate this community : cept for the changes in size of the branch caused by
Bulbophyllum scariosum which is easily seen, while growth or broken off pieces. Unfortunately this
the more tiny Polystachya dalzielii is harder to method is for practical reasons very difficult to use.
observe when not in flower. In this section of the The inaccessibility of the tree crowns has previously
crown the lichens cover the branches or hang down been discussed. The slow development of many of
( Usnea sp.) from the finest twigs. the epiphytic plants, e.g. the orchids, would also

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 8 7

make i t necessary to continue the observations for a Colonization stages

very long period of time. Thus the difficulties con­ orchids


pteridophytes other vase.
e p iphytes
nected with a detailed study of the epiphytic flora on I nitia l

a phorophyte in situ make this kind of investigation [] 1 8 '/, 1 0 '1.

less useful. There are, however, several indirect orchids +


pteridophytes + orchids pteridophytes
methods to use. Comparisons of the epiphytic flora other vase.
epiphytes + other vase.
epiphytes Intermediate
between phorophytes in the primary forest and on
phorophytes that have become more exposed D 1 8 '1. 1 0 '1.

through changes in their environment may give


pteridophytes + orchids + other vase. epiphytes
valuable information on the amount and direction of Final

eventual changes. Analyses of humus deposits may 4 0 '1.

also reveal traces of an earlier epiphytic flora.


Fig. 87. 'Colonization stages'. The number of phorophytes
Epiphyte succession in Himalayan Quercus forest in sample plots 1-111 have been split into seven classes
The sucessions of epiphytes h as been studied by based on the presence of different epiphytes. The oc­
Dudgeon ( 1 923) in a Quercus incana forest at Lan­ currence of the different classes is given as a percentage of
the total number of phorophytes and is also illustrated in
dour, Western Himalayas ( 1 900-2 1 00 m). Since the blocks proportional to each other.
earliest stages in the colonization of the branches are
not dealt with in this study it might be of interest to The moss stages (3 and 4) are non-existent in the
give an account of his observations. rain forest below 1 000 m altitude, but become in­
Six separate stages are distinguished : creasingly important at higher altitudes.
1 . C rustose lichen stage; starting when the branch is
three to four years old. Succession indicated by floristic composition
2. Foliose and fructicose lichen stage; starting almost This analyse is built on the assumption that sooner
as early as the crustose lichens, but begins to make a of later species from all of the three major groups of
characteristic presence three to four years later. epiphytes, (i.e. the pteridophytes, the orchids and the
3. Pioneer moss stage; pioneer mosses gain foothold other vase. epiphytes), will be represented on the
in favourable spots. Other trailing and erect mosses phorophyte. No phorophyte, except the oil palm,
gradually crowding out the lichens, probably by Elaeis guineensis, has given proof of being rejected
shading them. by any particular group of epiphytes.
4. Climax moss stage; starting when the thick con­ Three stages in the colonization of the
tinuous pads are developed. Under favourable con­ phorophytes are recognized, the initial, intermediate
ditions this stage is reached in about 20 years. and final. stage. During the initial stage species from
5. Fern stage ; starts when epiphytic ferns gain only one of the major groups of epiphytes are
footholds in the moss pads. The fern stems and roots represented on the phorophyte. In the intermediate
collect and hold more dust, forming a soil bed that stage species from two groups are present, and in the
may attain a thickness of 10 cm. final stage species of all groups are occurring on the
6. Flowering plant stage; starts with the appearance phorophyte. From the three sample plots (Tables 1 4-
of other vascular plants apart from the ferns. 1 6) the occurrence of phorophytes representing the
Dudgeon summarizes his observations : three colonization stages are given in Fig. 8 7 . The
"Successions shown by epiphytes are considered to colonization seems to start with the pteridophytes or
be unique, because of ( 1 ) the short time required for the other vascular epiphytes. This initial stage is
completion, (2) the unusual clearness of successive succeded by the entry of the orchids. As time passes
stages, (3) the small size of the plants involved, and these phorophytes in the intermediate colonization
(4) the adaption of the plants to repeated and stage will most likely be colonized by some species
prolonged desiccation." from the third group and enter the final stage.
Although Dudgeon's investigation refers to a
cooler environment than the rain forest in Nimba, Succession traced by analysis of humus deposits
the pattern described could with some exceptions be This technique was used in the attempt to
said to be valid for the rain forest in Nimba as well. reconstruct the development of the humus deposits

A cta phytogeogr. suec. 59


88 Dick Johansson

and the plants involved in this process. The nature of In two accumulations each of a Microsorium punc­
these deposits is very heterogeneous. They are built tatum and a Platycerium stemaria community no
up of material in all stages of decomposition, mostly traces of an older epiphytic flora could be found.
resembling peat. Occasionally the decomposition has
gone so far that a mineralization has begun. Such Oleandra distenta - Tridactyle armeniaca e.c.
soil-like deposits can be found at the basal parts on Accumulations from this community proved to be
the large branches of old trees. More commonly the most rewarding in the search for remains of an
deposits are built up by material that is partly earlier flora. Four such accumulations were in­
decomposed. It is rather surprising that parts of dead vestigated, two of which held remains that could be
plants incorporated in the accumulation several identified. In one of them sterile leaves and rhizomes
years earlier frequently can be identified. This shows of Drynaria laurentii appeared in the bottom layer.
either that the plants are very resistent towards The other accumulation was so rich in old plant
decomposition, or that the micro-organisms that are remains that it will be presented in detail. This com­
responsible for the break -down process are munity was located at the basal part of a large
restrained in their activities by the environment, branch I 7 m above the ground on a Lophira alata at
which is regularly subjected to drought. 600 m altitude. The thickness of the humus layer was
Several cross-sections were cut through the ac­ not less than 1 6 cm. In the bottom part of the ac­
cumulations at a right angle to the axis of the cumulation the material was partly mineralized. The
phorophyte, with a sharp broadblade knife. This size of the investigated part was 0.2 x 4.0 m.
gave a glimpse of the 'stratification' in situ. The con­ The surface vegetation was a dense aggregation of
tent from the bottom layer was sifted through a sieve Oleandra distenta with the rhizomes hanging down
with a mesh of 10 x 1 0 mm. Remains of plants, such from the branch on both sides. Ten more species
as old pseudobulbs and rhizomes, could then much were present: Asplenium megalura, A rthropteris
easier be discovered. A total of 1 3 deposits of five orienta/is, Begonia polygonoides, Bulbophyllum
different epiphyte communities were examined. cocoinum, Brachycorythis kalbreyeri, Habenaria
leonensis, Nephrolepis undulata, Medinilla mannii,
A ngraecum birrimense - Raphidophora africana e.c.
Polystachya polychaete and Rhipsalis baccifera. The
Two communities, both occurring on the trunk 3-4
topmost 10 cm of this accumulation had a felt-like
m above the ground on two tall trees, were ex­
structure, and was composed of living and dead
amined. The humus layer was thin and held in a ver­
roots and rhizomes. Below this layer large quantities
tical position by a network of roots. Six species of
of partly decomposed plant material were found.
epiphytes were present : A ngraecum birrimense,
Sterile leaves and rhizomes of Drynaria laurentii
Calvoa trochainii, Diaphananthe rutila, Nephrolepis
were accompanied by leaves and rhizomes from
undulata, Polystachya rhodoptera and Raphido­
Phymatodes scolopendria and pseudobulbs of
phora africana, with the first and last species
Graphorchis lurida and several unidentified
dominating.
Bulbophyllum spp.
In the bottom layer in one of the communities 3
Traceable remains of an earlier epiphytic flora
groups of pseudo bulbs from Polystachya affinis were
were found in 30.7 % of the accumulations ex­
discovered.
amined. It is interesti.ng to note that the species that
Drynaria laurentii - Asplenium megalura e.c. were overgrown were all able to colonize clean bark
Three humus deposits from this community were ex­ surfaces, so called pioneer epiphytes, e.g.
amined. Two came from the central part of the Polystachya affinis, Drynaria laurentii, and several
crown and one from the outermost part. In one of Bulbophyllum spp.
the deposits from the central part of the crown some The expansive growth of the Drynaria fern will
old pseudobulbs probably produced by Bulbophyl­ regularly overgrow or surpress other plants of slower
lum lupulinum and pseudobulbs and aerial roots of growth, e.g. the many Bulbophyllum spp. that fre­
Graphorchis lurida were discovered. quently occur in the same habitat.
In the 0/eandra distenta- Tridactyle armeniaca com­
Microsorium punctatum - Peperomia sp. and munity it looks like several other epiphytes also are
Platycerium stemaria - Nephrolephis undulata e.c. active in the accumulation of humus, and that Olean-

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 89

Table 33. Epiphytes recorded on three Lophira alata, A, B and


C. (See text.)

Epiphyte Phorophyte
A B c
Asplenium megalura + +

Drynaria laurentii + + +

Elaphoglossum isabelense + +

Elaphoglossum salicifolium +

Oleandra distenta +

Phymatodes scolopendria +

Vittaria guineensis +

Aerangis laurentii +

Angraecum birrimense +

Angraecum distichum +

Angraecum subulatum +

Bulbophyllum distans + +

Bulbophyllum linderi + + +

Bulbophyllum maximum +

Bulbophyllum oreonastes + + +

Bulbophyllum phaeopogon +

Bulbophyllum saltatorium +

Bulbophyllum schimperanum +

Bulbophyllum schinzianum +

Chamaeangis vesicata +

Cyrtorchis arcuata +

Graphorchis lurida +

Listrostachys pertusa + + +
Fig. 88. Drynaria laurentii on a phorophyte in secondary Plectrelminthus caudatus +
forest. Mt Tokadeh 700 m. Polystachya polychaete + +

Polystachya saccata + +

Tridactyle anthomaniaca + + +

Tridactyle armeniaca + +

Tridactyle crassifolia +

Medinilla mannii + +

Peperomia rotundifolia +

Rhipsalis baccifera +

Total no. of species 20 18 12

A B
dra distenta under certain conditions can utilize this
substrate. The fact that this community is restricted
to the basal area of the crown indicates that a large
accumulation of humus is not enough to promote a
successful colonization by Oleandra distenta.

Effects of environmental changes


The direction of the colonization process may be
reversed or changed if the phorophyte is subjected to
an environmental change. Such changes -occur fre­
quently when the forest is affected by human ac­
c tivities. Here and there isolated trees will be left for
one reason or another.
A telescopic survey of twelve large isolated trees
Fig. 89. Number of orchid species on three Lophira alata:
A, B and C. (See text and Table 3 3 .) The figures in the in Y ekepa revealed that the ratio between species of
overlapping areas give the number of species in common. pteridophytes and orchids was roughly 1 :3 ( 1 2 :39),

A cta phytogeogr. suec. 59


90 Dick Johansson

which may be compared to the ratio in the dense had only four orchid species (36.3 %' in common
forest that was exactly 1 :2 ( 1 9 :38, Table 1 7). This is with the phorophyte from the most protected en­
hardly surprising, since the higher light intensities vironment (A) (Fig. 89). This may indicate that most
will favour the orchids which also are able to with ­ of the orchids are adapted to rather narrow en­
stand the much drier environment of more exposed vironmental conditions, which also could explain
trees. Drynaria laurentii is the only fern that can their richness in species.
take advantages of these changes in the environment, The study of the changes during a two years
and occurs here in sizes and numbers never found in period in the epiphytic flora on a branch that had
the dense forest (Fig. 8 8). Rhipsalis cassytha is the fallen down from a large tree may serve as an exam­
only species of the other vascular epiphyte group ple of changes that occur when the environment
that seems to thrive on these exposed phorophytes. becomes darker and more humid. The branch had
Comparison of the epiphytic flora on three fallen down over a small creek and was lying about
phorophytes of the same species (Lophira alata) of 2 m above the small stream. A stretch of six meters
approximately the same sizes, from the same altitude was examined, in November 1 96 7 and re-examined
but growing in three different habitats, may ex­ at the same time during the years 1 968 and 1 969
emplify the effect of the environment in more detail. (Table 34).
One phorophyte (A) was 42 m in size and grew in One must remember when analyzing the changes
a high forest surrounded by several other trees of the in the flora on this branch that it does not represent
same size. The second phorophyte (B) was 40 m and the same substrate as a living branch. The effect of
was left in a forest subjected to logging operations the decomposition of the bark remains unknown.
that had created a rather broken canopy. The third Anyway, the changes may give an idea of the
phorophyte (C) was 39 m and grew in a very ex­ tolerance or adaptability of the various species in­
posed environment with several dead trees around it. volved.
A total of 3 2 species of epiphytes were recorded The original species composition shows a number
from these three phorophytes (Table 3 3). A more of sun-demanding orchids, pteridophytes and other
open environment seems mainly to limit the presence vascular epiphytes. Only after some months the
of non-orchid epiphytes. The number of orchid leaves of the orchids started to turn yellow. After one
species was about the same on all of the phorophytes year the Bulbophyllum species were dead and so
( 1 2, 1 2 and 1 1 ), but it should be noticed that the were the A ngraecum distichum and Polystachya sac­
phorophyte (C) in the most exposed environment cata specimens. After one year three species of

Table 34. Changes in the epiphytic flora on a dead branch during a two years period. (See text.)

1 96 7 1 968 1 9 69

Orchids Orchids Orchids


Angraecum distichum Graphorchis lurida Brachycorythis kalbreyeri
Bulbophyllum intertextum Listrostachys pertusa Tridactyle armeniaca
Bulbophyllum oreonastes Tridactyle armeniaca
Graphorchis lurida Pteridophytes
Listrostachys pertusa Pteridophytes Arthropteris monocarpa
Polystachya saccata Arthropteris monocarpa Asplenium barteri
Tridactyle armeniaca Asplenium megalura Asplenium megalura
Nephrolepis undulata Asplenium paucij ugum
Pteridophytes Oleandra distenta Elaphoglossum kuhnii
Asplenium megalura Oleandra distenta
Drynaria laurentii Other vase. epiphytes Vittaria guineensis
Lycopodium warneckei Begonia polygonoides
Nephrolepis undulata Peperomia sp. Other vase. epiphytes
Vittaria guineensis Remusatia vivipara Begonia mannii
Raphidophora africana Begonia polygonoides
Other vase. epiphytes C alvoa monticola
Rhipsalis baccifera Remusatia vivipara
Raphidophora africana

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 91

Fig. 90. Epiphytes on a dead branch (see text) . (A)


Remusatia vivipara, (B) Nephrolepis biserrata, (C) Olean­
dra distenta.

orchids were still alive, but the Graphorchis lurida


plants had developed very elongated leaves and the
plant did not flower later in the season. The few
Tridactyle armeniaca and Listrostachys pertusa
specimens were also in poor shape. The Drynaria
plants had been overgrown by Nephrolepis undulata,
and Raphidophora africana seedlings could be seen
in large numbers.
After two years the branch was completely
overgrown by epiphytes. Nephrolepis undulata and
Raphidophora africana dominated the picture (Fig.
90). Only one species of the original group of orchids
was still alive, but some small seedlings of
Fig. 9 1 . Very extensive epiphytic growth on a tall tree.
Brachycorythis kalbreyeri were discovered.
Nimba Range 800 m.

rich moss or lichen flora may indirectly contribute to


Effects of the epiphytes on the a higher humidity in the bark which would favour
phorophytes development of numerous fungi that saprophytically
or parasitically could endanger the existence of the
According to the definition of an epiphyte, it should
phorophyte. B acteria and insects may also take ad­
not affect the phorophyte in a direct way, e.g. draw
vantage of the humid environment.
water of nutrients from its living tissues.
The vascular epiphytes have generally been con­
sidered harmless and at most troublesome due to
EARLIER OBSERVATIONS
overgrowing (Fig. 9 1 ) and thus strangling the hosts
Indirect effects on which they live, e.g. Richards ( 1 964 : 1 1 0) :
Epiphytes may affect the phorophyte in a more in­ "Through epiphytes are wholly dependent o n other
direct way. There are many opinions concerning the plants, they in their turn have little effect on their
effect of cryptogamic epiphytes on the phorophytes, supports."
as related by Barkman ( 1 95 8 : 1 8- 1 9) and Ruinen Coffee growers in Java suspect that the common
( 1 9 5 3 : 10 1 ). It has generally been recognized that epiphyte Cyclophorus nummularifolius kills the
epiphytic growth is richer on dead than on living coffee trees as reported by Went ( 1 940 :92). The
trees, but without experimental proof it is difficult to latter thinks this is only an illusion since these
say whether the epiphytes are the direct cause of epiphytes are most common on the upper dead
death of the phorophyte, or only infest sick trees, branches and twigs. When the dead branches fall the
eventually accelerating their death. The presence of a roots and rhizomes keep them hanging around the

A cta phytogeogr. suec. 59


92 Dick Johansson

tree instead of falling down to the ground. This The prompt recovery of the host plant when the
creates the impression that the trees carrying this epiphytes are removed was experimentally proved.
epiphyte have more dead branches than others. Plants artificially infected with the fern Drymo­
glossum showed a rapid decline within a year. An
Direct effects anatomical investigation revealed that in the infested
Cook-Melville ( 1 926) considers the epiphytic orchids branches the vessels were partly blocked by gum and
to be parasites on citrus trees in Puerto Rico. fungus and the cortex contained many hyphae. The
The generally accepted belief that epiphytes as a resultant deficient nutrient and water supply would
whole are limited to a non-parasitic life has been amply account for the inhibitory effect.
carefully examined by Ruinen ( 1 95 3), whose opinion The invading hyphae were traced microscopically
and interesting results will be summarized here. She to the place of exit from the epiphyte in which it
states that the typical epiphytes preferably develop formed a mycorrhiza. The inference drawn from this
on young, living tissue and that plants declining in is that the mycorrhizal fungus of the epiphyte is
health usually recover, if the epiphytes are removed. potentially parasitic on the supporting trees. It
Contrary to Went, Ruinen confirms the observation depends on the constitutional conditions presented
by planters on Java that cleaning their coffee, tea by the supporting tree, whether or not it will appear
and citrus trees from epiphytes causes the trees to as such.
become healthier subsequent to this treatment. It would seem hardly possible to speak of a sym­
The early symptoms of decline under epiphytic biosis between epiphyte and supporting plant, were it
growth are difficult to define. They suggest a general not for the failures in cultivating some orchids, e.g.
deficiency, an aspect of ill-health, of failing. The Taeniophyllum sp. on inorganic substrate and,
leaves are no longer lush, still green but less so than moreover, for the marked preference shown by
usual, not as turgescent as the non-infested branches, epiphytes for some definite phorophytes under
but nevertheless not wilted. In this stage the develop­ otherwise identical environmental conditions. These
ing buds are slightly smaller. The leaf fall is conditions are undoubtedly inherent in the suppor­
premature, therefore the vegetative period is ting plant itself, which obviously of(ers favourable
shortened. The noxious effect is localized at first to symbiotic relations with the epiphytes, and
the parts directly under the epiphyte, the other parts therefore, from the point of the support the sym­
are only secondarily affected, viz. show the influence bioses should be qualified as antagonistic.
at a later date. Ruinen concludes that the generally accepted
The phenomenon, which is termed epiphytosis, belief that epiphytes as a whole are limited to a
may be described as a slow exhaustion of all partners saprophytic habit, seems not only open to discussion
of the biocoenosis one after the other (Ruinen but as such has been proved to be actually
1 95 3 : 1 08). As the epiphytes are known to be erroneous. (Cf. Harvais & H adley 1 967.)
mycotrophic, the most simple form in which the In a study of structural connections between
biocoenosis in its relations and deterioration might epiphytes and host plants, Furman ( 1 95 9 : 1 27) has
be represented is illustrated in Fig. 92. made observations similar to those of Ruinen. In a
Honduran cloud forest the interpenetration of root
hairs or root mycelium with host tissues was ex­
amined. Vascular epiphytes, chiefly orchids,
bromeliads and ferns, were found to penetrate living
e p iph yte + fu n gu s + sup p o rt hosts' stem tissues with root hairs, or share a com­
mon fungus mycelium with them. Frequently the
L
J
sy m biosis _j host tree stems had very thin, lightly suberized bark.
Schimper ( 1 935 : 3 1 1 ) have earlier stressed the fact
( antibiosis
that many epiphytes belong to those families known
for their mycorrhiza, e.g. Orchidaceae, Ericaceae,
Melastomaceae and 'perhaps' even other ones and
Fig. 92. Interactions between epiphyte, fungus and sup­ that the fungus might play an important role in the
port. (From Ruinen 1 95 3.) metabolism of the epiphyte.
A cta phytogeogr. suec. 59
Ecology of vascular epiphytes in West African rain forest 93

AUTHOR'S OBSERVATIONS
Any direct effect by the epiphytes on their
phorophytes has not been noticed. However, certain
species of phorophytes, e.g. Mitragyna ciliata and
Parinari excelsa (Fig. 1 8) with a rich epiphytic flora
often give the impression of suffering from
'epiphytosis' as described by Ruinen.
The 'preference' of certain epiphytes for a sub­
strate of young bark or living tissues, as reported by
R uinen and Furman, should be reflected in the dis­
tribution pattern. Species with such preferences
would have their main location in the outermost
parts of the crown, on the youngest branches and
twigs. Only one species, Bolusiella talbotii, shows
such a distribution pattern (Fig. 7 8).
There are, however, certain species of orchids that
seldom appear on the tall trees, but seem to prefer
younger ones and even shrubs. Such habitats are
found around creeks and small rivers, or in
'farmlands'. Five species are ± restricted to these
habitats : A ngraecum classensii, Diaphananthe den­
siflora, D. rutila, Eurychone rothschildiana and
Rhipidoglossum paucifolium.

The leafless orchids


Brief observations of the leafless orchid Microcoelia
exilis Lindl., (Fig. 9 3), in Tanzania may contribute
to the picture of the relationship between epiphyte
and phorophyte.
Along the Great Ruaha river in the vicinity of
Fig. 94. Microcoelia exilis (see arrow) in a typical habitat.
Kidatu at 400 m altitude the Microcoelia plants were Kidatu 400 m , Tanzania.
very common on certain trees in the riverine forest.
small trees (Fig. 94). Very frequently the section of
They generally occurred on the topmost branches in
the branch on which the orchid occurred was dead
very exposed positions, and on twigs or branches of
or dying.
Morris ( 1 970:88) reports that Microcoelia exilis
occurs "mainly on smooth barked trees
and . . . mainly on the outer branches". Several
species of Microcoelia are reported to grow on the
outermost twigs or thin branches of large trees, or
amongst bushes and young trees (Stew art 1 973 :802-
803). Leafless orchids of the genus Taeniophyllum
were observed to cause deficiency symptoms in the
phorophyte (Ruinen 1 95 3 : 1 05) and were .preferably
attached to the living epidermis of the younger
branches, twigs or stems. Taeniophyllum shows op­
timum growth at surprisingly low light intensities ;
Fig. 93. Microcoelia exilis. The young roots grow
1 /8- 1 /5 of the outside light (Wiesner 1 89 7). This
horizontally and freely probably to increase
photosynthesis and water uptake. Kidatu 400 m , Tan­ may indicate that it is not primarily the need of light
zania. that is responsible for its occurrence in the outermost

A cta phytogeogr. suec. 59


94 Dick Johansson

Fig. 95. Records of


Microcoelia species in
West Africa. (From
F.W.T.A. 3 ( 1 ) 1 968.)

parts of the branches. It has generally been assumed genera of orchids throughout the world has generally
that the photosynthesis taking place in the roots of been considered to be an adaptation for preventing
the leafless orchids is sufficient for the needs of the water losses through the leaves (Kerr 1 972 : 307,
plant (Dycus & Knudson 1 95 7). However, there Teuscher 1 97 2 :497).
seems to be no experimental proof of this assump­ If one looks upon the distribution of the species
tion. The rate of photosynthesis in roots of orchids among e.g. the West African leafless orchid genus
with green leaves (Erickson 1 95 7, Dycus & Knud­ Microcoelia Lindl., they all appear in areas with con­
son 1 9 5 7) clearly shows that these roots do not even siderable amounts of rain and with such minor
produce material that can support their own respira­ climatic disadvantages for epiphytes that a very
tion. Difficulties for leafless orchids growing on dead large number of leaf carrying epiphytic orchids with
substrate (Kerr 1 972 :307) (Teuscher 1 972 :497) or green leaves occur in the very same area (Fig. 9 5).
when transplanted (Stewart, 1 970 :90, 1 97 3 :803) can The possibility that the leaves have become obsolete
be used as argument that these orchids to a certain due to a partly parasitic life, should be carefully ex­
extent are dependent on living tissues. amined.
The reduction of leaves that occur in several

A eta phytogeogr. suec. 59


V. Environmental influence

Environmental influences can be referred to three altitudes (500-700). Comparisons will also be made
main groups : climatic factors, substrate and biotic with the epiphytic flora at altitudes higher than those
factors. Although environmental influences result in the investigated area, as observed on the high
from a combination of these factors, they will be ex­ mountains in East Africa. From Tanzania, the
amined separately. Their possible combined effect epiphytic flora in dry environments will exemplify
will be considered later (Chapter VI). climatic influences not represented in the investigated
area.

Climate Nimba Mountains


The epiphytic flora in the highest parts of the Nimba
Macroclimate Mountains has few similarities to the one at lower
The large altitudinal differences that exist in the in­ altitudes. The abundance of mosses, lichens, and
vestigated area will naturally give rise to a number of pteridophytes is very obvious (Fig. 96). Several
different macroclimates. For reasons of simplicity species of pteridophytes and orchids are ± restricted
the epiphyte flora of the highest parts ( 1 000- 1 300 m) to the higher altitudes e.g. : Elaphoglossum
will be compared with the one occurring at lower chevalieri, Lycopodium mildbraedii, Xiphopteris
oosora and X. villosissima, Bulbophyllum bifarium,
B. inflatum, B. scariosum, Podangis dactyloceras,
Polystachya dalzielii, P. leonensis, Rangae"ris
brachyceras and Tridactyle tridactylites.
Few species from lower altitudes are present, ex­
ceptions being A splenium dregeanum and Medinilla
mannii.
The proportion between the number of species of
epiphytic pteridophytes and orchids respectively is
also altered with altitude (Table 3 5).

East African Mountains


During a brief study tour to several of the high East
African mountains similar changes were observed in
the composition of the epiphytic flora. On Mount
Kenya, the trees near the forest boundary (at ap­
proximately 3200 m) were devoid of orchids but ex-

Table 35. Number of epiphytic pteridophyte and orchid species


in the foothills (500-700 m) and at the crest ( 1 000- 1 300 m) of
the Nimba Range.

Altitude (m) Pterido­ Orchids Pteridophytes:


phytes orchids

500-700 33 94 1 :3
Fig. 96. Parinari excelsa densely overgrown by epiphytes.
1000- 1 300 28 27 1:1
Nimba Range, 1 3 50 m.

Acta phytogeogr. suec. 59


96 Dick Johansson

Table 3 6. Epiphytes recorded in the Kiwira forest, Mt Rungwe


2200 m, Tanzania.

Pteridophytes
Arthropteris monocarpa
Asplenium aethiopicum
Asplenium barteri
Asplenium geppii
Asplenium monanthes L. Mant.
Asplenium rutifolium (Berg.) Kunze var. bipinnatum (Forsk.)
Schelpe
Asplenium theciferum (Kunth) Mett. var. concinnum (Schrad.)
Schelpe
Asplenium erectum Bory var. usambarense (Hieron.) Schelpe
Drynaria volkensii Hieron.
Loxogramme lanceolata
Lycopodium ophioglossoides Lam.
Oleandra distenta
Pleopeltis excavata (Bory) Sledge
Pleopeltis macrocarpa (Bory) Kaulf.

Orchids
Bulbophyllum stolzii Schltr.
Polystachya zambesiaca Rolfe
Stolzia nyassana Schltr.

Other vascular epiphytes


Peperomia tetraphylla (Forst.) Hook. & Am.

were observed : Polystachya nigresce_ns Rendle and


Ypsilopus sp.
Fig. 97. Moss-covered branches of Hagenia abyssinica
At 1 950 m several orchids were common, e.g.
(Bruce) J.F. Gmel. (Rosaceae). Mt Kenya 3 1 00 m,
Kenya. Bulbophyllum spp., Polystachya campyloglossa, P.
cultriformis, Tridactyle spp. Among the
hibited a few pteridophytes, e.g. : A splenium pteridophytes Drynaria volkensii Hieron. and
kassnerii Hieron., Xiphopteris flabelliformis (Poir.) Pleopeltis macrocarpa (Bory) K aulf. were noticed.
Schelpe and Pleopeltis excavata (Bory) Sledge (Fig. At 2250 m altitude in the Kiwira forest, at Mount
9 7). Pteridophyte nomenclature follow Schelpe 1 970. Rungwe in southern Tanzania, the ratio between the
At 3000 m in the bamboo zone a small species of orchids and pteridophytes was roughly
Polystachya sp. (probably P. campyloglossa Rolfe), one to five (Table 3 6).
· was observed, and at . 2600 m several species of At lower altitudes as in the Usambara Mountains
orchids were present such as Tridactyle scottelii in Tanzania the number of orchid species outnumber
(Rendle) Schltr., Polystachya cultriformis (Thou.) the pteridophytes. At Amani, 950 m, 1 4 species of
"Spreng., and P. campyloglossa. orchids and 1 2 species of pteridophytes were
On the eastern slope of Mount Meru in Tanzania, counted on one Parinari tree (Table 3 7). On another
at 2600-2700 m altitude, only one orchid, tree at the same place 1 1 orchids and 6
Polystachya campyloglossa, was observed. However, pteridophytes were observed.
several pteridophytes, e.g. Asplenium aethiopicum, Epiphytes disappear, according to Tixier
A . friesiorum C. C hr. and A . loxoscapoides B ak. ( 1 966:2 1 ) in tropical humid montane forests, at 3 5 00
were abundant. m in the Himalayas and Africa, and at 4000 m in the
At 2500 m the pteridophytes were very abundant, Columbian Andees and maybe in New Guinea.
e.g. Asplenium aethiopicum, Pleopeltis macrocarpa From Tanzania Moreau ( 1 943 :8) reports : "On the
(Willd.) Kaulf., and Elaphoglossum acrostichoides whole, epiphytic orchids seem most plentiful, in both
(Hook. & Grev.) Schelpe. Only two orchid species individuals and species, between 3000 ft and 5000 ft

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 9 7

Table 37. Occurrence of epiphytes in different sections of a 32


m tall Parinari excelsa, Amani 950 m, Usambara Mts, Tanza­
nia.

I No epiphytes

ll Asplenium nidus L.

Ill
Pteridophytes

Asplenium pellucidum Lam. subsp. pseudoporrectium (Hieron.)


Schelpe
Asplenium megalura
Drynaria volkensii
Elaphoglossum acrostichoides (Hook. & Grev.) Schelpe
Elaphoglossum lastii (Bak.) C. C hr.
Lycopodium warneckei
Microgramma owariensis
Oleandra distenta
Vittaria ensiformis Swartz

Orchids
Ancistrorhynchus refractus Kraenzl.
Bulbophyllum plathyrachis Schltr.
Cirrhopetalum umbellatum (Forst.f.) Hook. & Arn.
Oberonia disticha (Lam.) Schltr.
Polystachya adansoniae
Rangaeris muscicola
Tridactyle anthomaniaca
Tridactyle bicaudata
Fig. 98. Simplified drawing showing the epiphytes utilizing
the humus held by the root system of Asplenium
IV
theciferum on a Nuxia congesta R.Br. (Loganiaceae).
Ngong Hills 2400 m, Kenya. (A) Asplenium theciferum, Pteridophytes

(B) Elaphoglossum macrocarpum, (C) Polystachya cam­


Drynaria volkensii
pyloglossa, (D) Stolzia repens, (E) Peperomia sthulman­
Microgramma owariensis
nii.
Orchids
Ancistrorhynchus refractus Kraenzl.
Bulbophyllum bequaertii De Wild
in rain forest . . . Above 5000 ft in Usambara,
Bulbophyllum encephalodes Summerh.
species became less numerous again with the genus
Bulbophyllum plathyrachis
Polystachya predominating." Diaphananthe subsimplex Summerh.
Polystachya adansoniae
Possible influences of altitude Tridactyle teretifolia Schltr.

The effect of the altitude on the epiphytes has been Other vascular epiphytes
compared with either the water balance, e.g. drought Rhipsalis baccifera
correlated to low temperatures (Went 1 940 :94, Tix­
ier 1 966 : 3 6) or the temperature. V Orchids
Bolusiella iridifolia (Rolfe) Schltr.
The lower the temperature (Table 2) the higher the
Tridactyle anthomaniaca
relative humidity (Fig. 9) and the presence of local
cloud systems or mist, at the highest parts of the
Nimba Range (Fig. 1 1), will naturally influence the
evapotranspiration and favour less drought resistant
species such as filmy ferns, and certain pterido­ Temperature influence on the occurrence of
phytes. Mosses and lichens are also favoured by the orchids has been stressed by Rupp ( 1 969 : 1 ) : "The
higher light intensities that exist at all levels in the epiphytes reach their greatest development in the
more open forest type at the crest of the range. tropics, decreasing in numbers towards cooler

A cta phytogeogr. SUfi?C. 59


98 Dick Johansson

regions. This fact is well illustrated in Eastern Table 38. Occurrence of epiphytes in two different forest types.
Australia, where Queensland h as approximately 1 00 (Trees < 5 m are excluded). Great Ruaha River 400 m, Kidatu,
known species of epiphytes, New South Wales 5 2, Tanzania.

Victoria 5 , and Tasmania 2."


Habitat No. of trees % of trees Max. no. of
There are diverging opinions about the ability of
examined with epiphytes species on a
orchids to survive frost. Piers ( 1 968 :5) states that single tree
epiphytic orchids can not survive frost, while Stewart
Riparian forest 20 60.0 2
( 1 9 70 :54) reports that Bo/usiella iridifolia "may Wooded grassland 24 8.3
even withstand frost on the coldest nights of the
year".
Generally speaking, it is hard to believe that the 10 m from the shore of the Great Ruaha River,
pteridophytes, which as a group are poorly adapted parallel to the river. The second strip was 50 m away
to desiccation, would dominate over the epiphytic from the river 10 m higher on the slope in a
orchids with their advanced technique to survive dry grassland with scattered trees (wooded grassland).
periods, if drought was a main limiting factor on the The two species that occurred in the riparian forest
occurrence of epiphytes at higher altitudes. were Cyrtorchis arcuata subsp. variabilis and
Microcoelia exilis, which were also the only two
Kidatu common species in this area. In the wooded
A thorough study of the epiphytes in the Kidatu area
(1° 40' S, 36 ° 59' E) in Tanzania was performed in
April-June 1 97 1 .
The slopes of the Great Ruaha River Valley, east
of Kidatu, are covered with patches of forest. The
annual rainfall is around 1 000- 1 400 mm (Atlas of
Tanzania 1 96 7 :6), with a long dry season of 4-5
months. Close to the stream bed a riverine forest oc­
cur with a rather large number of tree-forming
species, e.g. Burkea africana Hook., Cassia sp.,
Combretum sp., Dip/orynchus condy/ocarpon (Miill.
Arg.) Pichon, Ficus sycomorus L., Lonchocarpus
bussei Harms, Markhamia obtusifolia (Baker)
Sprague, Milletia bussei Harms, Piliostigma thon­
ningii (Schumach.) Milne-Redhead, Pterocarpus sp.,
Stereospermum kunthianum Cham., Sterculia
quinqueloba �· Schum., Xeroderris stuhlmannii
(Taub.) Medonca & E.P. Sousa, Vitex doniana
Sweet.
m o Aera ngis
A total of eight species of epiphytes were recorded 2
flabel l ifo lia
in the riverine forest : A iirangis flabellifolia Rchb. f., • M icrocoelia
A i/rangis kotschyana (Rchb. f.) Schltr., Bu/bo­ exi l i s

phyllum sp., A nsel/ia gigan tea Rchb. f. var. nilotica

l
(Baker) Summerh., Cyrtorchis arcuata subsp.
variabilis and C. arcuata subsp. whytei (Rolfe)
Summerh., Microcoe/ia exilis Lindl., Polystachya Evaporation
tessallata. The only epiphytic fern was Phymatodes
sco/opendria (one 'stand'), which was partly growing
on some wet rocks in a shady habitat.
Trees along two parallel strips 50 m long and 10
Fig. 99. Simplified drawing showing the distribution of
m wide were examined ('distance' method) (Table epiphytes on a Combretum tree close to the Great Ruaha
3 8). The first strip was situated in the riparian forest River. Kidatu 3 90 m, Tanzania.

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 99

Fig. 1 00. Brachystegia woodland ('Miombo'). Makum­


bako 1 900 m, Tanzania.

grassland A iirangis flabellifolia was the only species


with a total of three 'stands'. The relative 'richness'
of epiphytes in the riparian forest can probably be
correlated to a favourable climatic effect which was
also indicated by a kind of 'reversed' distribution
pattern of the epiphytes on several phorophytes (Fig.
99).
Growing sites close to water are according to Fig. I 0 I . Tridactyle tricuspis. Makumbako I 900 m, Tan­
zania.
Teuscher ( 1 9 72 :497) a general habitat for leafless
orchids. Stewart ( 1 973), however, has pointed out and pseudobulbs remained. The occurrence and total
that a number of Microcoelia species are adapted to number of each species were estimated (Table 39).
different habitats. Three frequences of abundance were used : ( 1 ) less
than five 'stands', (2) 5-25 'stands', (3) more than 25
Makumbako 'stands'. The Polystachya species were very hard to
In the Southern Highland, near Makumbako discover when the old inflorescences were absent.
(8"0 56' S, 34° 5 2' E) at approximately 1 900 m The leafless pseudobulbs were more or less covered
altitude, large areas are covered with a Brachystegia with lichens. Therefore, the frequence of the
woodland (so-called 'Miombo', Fig. 1 00). The yearly Polystachya species are probably too low in this es­
rainfall is 800- 1 000 mm (Atlas of Tanzania 1 96 7 :6), timation.
and the dry season lasts for 5 -6 months. Fifty trees Every tree examined had at least one species of
selected at random, four meters or higher, were ex­ epiphyte. Also surprising is the very high number of
amined for epiphytes. individuals, particularly Tridactyle tricuspis. Sixty­
Two orchid species were observed, Tridactyle four per cent of the trees had 5-25 'stands' of this
tricuspis (Bolus) Schltr. (Fig. 1 0 1 ), and a species. The presence of such a large number of in­
Polystachya sp. of which only the old inflorescences dividuals indicates that they are well adapted to this
environment.
Table 39. Occurrence and abundance of epiphytes on 50 trees The climatic influence in these two areas in Tan­
in a Brachystegia woodland ('miombo'). (Trees < 4 m are exclu­
zania obviously limits the number of epiphytic
ded.) Classes of abundance explained in the text. Makumbako
1 900 m, Tanzania.
species although the abundance of certain species is
surprising.
Species % of trees Abundance Brief observations in even drier areas of Tanzania
with epiph. -------
in the Iringa and Dodoma districts (600-800 mm
2 3
yearly rainfall) revealed that A nsellia gigantea var.
Tridactyle sp. + Polystachya sp. I 00.0 nilotica (Fig. 1 02) was locally rather common, often
Tridactyle tricuspis (alone) 9 8 .0 22.0 64.0 1 2.0 on A dansonia digitata L. and Hyphaene thebacia
Polystachya sp. (alone) 98.0 48.0 48.0 2.0
(L.) Mart. In those very dry areas two species of the

A cta phytogeogr. suec. 59


1 00 Dick Johansson

figures, Jones states : "generally the figures to date


point to the ocean as the main source of rainwater
nutrients" since "the low nitrate figures eliminate the
possibility of large contributions from dust and elec­
tric storms". In Uganda, Visser ( 1 964 : 3 5) observed
that the heavier the shower the more nitrate-N was
precipitated, and therefore states that storms
preceding heavy rains take a lot of dust into the at­
mosphere from which NO) and other nutrients are
later washed.
A similar view is presented by Thornton
( 1 965 : I 025) who found in Gambia that the highest
values for potassium and calcium were obtained at
the beginning of the rains which indicate that the
elements are either washed from or fall in the form of
dust present in the atmosphere. No matter what
origin the nutrients in the rainwater should represent,
a valuable addition to the nutrients of the substrate is
Fig. 1 02. A nsellia gigantea. A large orchid of rather dry made.
areas in East Africa. Kidatu 400 m.

family Crassulaceae, Kalanchoe /anceo/ata (Forsk.) Microclimate


Pers., and K. pilosa Bak., occur as epiphytes. The The microclimate in a rain forest is of course depen­
latter species has also been recorded (J.B. Gillett, dent on the structure and density of the vegetation
1 5 1 23 K) as an epiphyte in the Marsabit area, and is thus very complex. There is a ± continuous
Kenya (780 mm yearly rainfall). vertical climatic gradient from the ground to the
highest treetops.
Rainfall importance Since none of the strata are uniform in density
The rainfall may not only be of importance for the there are also variations in microclimate from place
water economy of the epiphytes but also serve as a to place at the same horizontal level. Furthermore,
source of nutrients. It has been clearly shown that the microclimate in the crowns will vary according
rainfall nutrients play an important part in the to the species and age of the tree.
nutrient cycle of ecosystems on oligotropic sites In a study like this the largest interest is connected
(AUen et al. 1 968 :497). Analyses of the nutritional with the variation of the microclimate in various
content of rainwater are available from two countries strata of the forest, but no such investigation has
in West Africa (Table 40). Jones ( 1 960 :43 2) remarks been possible to perform. For reasons easy to un­
that his nitrogen, potassium and sodium figures are derstand there are in fact few and limited reports of
the highest so far recorded in rainwater. the microclimate in the higher strata of the rain
The origin of the nutrients in rainwater has been forest.
subject to various explanations. Commenting on his From the Shasha Forest Reserve in Southern
Nigeria Evans ( 1 9 3 9) has published results of in­
Table 40. Annual total nutrients (kg/ha) in rain according to vestigations at the ground level and at 24 m level. A
records from West Africa (from Gore 1 968). more detailed study has been carried out in the
lowland rain forest at Banco in the Southern Ivory
Location Rainfall lnorg. p Na K Ca Mg
Coast by C achan ( 1 963). By help of a 46 m high
(mm) N
tower, recordings from four different levels were ob­
N. Nigeria (Jones, 1 960)
1067 54.7 2.58 60. 3 36.76 1 .0 1 2.9 1
tained. The results from these investigations quoted
Gambia (Thornton, 1 9 65) here cannot be simply transferred to the Nimba area.
Yundum 1054 47. 1 0.3 1 9.5 5 . 94 4.37 However, some of the general observations can
Jenoi 1 006 44.9 0.6 1 8.9 4.26 2.69 probably also be applied to the situation within the
Yoroberi K. 709 1 4. 2 0.27 5.8 2.80 1 .6 8
forests in Nimba.
A cta phytogeogr. suec. 59
Ecology of vascular epiphytes in West African rain forest 1 01

Table 4 1 . Humidity of the air during dry and wet seasons at


two levels in an Ivory Coast rain forest (from Cachan 1 96 3).
-- 4 6-D m
- 4 6- 1 "
---- 4&-11 " Level Dry period Wet period
..... ..... 46-26"
/'',,,_ Minimum relative humidity 46 m 50-70 % 60-80 %
\\

\
/
I
/ 1 m 60-80 % 90- 1 00 %
·..
· ....
· · · · · · · · · · · · · ··.
Saturation deficit 46 m 1 2- 1 6 mm 6- 1 2 mm
\\
,, 1 m 6 - 1 0 mm 0-4 mm
··-·· · · · · · · · · �-�:-:':'::.
Hours with rei. humidity 46 m 1 0- 1 4 hrs 8- 1 2 hrs
less than 9 5 % 1m 1 0- 1 2 hrs 0-2hrs
700 goo 1 100 13 o o 1 500 1 700 1 9 oo

Fig. 1 03 . Daily temperature differences between different


levels in a tropical rain forest (from C achan 1 96 3). (Cachan 1 963 : 1 52). The following characteristics
were observed for a dry resp. a wet period (Table
Temperature
4 1 ).
Evans found that the temperature showed a smaller
range of variation in the undergrowth than at the up­ The rate of evaporation at various levels in a rain
per levels of the forest. In the dry season the max­ forest, given as a percentage of the value obtained at
imum temperature is about the same. In the wet 46 m level (above the canopy), showed a steady
season the differences are much more accentuated decrease from the canopy level towards the ground.
and the minimum temperatures in the undergrowth At 26 m the value was 58 %, at 1 1 m 35 %, and at
are in every case from one to two degrees higher 1 m 2 1 % of that at 46 m (Cachan 1 963 : 1 49).
than in the treetops.
The vertical temperature gradient shows, in the Light intensities
middle of the day, a fairly steady decrease from the In the undergrowth
canopy down to the undergrowth. The temperature The instrument used was a photoelectric cell of un­
is 2-3 degrees higher at 24 m than at ground level. At known spectral range. The cell itself was connected
nighttime the undergrowth at 0. 7 m was always to a galvanometer with a 3 m long cable. The light
warmer than the air 24 m above the ground. cell was placed on top of a 1 .3 m high stick, that was
Cachan noted a similar temperature pattern in the pushed into the ground during the recordings. Lower
Ivory Coast. The daily variations of the differences down on the stick two waterlevel instruments were
between the temperature at four levels is shown in attached to ensure that the stick and the light cell
Fig. 1 03 . were held in an absolute vertical position at each
measuring point. The recording was done along three
Humidity parallel strips at 5 m distance from each other, at
The humidity of the air can be expressed in three every 5 m of the 1 00 m long strips. Three recordings
ways : relative humidity, saturation deficit and during one minute at each recording spot were taken
vapour pressure. In general, the minimum relative and the median value chosen. The investigation was
humidity and the maximum temperature are done at noon on a sunny day at two plots, one at
simultaneous, the daily variation of these elements 600 m (Nov. 2, 1 969) and another at 1 300 m (Nov.
are in versed. The variation of the saturation deficit is 1 2, 1 969) Table 42. The samples from 600 m were
in versed to that of the relative humidity, while the
vapour pressure shows a double, daily oscillation Table 42. Light intensities in the undergrowth at two different

with a maximum in the morning and also in the after­ altitudes, given as a percentage of the light in the open.

noon (C achan 1 96 3 : 1 5 2).


Altitude No. of Light intensity
The humidity of the air determined from the
relative humidity, the saturation deficit, and the records Max. Min. Mean
duration of periods with a relative humidity of less
600 m 61 0 . 70 0. 1 7 0. 3 6
than 95 %, are usually identical between the 46 m 1 300 m 63 1 7.24 2.59 6.55
level on a humid day and the 1 m level on a dry day

A cta phytogeogr. suec. 59


1 02 Dick Johansson

taken in a high forest with the canopy around 40 m and was managed by an assistant. The recording
above the ground in the Seka Valley. Common was performed by holding the cell at the growing site
species of trees in this forest were : Parinari excelsa, of the plant to be investigated. To reach the plants
Heritiera uti/is, Calpocalyx aubrevillei and growing in more inaccessible places, the light cell
Chidlowia sanguinea. The ground was covered with was attached to a four m long aluminium pipe. At
dead leaves and the herb flora was sparsely the end of the pipe, where the light cell was attached
represented, e.g. : Selaginella vogelii Spring and at a 90-degree angle towards the pipe, a V -shaped
Selaginella versicolor Spring. On the lower parts of metal piece was connected to a ball bearing on the
the trunks a few ferns were noticed, Hymenophyllum pipe. When the metal piece with its open end was
kuhnii and A splenium barteri. held against the substrate, the whole pipe could be
At 1 300 m the samples were taken at the crest of turned around on its axis, and thus moving the light
the main ridge 300 m SW of the point where the cell at the same time. By placing the light cell at the
borders of Liberia, Guinea and the Ivory Coast actual growing site of the epiphyte, and then turning
meet. The forest in this locality almost exclusively it, the maximum light intensity was found.
consisted of Parinari excelsa with a somewhat The tree chosen was a Parinari excelsa growing in
broken canopy at 1 0- 1 2 m level above the ground a typical section of the forest that covers the narrow
which was irregularly covered with herbs. crest of the main ridge at 1 300 m altitude from the
The high light intensities in the montane forest is mark of the three national borders down to the min­
remarkable, but judging from the abundance of ing area. The canopy in this single dominant
plants on the ground it was not totally unexpected. Parinari forest was at 1 0- 1 5 m above the ground
In a montane forest in Ecuador, Whitmore and partly open. There were a few trees and shrubs
( 1 968 :238) found that the canopy was more of smaller dimensions. e.g. Ochna membranacea
transparent than in a lowland forest. Oliv., reaching up to the lower parts of the crowns of
Particular interest has been paid to the illumina­ the emergent trees. The tree investigated had a size
tion of the undergrowth and the effect of the sun­ of 23 m and a circumference of the trunk at breast­
flecks on the forest floor, e.g. Evans ( 1 939) in height of 1 62 cm and was heavily branched, with the
Nigeria, Gusinde & Lauscher ( 1 94 1) in Congo, Eid­ first ramification starting 3 . 5 m above the ground.
mann ( 1 94 1) on Fernando Po, Evans et al. 1 960 and The foliage was concentrated to the outermost
C achan ( 1 963) in the Ivory Coast. The shade il­ branches which made movements easier in the cen­
lumination was found by all of them to be 1 % or less tral part of the crown. At the basal part of the trunk
of the outside light. The light intensity in the sun­ the cover of epiphytic mosses and filmy ferns was
flecks in British Guiana varied between 1 0.4 and 1 00 %. Higher up on the trunk ( 1 -3.5 m) the
72 % (C arter 1 934). The spectral composition of the epiphytes were more or less concentrated to one side
light reaching the forest floor as examined by Evans (S), with the plants appearing in small patches with
( 1 939) showed approximately 40 % portion of the clean bark surface in betwe(;!n. In the basal parts of
wave-lengths above 7000 A (infra-red light). Similar the crown Tridactyle tridactylites formed dense, un­
observations have been reported from Puerto Rico tidy 'stands'. In the central part of the branches the
(Johnson & Atwood 1 970). This indicates a con­ mosses e.g. Porothamnion hildebrandtii (C.M.)
siderable increase in the transmission of the forest Fleisch, were very abundant, forming large cushions.
canopy just beyond the red end of the visible spec­ In the apical part of the branches the lichen cover
trum. The effect of this portion of light on the plants was pronounced especially on the twigs, where an
is unknown but it can probably not be used in the Usnea sp. with hanging growth habit was noticed.
photosynthesis, ·since it is outside the range of Bulbophyllum scariosum and Polystachya dalzielii
chlorophyll's absorption. As the forest became were also abundant.
denser a decrease in the proportion of blue light was A total of 1 5 species of epiphytes were growing on
also noticed. the tree, 7 pteridophytes, 5 orchids and 3 other
vascular epiphytes. Thus the pteridophyte/orchid
A t the growing sites of the epiphytes ratio was 1 .4 :1 . A total of 1 1 9 recordings, which
The same photoelectric cell as described earlier was were estimated to cover 80 % of the plants present,
used. The recording equipment stood on the ground were performed during a 3 hours period in the mid-

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 1 03

Table 43. Light intensities as a percentage of the light in the between maximum and minimum intensities should
open at the growing sites of epiphytes on a Parinari excelsa. be noted with caution, since sun-flecks with a very
Nimba Ridge, 1 300 m. short duration may be responsible for the high
figures.
Epiphytes No. of Light intensity

records Mean Max. - Min. At various levels above the ground


The general knowledge of light conditions in the
I + 11
Asplenium dregeanum 9 1 4.2 1 6. 8 - 1 2. 6
crowns of the trees in the tropics is very scanty. The
Begonia rubro-marginata 2 1 7. 7 23.2 - 1 1 .9 need for the erection of large structures to reach the
Bulbophyllum inflatum 5 1 3 .5 1 9. 7 - 7.1 upper parts of, or preferably above, the canopy has
Elaphoglossum chevalieri 6 1 5 .2 24. 5 - 7.1 naturally been difficult to overcome. There are ac­
Medinilla mannii 5 1 3. 2 1 7.4 - 9.4
tually only a few such investigations (Allee 1 926,
Peperomia sp. 8 1 6. 1 2 3 . 2 - 1 0.3
Polystachya leonensis 6 1 4.2 1 8. 4 - 9.0
Dirmhirn 1 96 1 and C achan 1 96 3). The decrease in
the light intensity from the canopy towards the
Ill
ground is by no means a gradual process, but rather
Asplenium dregeanum 5 26.8 29.3 - 23.2
Lycopodium mildbraedii 4 20.0 26.8 - 9.3
abrupt changes occurring at various levels, depen­
Medinilla mannii 6 25.5 29.7 - 1 6. 8 ding on the structure of the forest (Richards
Peperomia sp. 3 2 1 .6 28.4 - 1 6.4 1 93 9 :30). C achan ( 1 963 :93) reports on the vertical
Tridactyle tridactylites 7 2 1 .6 29. 7 - 1 5 . 8 variation of the light intensity from a tropical rain
IV forest in the Ivory Coast. The emergent trees in this
Asplenium aethiopicum 2 24.5 25 . 1 - 23.9 forest reach a height of little more than 40 m, e.g.
Asplenium megalura 4 28.4 36.8 - 1 9.0 Combretodendron africanum, Piptadeniastrum
Bulbophyllum scariosum 6 27.7 3 6.4 - 1 6.4
africanum, Guarea thomsonii Sprague & Hutch. and
Begonia rubro-marginata 2 26.4 28.7 - 23.9
Elaphoglossum isabelense 3 2 1 .0 23.9 - 1 8 . 7
Turraeanthus africanus (Welw. ex DC.) Pellegr. The
Polystachya dalzielii 8 29.3 3 7 . 1 - 24.5 light intensity above the canopy varies continuously,
Tridactyle tridactylites 5 22.3 27.2 - 1 8.4 since the sky is more or less hazy. Values above
Xiphopteris vilosissima 4 1 7. 7 22.6 - 1 3 .9 1 OOJ,OOO Lux were exceeded during rather short
V periods. The daily variation in the light intensity at
Bulbophyllum scariosum 11 3 7. 4 43.2 - 28. 1 44 m and 33 m above the ground, is presented in Fig.
Polystachya dalzielii 7 37.7 45.5 - 3 2.9 I 04. Wide variation in the light intensity occurs in
Usnea sp. 7 42.6 5 8 . 7 - 30.0
the middle of the day at 46 m, where the daily
die of the day. An additional 7 recordings of the light - 0 00 LUX

intensities at the more exposed positions of the


typical Usnea lichens were also included for com­

(\
60
parison (Table 43).
It is interesting to note that the arbitrary division
of the trees into 5 sections that has been used
throughout this study, was in this tree not correlated I �
to the light intensities (Table 44).
No marked difference in the light intensities could �J �

\
30

be found between the basal and middle parts of the


branches. The very wide variation in the values

Table 44. Light intensities in the five sections of a P arinari ex­


3 3 rn
celsa at 1 300 m. Based on the mean light intensity in each sec­
tion (Table 43 ) .

goo 11 00 IJOO 15 0 0 1700


Section I + ll Ill IV V
Fig. 1 04. Variations in daily light intensity above the
Light inten­
canopy (4 6 m above the ground) and at 33 m in a tropical
sity 1 4.9 23. 1 24.7 3 7. 6
rain forest in the Ivory Coast (from Cachan 1 963).

A cta phytogeogr. suec. 59


104 Dick Johansson

Daily variations in light intensity


Through a special 'writing unit' the daily continuous
changes in light intensity were recorded. The same
light cell as previously described was simply con­
nected to the 'writing unit' through a cable.
The light cell was placed at the growing site of the
epiphyte to be investigated pointing in the direction
of the supposed main light influx, which in most
cases meant a vertical direction (Fig. 1 05). The
graphs obtained with this equipment should only be
compared with each other, since the spectral range
of the light cell was unknown. No extra instruments
that could be used for simultaneous recordings in the
open were available. Therefore, all the curves shown
are from different days, which of course include a
source of error. To reduce the error the comparisons
shown are all taken from cloudfree days. The
seasonal effects must also be considered, since the
recordings are from different times of the year with
different solstices. The aim of these recordings was
to find out if there were differences between various
species of epiphytes in their light climates over a
longer period, and to get an idea of the frequence and
duration of direct light (sun-flecks). The light inten­
sities of three species of epiphytes, Diaphananthe
Fig. 1 05 . Daily variation of light intensity measured by a pellucida, Vanilla crenulata, A rthropteris monocar­
light cell at the growing site of a Diaphananthe pellucida
four meters above the ground (see text).
pa, from the lowest parts of a high forest (Seka
Valley 600 m) compared with the light outside the
forest is given in Fig. 1 06. The Diaphananthe
average was 3 8 ,000 Lux. The average light intensity
specimen was growing on a liana, four meters above
at 3 3 m level is less than 1 0 % (3082- 3 8,000 Lux) of
the ground (Fig. I 05), and the Vanilla vine on the
the intensity above the canopy. Observe that in such
rotten stump of a large tree at about 1 .5 meters
a short vertical distance as 1 1 m a 90 % reduction of
height, while the A rthropteris monocarpa specimen
the light intensity has already taken place. At ground
occupied the basal part of the trunk of a large tree.
level, the light intensity never reached above 1 % of
The daily variation of the light intensity in the
the light above the canopy.
open on a clear day shows a fairly even rise and
decline with the maximum in the middle of the day.

1 I n the open
2 D i apha n a nthe pel l u cida
3 V a n i l l a c re n u l a ta
4 Art h ropte r i s m o n o c a rpa

Fig. 1 07. Daily variations in light


intensity at the growing site of a
Bulbophyllum inflatum. Nimba
06.oo 12 .oo 1 8 _00
Range 1 300 m.

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 1 05

18 . 3

06. 0 0 12.00 1 8 .0 0

Fig. 1 07 . Daily variations in light 20 . 3


intensity at the growing site of a
Bulbophyllum injlatum. Nimba �
Range 1 300 m. 06.00 1 2 .0 0 1 8.oo

The minor variations that can be observed Bulbophyllum inflatum, growing on a trunk I .5 m
throughout the day are probably caused by haze. above the ground at 1 300 m altitude, give the varia­
The more gentle peaks in the curve for tion in light intensities for the orchid which grows in
Diaphananthe pellucida probably indicate that this the most shady habitat of all the orchids in the Nim­
species recieves a more indirect type of light. The ba area (Fig. I 07). These two curves can also serve
series of distinct peaks with a very short duration as an example of the 'micro light climate' that
that occur between 1 2°0- 1 4 °0 on the Vanilla curve is probably is of great importance for the epiphytes.
a result of direct sunlight that has been able to Note the similarity between the two curves of March
penetrate down to this level. In denser shade closer 1 8 and 20, 1 970 (March 1 9 was a cloudy and rainy
to the ground the penetration of direct light is less day) when the light that was registered reached the
likely. The light level tends to be more uniform plant between 1 4°0- 1 530• For many plants that live in
throughout the day as can be seen for A rthropteris a superficially dark environment these short periods
monocarpa. of higher light intensities may be of critical impor­
In the montane forest the light regularly penetrates tance in their survival. It may also influence the oc­
the lower parts of the vegetation. The curves for currence of small groups of several species that com-
1 A s p l e n i u m m e ga l u ra
2 Aspl e n i u m afri ca n u m

3 Aspl e n i u m b a rteri

06.00 1 2.oo 18.oo

1 P o l yst a c hy a pa n i c u l a t a
2 P o l ysta chya l a x iflora
3 Po l ysta chya ra m u l os a

Fig. 1 08. Daily


variations in light inten­

sity at the growing sites


of (A) three
pteridophyte species,
(B) three orchid species.
12.oo 18.00
Nimba Range 600 m. 06 .oo

A cta phytogeogr. suec. 59


1 06 Dick Johansson

monly are found in patches of small and restricted As p l e n i u m O l e a nd ra P l a t yceri u m


range. To find out if there are any correlation b a rteri i · d i ste nta a ng o l ense
between the observed stratification of the epiphytes
and the light -intensities, three species in each of the

J.
genera A splenium and Polystachya were chosen for
study. One species in each genera represented a
specific level in the rain forest. The lowest level (sec­
tion I) was represented by Asplenium barteri and
Polystachya ramulosa, the intermediate level (11) of
Asplenium africanum and Polystachya laxiflora, and
the crown stratum (Ill + IV) of A splenium megalura B ra c hyc o ryt h i s Po lyst a c hya B o l u si e l l a
and Polystachya paniculata. ka l b reye ri i l a xifl o ra ta l bo t i i

Two 'different phorophytes that each carried these


three plants of the same genus were selected for the
study. The light cell was then applied in the same
way as earlier described. The result is presented in
Fig. 1 08. The differences in light intensities as
observed for certain species with the aid of recording
instruments (short or long period recordings) seem to
confirm the estimations made in the field.
A limitation to one or rarely two of the particular C a l voa B e go n i a R hipsalis
m o n ti c o l a ru b ro­ b a ccife ra
light groups that are recognized was common among
1 0 0 '/o m a rg i n a ta
the species in all major groups of epiphytes (Fig.
1 09). The majority of pteridophytes, however, occur
in diffuse light while the orchids are more evenly
50
divided between habitats with diffuse or bright light
(Table 46).

Substrate
Fig. 1 09. Records of epiphytes distributed between the
different light classes (from Table 27, Brachycorythis
PROPERTIES OF THE SUBSTRATE AND kalbreyeri from Table 29). (A) heavy shade, (B) open
THEIR INFLUENCE ON THE EPIPHYTES shade, (C) full sun.

The bark cases is proportional to the age and roughness of the


The properties of the bark affect the epiphytes in bark) and the number of epiphytic species (Fig. 69).
several ways, e.g. through ( I ) relief and growth habit On trees 10- 1 5 m in size, e.g. Lophira a lata,
(2) structure or porosity, and (3) chemical composi­ Triplochiton scleroxylon, Ceiba pentandra, there are
tion. When comparing the relief of bark between seldom any epiphytes, although these species when
various species of trees one must remember that the older are very rich in epiphytes. This refers to trees
bark by no means is uniform over the entire tree but of the primary forest as well as those in more open
varies with age. environments. However, several species of epiphytes
( I ) The importance of the bark's relief can be seem to have a marked 'preference' for smooth bark
suspected to be connected with the establishment of surfaces, e.g. A ncistrorhynchus cephalotes (Fig.
the seedlings. A deeply fissured bark could provide 1 1 0), A ngraecum podochiloides, and Polystachya af-
both a suitable microclimate for the germination of finis.
seeds and spores, and also more easily prevent the From other parts of the world observations of a
young seedlings from being washed away. rich epiphytic flora on trees with rough bark are
In the Nimba area a clear correlation has been es­ frequent (Oliver 1 930: 1 4, Eggeling 1 947 : 5 6). Went
tablished between the size of the tree (which in most ( 1 940 :9 5) observed that phorophytes with a poor

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 1 0 7

palms they readily attach themselves and seem to


thrive.
Sulit ( 1 950:8) even states that the 'palm trees' are
especially suitable for growing orchids on.

(2) The structure or porosity of the bark will n atural­


ly effect its water capacity.
The absorbation ability of the bark varies natural­
ly from tree to tree, depending on age, species and in­
clination (Freise 1 93 6 :303, Voth 1 939). The subse­
quent water content of the bark is also dependent on
the amount and type of rainfall, and the rate of
evaporation, and thus varies with weather conditions
and the season. Another fact to consider is that
water does not flow evenly over the bark surface but
is concentrated into so-called rain tracks or drainage
channels (Richards 1 964 : 1 1 8, Yarranton · 1 967}.
The water content of the bark is thus dependent
on a large number of variable factors. Measurements
must therefore be carried out during a long period
and involve samples from different sections of the
trees. This study does not include measurements of
the water capacity or the water content of the bark.
Water capacity of bark have been investigated in
various parts of the world with different methods.
The results, however, show large variation between
Fig. 1 1 0. A ncistrorhynchus cephalotes (lower down) and trees of the same species (Barkm an 1 9 5 8 :7 5-7 6).
Graphorchis lurida (above) are two epiphytes which often (3) No analyses of the chemical composition of bark
occur on bark without humus deposits. Seka Valley 700
have been undertaken, but pH measurements of bark
m.
samples in water suspension were performed.

epiphytic flora frequently were found among species Bark pH and the epiphytic flora
with smooth or hard bark, and fast growing species. The samples were taken from representative areas in
Boyer ( 1 964) concluded that it probably was the the vicinity of epiphytes in various sections of the
smooth nature of the bark that was the cause of the phorophyte. They were all taken from trees felled the
absence of Platycerium ferns on Musanga Smithii. same day, except for a living Parinari excelsa tree at
In the Nimba area trees with a defoliating b,ark (e.g. 1 300 m altitude which was climbed. The bark pieces
Distemonanthus benthamianus), seem to be very dif­ used in the survey had a size of 6- 1 0 cm2• They were
ficult for the epiphytes to colonize. This has also pulverized and dried in 24 hrs in 5 5 % relative
been observed elsewhere (Dudgeon 1 923, Pessin humidity. Distilled water twice the dry weight of the
1 925 :34, Oliver 1 930: 14, Went 1 940 :95, Rupp sample was added. The measurements were per­
1 969 :XI). Voorhoeve ( 1 96 5 :200), however, states formed 24 hrs later on an electric pH-instrument.
that colonization of flaking bark does occur, and in The sampling procedure, e.g. cutting a piece of bark,
some cases the epiphytes even prevent the shedding involves a great deal of subjectivity, since the actual
of the bark, e.g. on Gilbertiodendron preussii. The choice of bark pieces may influence the result. It was
absence of epiphytl c orchids on the oil palm is hard found that dead, corky pieces gave higher pH values
to correlate to any morphological characteristics of than living bark, a fact that is well known. There are
the phorophyte, but may possible be attributed to also difficulties in utilizing a completely identical
germination difficulties of the orchid seeds. When method of sampling due to the variation in the bark
full-grown epiphytic orchids are transplanted on oil morphology. Some species have a thick easily

Acta phytogeogr. suec. 59


108 Dick Johansson

pH dent (Barkman 1 95 8 : 1 03). The acidity may influence


3 4 5 6 7
the germination of spores and seeds, or affect the
I I
11
F a g a ra
Ill : I growth of established plants. Pessin ( 1 925 :24),
tess m a n n i i IV
...
however, remarks : "It appears that the acidity or
V

. . . • .•
I 1
H e r i t i e ra 11
.. alkalinity of the substratum does not influence the
.I • •
Ill
uti l i s :
IV
V

I
11 •• •
occurrence and distribution of Polypodium
Loph i ra : 1: I
.. .. . polypodioides."
Ill
a l a ta ..
IV
V

I I
11
M itra g y n a .
. . ..
Ill
c i l i a ta . I Exudates
IV
V

1••
P a r i n a ri
I
11 I
.. . .
"Orchids seem to dislike trees which exude a latex
..
Ill
excel sa IV
V : I• juice such as most Ficus species and tree Euphor­
I•
I
P a ri n a ri 1 3 00 11
.. bias." (Piers 1 968 :4).
... .
Ill
excel sa m
..
IV
V

I
I
I ..
In the Nimba area several species of phorophytes
..
11 11
P i pt a d e n iastru m Ill
:i which · exude a latex juice have a rather rich epiphyte
africa n u m ..
IV
V
flora, e.g. Chlorophora regia, Uapaca guineensis and
even the rubber tree, Hevea brasiliensis, while others
Fig. 1 1 1 . pH in water extracts of bark from different show an opposite tendency, e.g. the figs.
phorophytes. Yekepa. The toxic effect of certain exudates must also be
detachable bark while others have a thin flaky, corky considered. Barkman ( 1 95 8 : 1 34) suspects the 'robin'
or deeply fissured one. Differences in pH value for in the bark of the R obinia tree to be the cause of the
samples of large and small pieces of bark have also poverty of non-vascular epiphytes of this tree.
been reported (Sjogren 1 96 1 : 1 02). Large differences in the presence of orchids
The pH values exhibit a ± significant pattern between various species of trees of the genus Quer­
within the phorophytes (Fig. 1 1 1 ). The lowest values cus were observed by Frei ( 1 9 7 3 a : 3 1 1 -3 1 2) in a
were obtained from the upper part of the main trunk cloud forest in Mexico. Two species in the genus had
and the outer branches. only a few species of orchids, two. others carried
The pH of the bark has been studied by many many orchids, and one was never observed with any
authors using various methods. It has been observed orchids whatsoever.
that the pH values decrease from the bases of the When analyzed, the bark from the species with
trees and upward, probably due to decreasing dust many orchids showed no presence of inhibitory sub­
content. The pH has also been found to vary accor­ stances. The bark from the phorophytes devoid of
ding to the time of day (Pessin 1 925 :34). orchids, contained gallic and egallic acids, which are
Du Rietz ( 1 945 : 1 98- 1 99) distinguishes three known to be growth inhibitory substances
groups of bark based on the pH values, 'rikbark' (hydrolyzable tannins). (See also Frei 1 973b : 70 1 -
(rich bark) pH 5-7, 'overgimgs-fattigbark' 708.)
(intermediate bark) pH 4-5 and 'extrem fattigbark' B ark from several tropical species of trees e.g.
(extremely poor bark) pH <4. When trying to Ficus pumila contain strong growth inhibitors
evaluate the influence of the pH value of the bark on (Bovey & Diaz-Colon 1 959 :256).
the epiphytic flora one should remember that the pH The importance of the leaching of growth­
value might have changed since the time when the regulating substances from the vegetation has also
seedlings were established. The long span from ger­ been stressed by Tukey Jr ( 1 970 : 1 60).
mination to a fully developed plant exemplified by Went ( 1 940 :90) assumed that the rich epiphytic
the orchids emphasizes this problem. flora on Castanopsis argentea could be connected to
The epiphytes themselves influence, in a complex the high amounts of tannin in the water on the bark
and imperfectly known way, the acidity in their en­ surface. The excretion of water-soluble substances
vironment. This can be done by collecting dust and by the bark or the leaves has seldom been measured.
forming humus, through photosynthesis, by release Schweitzer (quoted from Ruinen 1 95 3 : 1 5 1 ) . found
of lichen acids and exchange of ions (Kolkwitz 1 9 3 2, that a total leaf surface area of one sq. m of coffea
Skye 1 968 : I 07). The acidity also regulates the leaves, leached in distilled water during one hour,
nitrification since this process is strongly pH depen- yielded 2 1 -3 5 mg mineral substances, Erythrina

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 109

3 3 . 8 mg, Hevea · 5 5 mg. In many instances sugar L R a i nwater

could be detected in rainwater dripping from the


leaves. Rain wash in a mature forest in Ghana
carried the following amounts of nutrients out of the
vegetation, in lb/acre : N 1 1 , P 3, K 1 96, Ca 26, Mg �b r i s from

1 6 (Greendale & Nye 1 964 :248). See also Eaton et


I
tr.e p h o rophyte
L_ _____J
al. ( 1 9 73). Ruinen ( 1 95 3 : 1 5 1 ) states that germina­
tion and settlement of the epiphytes are possible even I M t r l tra nsported
from the ground
in a milieu with minimal nutrient value, because of by Excreta fr:Jm
a n t s and termi tes
a n imals
the rich substrate offered by the living tissues and ----

their excretions.
An efficient adaptation to intercept leachates from
the overhead canopy is shown by certain bromeliads.
It has been shown experimentally that metabolites
fall in the well (formed by the diverging leaf bases)
from which they are absorbed and utilized in growth
(Tukey Jr 1 9 70 : 1 5 5 - 1 60).

Humus deposits
Origin and composition
The accumulation of humus starts with the decom­
position of the outer layers of the bark. The bark is Fig. 1 1 2. O rigin of substrate and nutrients of importance
for the epiphytic flora.
continuously renewed from within, so that it
automatically provides a steady supply of new organic in origin. This can be compared to the soil
material for decomposition. This is m ainly carried under the trees containing 3 .0- 3 . 9 % of organic
out by fungi, which in many cases are connected to matter. I 0-23 % of the material belonged to a frac­
the roots of the epiphytes. tion less than 2 mm in diam. The rest consisted of
To the material from the bark external material is living, or dead and more or less decomposed parts of
added. Dead leaves, twigs, small branches, etc., from leaves and roots.
the phorophyte itself can be assumed to form the The accumulation of organic material through
bulk of the deposits. In this more or less accidental wind, animals, and water running along branches
process the growth habit of the phorophyte and and stem amounted to 5- l 0 % of the total weight.
structure of its bark are of importance. The root The bulk of the organic material was derived from
systems of the epiphytes also help to catch debris in the tissues of the fern itself. The addition of organic
the many niches they form. Certain epiphytes play a material from sources outside the fern was con­
more active role in the catching of falling debris. sidered to play a minor role in the nutritional supply
Usually a niche between the plant itself and the sub­ for this epiphyte.
strate is formed in which debris from above can be The formation of the humus was slow. It was es­
caught, e.g. Drynaria laurentii, Platycerium spp. tim ated that it takes 1 0 years to form a layer of
(Fig. 1 1 2). The term 'sole epiphytique' has been used I 0 cm thickness. It was also noticed that the material
by Boyer ( I 964) for the humus collected by that was incorporated was only slowly decomposed
Platycerium ferns. She is of the opinion that these ac­ despite the rich microflora. The external contribution
cumulations of organic material can hardly be com­ to the assemblage of humus may be small in volume
pared to the soil in its proper sense. These ac­ but important in its nutritional effect on the epiphytic
cumulations have attracted the interest of several flora.
authors, e.g. Miehe ( I 9 1 1 ), Paulian ( 1 945) and Dust carried by the winds may be Jf considerable
Klinge ( I 9 6 3). importance in certain areas. Every year in the winter
In the Ivory Coast Boyer ( 1 9 64) found that in the months the Harmattan winds fill the air with a fine
humus of the 'leaf basket' of Platycerium angolense dust in vast regions of West Africa. It is easily
and P. stemaria 6 3 .4-90. 9 % of the dry weight was observed that trees close to dusty roads often show a

Acta phytogeogr. suec. 59


1 1 0 Dick Johansson

'/, but prevented from contact with rainwater. The


100
A D ryna ria l a u r e n t i i
weight of the samples were taken every second day
a M i crosori u m p u n c t a t u m
during a 30-day-period. The water holding capacity
c P l athyce ri u m a ng o l e n se
for the humus is shown in Fig. 1 1 3. It is obvious that

�A the h u mus deposits may serve as a water reservoir.
The ability of rapid water absorption during a
50

+�

855
shorter rainfall and the slow release of it is naturally
of great importance for the epiphytes.
30 Cs5

0
pH of humus deposits
0 15 30
Thirteen different humus deposits uti lized by
DAYS
epiphytes were examined. The methods used in these
Fig. 1 1 3. Water holding capacity o f humus from different determinations are the same as described in the in ­
epiphyte communities, in o/o of the wet weight. vestigation of the bark, i.e. the bark piece has simply
been replaced by a piece of humus (50 g dry weight).
l u xu riant growth of epiphytes (Thorold 1 9 5 2 : 1 3 9, Rather high pH values were obtained, con­
Richards 1 964 : 1 1 4 ). Ants and termites carry siderably higher than those of the bark, Table 45.
material of inorganic as well as organic origin up Schnell ( 1 95 2 :5 8-5 9) gives the pH in the humus of·
into the trees. Termite nests built from earth, and the a Bulbophyllum sp. as 4.8 and for a Listrostachys sp.
earth-covered tracks of these insects are common on to 5 . 7. Boyer ( 1 964) reports the pH of humus from
the trees. Richards ( 1 964: I l l) states that the roots Platycerium stemaria as 4. 6-6.4 (6.0 for P.
of flowering epiphytic plants and ferns are one of the angolense).
chief nesting places for arboreal ants. Adamson
( 1 943) suggested that certain epiphytes in the rain Correlation between substrate and epiphytic flora
forest of Trinidad m ay be at least partly dependent Many epiphytes show a high frequence in the recor-
on soil carried up into the trees by termites. In an
Table 45. pH in water extracts from humus deposits with gro-
analysis of the mineral content in the humus from a wing epiphytes. Median value of five samples.
Platycerium fern in the Nimba Mountains Schnell
( 1 95 2 :58) found that 3 8 % of the mineral content Phorophyte Section pH Epiph. utilizing the humus

had a granular size of gritty sand and 68 o/o was of


lesser sizes. The mineral particles appeared as grains Heritiera utilis IV 6.6 Drynaria laurentii
of quartz, hematite, and loam. Schnell did not Nephrolepis undulata

elaborate on the origin of the larger sized particles Heritiera utilis V 6 .0 Drynaria laurentii
Nephrolepis undulata
that can hardly be transported by the wind, but it
Lophira alata Ill 5.8 Vittaria guineensis
can be mentioned that the earth in this particular Lophira alata IV 7.0 Oleandra distenta
region is very rich in hematite. fridactyle armeniaca
Lophira alata V 6.8 Drynaria laurentii

Water holding capacity Polystachya polychaete:


Parinari excelsa lii 6.0 Oleandra distenta
Pieces of humus of roughly the same volume
Arthropteris orientalis
( 1 .5 dm3) ( 1 00 x 1 00 x 1 5 0 mm) were cut out from a Parinari excelsa IV 5.6 Drynaria laurentii
humus deposit in a Drynaria laurentii, a Microsorium Microsorium punctatum
punctatum and a Platycerium angolense epiphyte Parinari excelsa Ill 6.8 Asplenium megalura

community. Four samples from each type of humus Peperomia sp.


Parinari excelsa IV 6. 1 Asplenium dregeanum
were taken. The pieces were placed in baskets of
Polystachya leonensis
'chicken wire' and then saturated with water for 3 Piptadeniastrum Ill 6.2 Asplenium geppii
hours. The water was allowed to drip off, and when africanum Begonia polygonoides
no more water was given off, the weight of the twelve Piptadeniastrum IV 6.3 Oleandra distenta

samples were taken. The samples were divided in two africanum Microsorium punctatum
Triplochiton scleroxylon IV 5.5 Platycerium angolense
groups, one was placed in an air-conditioned room at
Triplochiton scleroxylon IV 4.8 Platycerium stemaria
5 5 % relative humidity, the other placed outdoors
A cta phytogeogr. suec. 59
Ecology of vascular epiphytes in West African rain forest 1 1 1

D ryn a ri a Pyrros i a Asp l e n i u m O l eandra


Table 46. Most frequent light-substrate class at the growing
l a u re n t i i m echowii m ega l u ra d i stenta
sites of each epiphyte species. Given as a percentage of the total
number of species in each of the three groups. Based on Table
27 a and b, for species with 10 records or more. For species
with less records the sum of records from Table 27 a and b, and
Table 28 a and b are used if it reach 10 or more. For the remai­
ning species the sum of records from Tables 2 7 a and b, 28 a
and b and 29 are used.

B u l b o p hy l l u m B u l bophyl l u m B u l bophyl l u m Polysta chya


b u nti n g i i b ufo cocoi n u m tessa ! l ata Substrate Light Total,
substrate
Heavy Open Full
shade shade sun

The Pteridophytes (3 9 spp.)


Large humus
deposits 7.7 1 0. 3 2.6 20.Q
Minor humus
R h i psa l is M ed i n i l l a Rem usatia Pepero m i a
deposits 1 0. 3 38.5 2.6 5 1 .4
bacc ife ra mannii vivipara sp .
10 0'/,
Bark 5. 1 5.1 1 7. 9 28. 1

Total, light 23. 1 5 3.9 23. 1 1 00. 1


50
The orchids ( 1 0 1 spp.)
Large humus
deposits 1 .0 7.9 1 .0 9.9
Minor humus
Fig. 1 1 4. Records of epiphytes distributed between the deposits 1 .0 25.7 1 2. 9 39.6
different humus classes (from Table 27). (a) bark, (b) Bark 1 5.8 34.7 50.5
minor humus deposits, (c) large humus deposits.
Total, light 2.0 49.4 48.6 1 00.0

The other vascular epiphytes ( 1 3 spp.)


dings from only one substrate class (of the three that
Large humus
are recognized), as can be assessed from Tables 27- deposits 1 5 .4 38.5 53.9
29. Less frequently an even distribution between the Minor humus
substrate classes is observed. This 'preference' for a deposits 30.8 7.7 38.5

particular type of substrate occurs among all the Bark 7.7 7. 7

major groups of epiphytes (Fig. 1 1 4). Total, light 1 5 .4 7 7.0 7.7 1 00. 1
Most pteridophyte species (5 1 %) grow in minor
humus deposits. The orchid species occurred on bark
(50 %). The other vascular epiphytes were found in will be used (Table 4 7). Phorophytes with a high
large humus deposits (54 %) (Table 46). number of species of epiphytes are :
Regarding the interrelationship between epiphytes A mphimas pterocarpoides Cryptosepalum tetraphyllum
and phorophytes three case may be distinguished in A ntho_notha fragans Entandrophragma utile
Calpocalyx aubrevillei Heritiera uti/is
the area studied (denoted A, B and C below).
Canarium schweinfurthii Lophira alata
Ceiba pentandra Mitragyna ciliata
A. Phorophytes with an abundance of epiphytes Chlorophora regia Parinari excelsa
In the Nimba area a 'preference' by epiphytes for Combretodendron macrocarpum Piptadeniastrum africanum

colonizing certain species of phorophytes is easily Coula edulis Triplochiton scleroxylon

observed, but somewhat more difficult to express.


The abundance can be expressed as previously An often striking difference in the abundance and
described, either by the number of species of floristic composition of the epiphyte vegetation
epiphytes on a certain species of phorophyte' or the depending on the species of phorophyte is well
number of ' stands' of epiphytes on the individual documented from various parts of the world.
phorophyte. Since the results from these two In the Phillippines the 'preference' for certain
methods correlate for the most part, the first method phorophytes that the orchids exemplify, has even led

A cta phytogeogr. suec. 59


1 1 2 Dick Johansson

Table 47 a. Records of pteridophytes and other vascular epiphytes growing on different phorophytes in the Nimba a r e a .

<f)
"'
H
.s
.s:::
bl)
:;:;
"'
.s:::

"'
"0

� �]
�- 3 .g
E o �

_g � �
0.. C.) U
Pteridophytes

X • + X " 0 0 " 10
A n trophyum i m m e rsum
. • •
A. n1anntanun1 0 + 7
A r thropteris monoca rpa + X + + + + + X + + + 20

A . orienta l i s + · o + + + x + + X x · x · + x x + o + X + 29

A . palisoti I
A splenium :lethiopicum X " 4
A. nfricanum X + • + + X 0 · 14 . 3

A. barteri + + x · x + + X 20
A. drege:mum + + 7
A . geppii + X + X 0 + " 23 + 2
A. hemitomum 4

A. megalura + • X 0 X 0 0 x x + x x o + X X + 34 + + 4
• • • 0 0 • • •
A. variable var. paucijugum
e
Da va Ilia cha rophy \ loides • + 0 X

+ + 0 • • + 0 • 0 • 21 + + 4
Drynaria \au re n t i i + x x x + x x x o x x x x x x 0 X X X X X X X X X + X X X X X X X + + X + + 43
Elaphoglossum burteri X ' • + + X ' ' ' ' X • • 0 10
E. cheva lieri • • X • 4
E. isabelense X + x + o · + · + o x + x x + x x x x x o X + · + 2� r 3

E. kuhnii • + . 2
+ •
E . salicifolium x + + + · o + o + o o x x • + X X + + 3� 2
Lomariops;s guinecnsis 8
Loxogramme lanceolata · x x + · x x · · · + + + X + X + 0 • + X 27 + .

Lycopodium m i ldbraecli i 2
L. warneckei x · o x + · · o + · o o x o o o x x o x · -t o · o + 0 + 0 X 0 X + 33 + + 4
Microgramma owa r i e n s i s x · x x o o o o · · · o + · x x · x + • o + · · + u x x x t + 0 X 30 . + 4
Microsorium punctatum x x o x o x x x · x o o x x x x · o o x x + x x x x x x x x x x x · x 38 + + 5
• • X • • • • 0 • • • 0 X ·t· • • •
Nephrolepis bi serrata X 0 + X • • + X • 0 ' X + + + + • X · • X • + 0 + X + 24
N. undulata o x x · x + x · + + x o o x x x x x + x x x x x 'l: x x x x · x · · o · x x 0 X 0 X 38 r + 3
0\eandra dis Le n ta + x x o + x · x • x x x • o x + + + + + 0 X • + X X X X + • 34 + + 4
Phymatodes scolopendria x o x o · + · + x + X + ' X X X ' + X ' X X + 30
Plat.vceriom angolense 0 X 17
P . slemari::t • 0 X + 0 23

Pyrrosia mcchowii 3 2
• •
Tectaria angelicifolia X 6
T. fernandensis
Vi ttaria guineensis -t o x + x + x x x x + x x x x + + x + x x x x + x + X + X X X 42 + + + 5
Xit:.hopteris oosora 2
X . serrulata
X . villosiss ima
0 9 N ffi 0 00 0 � M O � � � M m N 9 9 0 N N � M� 9 - � ffi • M m - � • M � m 9 � ,......., ......, M ':"l :" I L::'
Total no . of pteridoph.vtc s pp. M 1"""1 � !"""' .-1 1"""1 1"""1 M ....-! C\1 J""-4
..-( ,..... ,....; 1'""""4 r-t .....-t C\l .--t C'? N ...-! � .....-t C'. J M C\1 .--t �
,....; C'\1 1"""1
....... M CO 1'"""4 .....-1 CO N CO 1"""1 ,....; M t-- C\1 .......
....... 7'1 ,.......; .....-1 - .....-I

Othe r vase . epiphytes


• • • • • • • . • • • • • . • • • •
0
Begonia mannii + + X 7
0 • • 0 + + X I + 0 • • • 0
B. polygonoides + 0 • + 0 () 0 • X . • 20
B . rubro-margina t..'l · o o + o x x · x 0 0 x x x o o x x · o + X 28
Cnlvoa monticola + .. . + . • • . . + . . • • • . . . 6
C. trochainii +· + • + + · + x + + + 0 · x x + · · x + x 22
Medinilla mannii X + • • + x x o · · + o o x • 0 + 0 0 • X + 24 + . 3
Peperomia fernandopoiana + X + • + · x + + • x + x · + X + + 23
P. molleri . + . 7
0 • l·
P . rotundifolia • X • · o + x · x + + 23
0
. + +
P reussie l la cheva l i e r [ + l t)
P . kamerunens i s + • 1
Remusatia v i v i para • 7
Rhipsalis baccifera + + o o + o x x · + o o + x x o o o x + 32 + . 4

Total no . of other 3 4 5 8 3 5 2 5 2 3 7 8 7 6 4 4 5 5 6 4 1 D 5 1 2 9 2 9 3 1 9 1 3 S 8 3 4 3 7 G 3 3 2 - 9 5 4 5 :2 3 - s 1
vas e . epiphyte spp.

x = close observalion, o = d i s L1.n<"e obse r v a t i o n , + o occasional observation. Only one k i n d o f record is g i v e n , i . e . when a c l ose observation record occu r , d i s La nce and
occasional observations a re excluded. If a distance observation is given no close record ex i s t and onlv when close and distance records a re lacking, occasional records

are show n .

A cta phytogeogr. suec. 59


;; � '1:l �
d' � Z t"' t"' Cl n r:

i
0 0 0

ll
0
c ::' i� "' "' c: 0 > >

1
"'

1 �! �-�;

�� �� �
i
"'
:r. ..;,.
""

I !
"

� �
:;;; =

i

i l
.:::: :r.
2 c

�� �
!i �
�1 l
X c

"'

� ·�
��
2-: T.'
� c
�� :r.. "'

t
¥. -:;,
;:
lf 2 -g � §
J � 1 3 �
-;:

· "
3 �
�· �g �
- -§
7.

·7.
�- �
a;· �

�- c


� ==

� � �7
high fo re s t
��
Phoroph\·tcs o f the

t\lbi7.i:."t glaberrim:1
Albizia 7..\·gia
so l e �

u ��
20
A l t n i : boon i
=. � A m phimas pleroc:tl'poidt':-;
Anthocleista nobi l i s
+ 0

A nthonnth:l f l':tgan s
. 0 A n L i a r i s toxic<l l'i;J
:: .- ')')
� iG

5·f 5·a ��
l3el' l i n i a confus:1
Bomba:x buonopozense
llusse:1 occidcntalis

%
C a l pocal.\'X aubre\·illci
Can;trium schwe i n l u r l h i i
42 CC'ib:l pentandra
C' h itl lowi a s a ngu i nc a
� � !� C h l o r·ophora regi:l
� g. 2 6
. C h r' vsoph_,· l lum perpulchrum
'"'
� � 29 C o m h rctodendron macroca rpum
:J � -!.) 'L
3
C o u l a cdulis
5 =. 24
g-
��
C r·,,·ptns(•p:llum '..C li'aph, · l l u m

i7 Uani e l l i n ogea

� � 26 39
Distcmonantbus benlh:rmianus
· E n t a nd l'Ophr·agma u t i l e
f':l 0

[ � �5
Et·_vthrophleum ivorcnse
F:t�·a r:r t c s s m a nn i i

a. � ��
Gua 1·ea ced rata
HeJ"iticra u t i l i s
Khav:1 a n t h o theca

i i �; � \'
Lophira :.1 la w

I
Lo oa trichi lioidcs
0 o lH
!\'Jammca africana
=.. 2. 47
• 0 /.

Mi t ra g.vna c i l i a t:l
c;;· 23 + . N:IUclca d id c rr i ch i i
[� � �� P:n inari :1uhre v i l l e i

:-·
Purin:lri excclsa
�� 13
Pa rina ri glabra
g 1 :.:!
('tl f':l
Parki;I bicolor
''0


. i4
Pen t.a c le thra m a c r u p h v l l : l
::: Pentatlesm;1 blltytacca
Q. g- 47 PipL;tdcn i : t s u·um a!'ric:rm1m
;;:· ::: 3
5� 20
P\Tnanthus angolcnsis
Rhodognaphalon b re v i cu s �
�5
g 0 Sacoglo l t i s gaboncnsis
., ;g !)
9
Tcrminalia i\·ot·e n s i s
� 0 Termin:tlia supe rba
� � 41
g. 3 26
Triplochiton sclct·ox\'lon
l':.�paca gu incen s i s
::,:1 t.l 2 :1

� g·
Vilex micrantha
- w !...: X t � _. 1..;. -1 � _. t' �� W ;: -1 ,_. '.ZJ � � I'<J ;:., .;.,, 1..;. � I'<J 1..:1 -1 =": ,_.

�!
,:': :.Jl � � ::': .;; ::: ,_. _. � � ,_. ,_. 0) � :j ; � �-=; � .l::o. � -; ::; :-: - ::;:: _. I � ::; t.;. � 1...;1 ':.1. t .;.,; �1 .l::o. �J. :: � ,_. :.a C': ;: ;:; � � � � � � _. :.;1 =t; � o;..;: .t:.. :,.,) ::::: ¥1 C: � ::-: 1 w 1.:.. rv <x :::t. 1..; l Total no. o f phorophv1c spp

Phoroph,Yles oulsitlc the h i g h fo1·e s 1 .


3!) Col:t n i l ida
�5
Cussonia bu rteri
10
Hevea b r a s i l i e n 3 i s
Musanga cecwpioidc!:i
n
:
Dead Lrces
I
El:1cis guineensis
1 ..... I - 1 ..: c..; ..::0 1
'l'ot a l no, of phoroph_vte spp
1 1 4 Dick Johansson

to the construction of a list of favourite 'host trees' from all over the world. Piers ( 1 968 :4) reports that
for certain popular orchids, for aiding the collector in Acacia and Ficus trees are avoided by most
field-work (Sulit 1 950 :6-8). Morris ( 1 970 :5) uses the epiphytic orchids. The same applies to tree Euphor­
term 'orchid-prone' for trees that are attractive to bias, exotic conifers and Eucalyptys trees. Moreau
epiphytic orchids. The following species of ( 1 943 : 8), however, remarks that conifers such as
phorophytes have been reported to carry an abun­ Juniperus bermudiana are colonized by orchids with
dance of epiphytes : a surprising frequency.
From Australia Rupp ( 1 969 :XI) confirms that
Liberia: Cola nitida (Harley 1 95 5 :7 1 ), Mitragyna ciliata
(Voorhoeve 1 965 :324), Parinari excelsa (Voorhoeve Eucalyptys trees are usually barren of orchids. The
1 965 :3 1 9) absence of epiphytes on figs has been observed by
West Africa: Parinari excelsa (Jaeger et al. 1 968 :270, several authors. Schnell ( 1 9 70 :309) mentions that
Schnell 1 970 :309) Ficus mucuso lacks epiphytes and Went ( 1 940) notes
Malawi : Brachystegia spiciformis, Bridelia micranta, /lex that humus epiphytes are lacking on Ficus in­
mitis, Parinari mobola, Podocarpus milanjuanus, volucrata Bl.
Syzygium cordatum, Uapaca kiekiana (Morris 1 9 70 :5)
Java: Castanea argentea, C. javanica (Went 1 940)
C. Spe�ific relationships between phorophyte and
West Borneo : Vitex sp. (Schuitem�ker, in Went 1 940 :95) epiphyte
Florida, U.S.A. : A nnona glabra L., Taxodium distichum In the Nimba area, no · specific relationship between
L. (Frei 1 9 73b : 70 1 )
phorophyte and epiphyte has been observed, even if
Haiti : Eugenia jambos L . (Curtis 1 946), .Euphorbia
the total dominance of Parinari excelsa at the
lactera Haw. (C urtis 1 94 7)
highest parts of the mountains has resulted in a near­
S. America: Crescentia cujete (Schimper 1 888 :95)
ly absolute correlation between this tree and some
B. Phorophytes with few epiphytes epiphytes naturally occurring at these altitudes (Ta­
In the Nimba area epiphytes are poorly represented
ble 47).
on a number of species of phorophytes (Table 4 7), in
There are several reports of a relationship between
particular on Fagara tessmannii, Terminalia ivoren­
a particular species of phorophyte and a particular
sis, T. superba and Musanga cecropioides, con­
species of epiphyte. Piers ( 1 968 :4) states: "Some
sidering the abundance of full-grown individuals of
epiphytes have marked idiosyncrasies regarding their
these phorophytes. Less frequent phorophytes with a
host tree, e.g. A nsellia nilotica has a strong
sparse epiphytic flora are e.g. A lbizia glaberrima, A .
preference for the Doum Palm (Hyphaene thebaica),
zygia.
the Baobab tree (A dansonia digitata) is the typical
The absence of epiphytic orchids on the trunks of host of A ngraecum dives and Polystachya adan­
Elaeis guineensis is obvious. However, they support soniae, while Aiirangis thomsonii attaches itself
a rich pteridophyte flora (Fig. 7 1). Only two almost exclusively to the rugged stems of the gigant
epiphytic orchid species have been observed on this ceder (Juniperus procera) of the Kenya highlands."
palm : Graphorchis lurida and Habenaria procera. Rupp ( 1 969 :XI) gives a similar observation :"one or
(The Kew Herbarium. has 26 collections of the latter two of the ironbark eucalypts are favoured by Den­
species. The growth habitat is given for fifteen of drobium aemulum, which for this reason is often
them. Eighty per cent of these were epiphytes, all of callen 'Ironbark Orchid' ". More examples of this
them on Elaeis guineensis.) kind are listed by Richards ( 1 964 : 1 1 9). Morris
In a study of the epiphytes on Elaeis guineensis in ( 1 9 70 :5) found few specific host/orchid relationship
the former Belgian Congo (Zaire), Van Oye ( 1 924) in Malawi, but observed : "There are however a few
did not mention any orchids, but several orchids which would appear to associate with
pteridophytes and figs. In the Philippines it is even specific hosts . . . Bulbophyllum intertextum on New­
stated that "Orchids will choose any other tree in the tonia buchananii, Polystachya johnstonii on Vellozia
wild than the palm" (Sulit 1 950 :8). spendens come to mind."
Ruinen ( 1 9 5 3 : 1 02) has suggested the term 'axeny' Moreau ( 1 943 :8) is a little more hesitant : "there is
for this condition of general inhospitality of trees little evidence of specific relation between orchid and
which never or only occasionally bear epiphytes. tree host, except that certain aberrant Polystachyas
Trees belonging to this group have been reported that are otherwise terrestrial also grow on the curios-

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 1 1 5

l y looking fibrous stem o f Vellozias, but no other leaves, seem to be of minor importance for the rain
host." (See also Richards 1 95 7 :5 6 8.) forest species in general (Odum 1 970 b). Voute
B oyer ( 1 964) states that the occurrence of the ( 1 946, 1 9 64) states that insect outbreaks do not oc­
epiphytic ferns Platycerium stemaria and P. cur in tropical forests because of the stability of
angloense on various phorophytes in the Ivory checks and buffers that act on any species that
Coast, does not permit any conclusions regarding becomes overabundant.
their preference for particular species of
phorophytes. Nutritional importance of animals and micro-
Whether a relation between a certain species of organisms
phorophyte and epiphyte really exist is doubtful. In The presence of a rich fauna in the crown of the rain
many cases the epiphytes have for various reasons a forest trees naturally influences the nutritional supp­
very limited choice of host trees. Given the oppor­ ly of the epiphytes through excreta from m ammals,
tunity, many of these epiphytes now occurring on a birds, snakes, lizards, frogs and insects. The effect of
specific host tree, might very well thrive on other bird excreta was early recognized. Sernander ( 1 9 1 2)
trees. Another fact that favours the theory of a coined the term 'ornithocoprophily' for such plants
specific relation between host and epiphyte is the that were favoured by bird excrement (or rather their
poor knowledge of the presence of the epiphytes. In­ nitrogen content). Du Rietz ( 1 93 2) showed that bird
creasing information on the ecology and presence of excrements also increased the pH of the substrate.
the epiphytes will probably decrease the number of Trees with nesting birds, edible berries, as well as
'specific relationships'. those used as sleeping trees and observation posts,
are known to attract nitrophytic lichens. It is im­
possible to even try to estimate the total effect of
Biotic factors nutrition provided by the animals in the diversified
rain forest. It should at least be kept in mind during
Several of the biotic factors have already been the analysing process.
presented and discussed in Chapter Ill. The role of the microorganisms in the nutritional
balance is unknown, but Ruinen ( 1 95 6) found
Human influence bacteria of the genus Beijerincka on leaves on nearly
The human influence on the rain forest effects the all trees and shrubs tested in Java and Sumatra.
epiphytes indirectly. In the Nimba area patches of These nitrogen-fixing bacteria obtain their minerals
scattered trees that remain after forest destructions and some organic material from substances excreted
are the particular growing sites of several orchid by the leaves. Blue -green algae of several kinds are
species. These are species of the deciduous forest, always abundant on the bark of trees in the wet
e.g. A nsellia africana, Plectrelminthus caudatus, tropics but their nitrogen-fixing capacity is unknown.
Polystachya paniculata, P. puberula and Rangaeris (cf. Ruinen 1 96 1 ).
rhipsalisocia. Bond ( 1 959) reports : "There are also nitrogen­
fixing micro-organisms present on leaves and
Animal influence branches of trees, which doubtless adds to the
Direct effects, as judged by signs of being eaten or nitrogen available to the roots of epiphytes, as
used by animals, on the epiphytes is . very seldom similar microorganisms do in the soil."
observed. Some species, however, show signs of fre­ The high amounts of nitrogen in the humus from
quently being eaten (probably by insects), e.g. Polypodium polypodioides puzzled Pes sin ( 1 925 : 3 2) :
Pyrrosia mechowii, A ncistrochilus rothschildianus, "It is indeed remarkable that the nitrogen o f the
A ngraecopsis elliptica and Polystachya ramulosa. humus on the bark of the tree is frequently as high
Effects on the leaves by herbivores or other and at times even higher than that of the most fertile
agents, as can be judged by holes in the falling soils."

A cta phytogeogr. suec. 59


VI. Distr ibution on the phorophytes

General pattern · 1 1 6), Tectaria angelicifolia and T. fernandensis are


At the same time as it is possible to recognize a dis­ frequent. The orchids are represented by
tribution pattern on the phorophyte according to the Polystachya ramulosa and P. rhodoptera which are
occurrence of a particular species of epiphyte, one almost exclusively located in this section.
can al so distinguish a pattern for whole groups of Also Brachycorythis kalbreyeri, Habenaria
epiphytes (Table 48). procera, Liparis nervosa and Vanilla crenulata
Although lichens, climbers, filmy ferns and belong to this section, being epiphytes which grow in
mosses were not consistently recorded, their large· humus deposits. The group of other vascular
presence was noted. Generally speaking, each of the epiphytes is represented by Calvoa monticola and
five zones of the phorophyte are characterized by a more seldom C. trochainii.
particular group or groups of epiphytes (Fig. 1 1 5).
Species normally located on the main trunk
Specific patterns Several species may appear in this section, e.g. A n­
Species seldom observed on tall phorophytes cistrorhynchus cephalotes, A splenium africanum,
Certain species of orchids are very rarely observed Diaphananthe bidens, D. pellucida. However, there
on tal l phorophytes, instead they frequently occur on are rather few species that are restricted to this ver­
smaller trees and even shrubs that often border small tical section, which is mostly devoid of humus
rivers and streams, or are found along tracks in the
forest. A ngraecum classensii, Diaphananthe den­
siflora, D. rutila, Eurychone rothschildiana and
R hipidoglossum paucifolium exhibit this pattern.

Species normally restricted to the basal part of the


trunk Li c h e n s

This section of the tall phorophytes is occupied by a


num ber of species very rarely found in any other sec­
pte r i d o p h yt e s +
tion. Among the pteridophytes A splenium barteri is
oth e r vasc u l a r e p .
the most common, but several others e.g.
Elaphoglossum kuhnii, Lomariopsis guineensis (Fig.

Table 48. General distribution pattern (section) on the phorop­


hyte of the pteridophytes, the orchids and the other vascular
epiphytes. Based on the most frequent growing site (section) of C l i mbe rs

each spe� ies (Tables 27 a + b, 28 a + b, 29). The result is pre­


sented as a percentage of the total number of species in each
group.

V F i l my ferns

4.0
Fig. 1 1 5. General di stribution of epiph ytes on tall
phoroph ytes.
Ecology of vascular epiphytes in West African rain forest 1 1 7

Fig. 1 1 6. Lomariopsis guineensis normally occurs on the


lowest parts of the trunks. Nimba Range 700 m.
deposits. Two species of ferns and two orchids can
be considered to have the center of their distribution
in this section : A ntrophyum immersum,
Elaphoglossum chevalieri, Polystachya saccata and
Tridactyle crassifolia. Two species with a climbing
habit, Culcasia angolense and Raphidophora
africana (Fig. 1 1 7), which are included among the
facultative epiphytes, play an important role through
their substrate holding capacity.
Species of the crowns
In this analysis the results from the close obser­
vations for species with at least 20 separate records
from the crowns have been used. For species with
fewer observations the results from the distance
observations are added and then used, if the total Fig. 1 1 7. R aphidophora africana and Culcasia angolense
sum is at least 20 observations. It might seem hazar­ are common climbers on the trunks of tall trees. Y ekepa.
dous to use such a low number of observations as a
base for a detailed distribution pattern. However, in fourth and fifth pattern, D and E, are formed by
most cases the information obtained with these species that occur in almost equal numbers in two
rather accurate methods are checked against the sections with very few individuals in the third, (D, for
more numerous but ecologically not as well­ species in sections Ill + IV and E for species in sec­
documented, so-called occasional observations. tions IV + V). The last pattern, F, is formed by
When there are uncertainties in the distribution species that occur more or less evenly in the various
pattern, e.g. differences in the results betweep the sections of the large branches (Ill + IV + V).
various observation methods used, the discrepancies Species exhibiting the various distribution pat­
are discussed in the text. terns are:
Since the absolute majority of the records of A (Ill)
A ncistrorhynchus capitatus
Asplenium africanum
species belonging to this group are located in the A ngraecum chevalieri
A. dregeanum
crown, the records from other sections of the A . geppii
Bulbophyllum bufo

phorophyte have simply been omitted in this A rthropteris monocarpa B. falcatum


B. josephii
analysis. A. orientalis
B. saltatorium
Elaphoglossum salicifo!ium
The distribution pattern of the individual species B. winkleri
Microgramma owariensis
can be divided into six major patterns (A-F). The Calyptrochilum christyanum
Nephrolepis biserrata
first three, A, B and C are represented by species N. undulata
Diaphananthe pellucida

that are ± restricted to only one section of the Polystachya laxiflora


Vittaria guineensis

branches, (A, for species in section Ill, B, for species Xiphopteris oosora Begonia mannii

in section IV and C, for species in section V). The A ncistrochilus rothschildianus B. polygonoides

A eta phytogeogr. suec. 59


1 1 8 Dick Johansson

B. (IV) tii and Tridactyle tridactylites seem to follow this pattern,


A splenium megalura Cyrtorchis aschersonii but it should be noted that these two species deviate from
Elaphoglossum isabelense Diaphananthe bidens this distribution in the distance observations where A .
Lycopodium mildbraedii Graphorchis lurida laurentii shows a n E - an d T. tridactylites and A-pattern.
L. warneckei Nephrangis filiformis
Platycerium angolense Polystachya galeata
P. stemaria P. paniculata Species distribution and environment gradients
P. puberula
A ncistrorhynchus clandestinus along large branches
A ngraecum distichum Rangae'ris muscicola
R. rhipsalisocia
In the absence of data about the most important en­
Bulbophyllum barbigerum
Tridactyle anthomaniaca vironmental factors concerning the changes along
B. buntingii
B. intertextum
T. crassifolia the branches a theoretical discussion may throw
B. maximum
T. tridentata some light on this problem.
B. oreonastes Medinilla mannii The environmental changes that occur are listed
B. schinzianum Peperomia rotundifolia here in two main groups, i.e. factors that decrease
B. schimperanum R hipsa/is baccifera
respectively increase towards the outer parts of the
Chamaeangis vesicata
branches. In the first group the roughness of the sub­
C(V) Only one species exhibits this pattern, Bolusiella strate, humus deposits, nutritional content of the
talbotii (Fig. 1 1 8). substrate, humidity of the substrate and the relative
D (Ill + IV) humidity of the air can be mentioned. In the second
group the temperature, light intensity and wind
Dl'ynaria laurentii B. linderi
Microsorium punctatum
velocity are important.
B. lucifugum
Phymatodes scolopendria Cyrtorchis arcuata In Fig. 1 1 9, the decreasing and increasing factors
A erangis biloba
subsp. variabilis are each represented by a straight line. The steepness
A ngraecopsis elliptica
Polystachya leonensis and irregularities of these gradients are naturally
P. polychaete
A ngraecum birrimense dependent on a number of variable factors, e.g.
P. tessallata
Bulbophyllum cochleatum species of phorophyte and the physiognomy of the
B. cocoinum B. rubro-marginata
forest.
B. distans
The distribution pattern of six species of
Bulbophyllum linderi exhibit an A-pattern in the distance epiphytes, each representing one of the major
examinations. A erangis biloba shows a C-pattern in the
patterns that previously were described, are also il­
distance observations and in the occasional records it
appears in all sections of the phorophytes. lustrated with the decreasing and increasing factors
shown as dotted lines.
E (IV + V) Only orchid species exhibit this distribution
A. This pattern follows the decreasing gradient.
pattern : Bulbophyllum scariosum, Polystachya adan­
soniae, P. dalzielii. The proportionally large number of pteridophytes,
but also some orchids that belong to this group, seem
F (Ill + IV + V) Species representing this pattern are
to be primarily limited in their distribution by the
very rare, a fact that suggest that each of the epiphytic
species is confined to restricted habitats. A erangis lauren- decreasing factors.
B. This pattern seems to arise from the interaction
of the increasing and decreasing gradients. In the
basal parts of the branch the increasing gradient, and
in the outer parts the decreasing gradient limit the
presence of epiphytes. A large number of orchids ex­
hibit this pattern which probably can simply be
described as a delicate balance between the need of
light and the influence of evapotranspiration.
C. This pattern follows the increasing gradient. It
is probably the higher light intensities that favour the
only species that exemplifies this pattern.
Fig. 1 1 8. Bolusiella talbotii normally occurs on the outer D. This is a modified A-pattern where the in­
parts of large branches among lichens. A small R hipsalis fluence of the increasing gradient effects the pattern
baccifera has also established itself. in the basal parts of the branch.
A cta phytogeogr. suec. 59
Ecology of vascular epiphytes in West African rain forest 1 19

1 0 0 '1. E. This is a modified C-pattern. The effect of the


decreasing gradient is noticed at the outer part of the
50
branch. Only three orchids show this pattern. The
evapotranspiration in these more exposed parts of
branches represents a severe drawback for the sur­
B a sa l p a rt M i d d l e p a rt O uter pa rt
vival of epiphytes. Therefore it is interesting to note
that two of the orchid species, Bulbophyllum
scariosum and Polystachya dalzielii, only occur in
high altitudes where low temperatures and mist sur­
A press the evapotranspiration. The third species,
Polystachya adansoniae, is adapted to dry en­
vironments with a geogr�phical distribution in rather
dry areas over large parts of Africa.
F. This pattern is the average of the increasing and
B u l bo p hy l l u m o reo n astes __
_ decreasing gradients, suggesting a wide tolerance
towards environmental influence. However,
B A erangis laurentii and Tridactyle tridactylites, both
show a different pattern in the records from the dis­
tance observations which may indicate that this
pattern is less distinct than the others.

B o l u si e l l a ta l botii The importance of any single factor in the total en­


vironment is naturally very hard to estimate. In a
c detailed study of the epiphytic fern Polypodium
polypodioides the following statement was presented :
"It is safe to conclude from the results obtained that
neither the light, the relative humidity of the air, the
temperature of the air nor the substratum
B u l bophy l l u m l i n d e ri
temperature exert any direct influence on the oc­
currence and distribution of Polypodium
D polypodioides nor on the epiphytes associated with it.
It is the sum of the total of the effects of the first
three conditions on the evaporating power of the air,
in the immediate vicinity of each station on the tree,
which is responsible for the particular epiphytic flora
Polyst a c hya a d a n s o n i a e
at the given station" (Pessin 1 925 : 3 1 ).
The evaporating power normally increases from
E
the base of the large branches towards the outer part
through influences previously described.
Most pteridophytes in the area investigated show
a distribution pattern on the branches that is
reversed compare? to a possible increase in
Aera n g i s l a u re n ti i
evaporating power.
.............. .... ..........�•• !�-:.: ............... ...... For most of the orchids the distribution pattern
F ---
_....
.. -.
....... .

seems to be a result of a need for high light intensities


and the tolerance of the evaporating power of the air,
while the composition of the substrate can be con­
sidered of secondary importance.
Fig 1 1 9. A theoretical model presenting a possible ex­
planation for the different distribution patterns of To make the distribution solely dependent on the
epiphytes on the large branches (see text). evaporating power of the air without any considera-

A cta phytogeogr. suec. 59


120 Dick Johansson

tion of light intensity and substrate composition is with differences in the evaporating power of the air.
not accurate for the majority of epiphytes. This in­ It is more probable that low light intensities are the
vestigation showed that the different epiphytes often limiting factor since substrates of similar composi­
have a marked preference for a certain kind of sub­ tion and evaporating power of the air equal to the
strate and light intensity. The almost total absence of conditions in the crowns also exist at various places
epiphytes from the crowns of the trees in the lower in the lower parts of the forest.
strata of the forest cannot be satisfactorily explained

A cta phytogeogr. suec. 59


Resume

Ecologie des epiphytes vasculaires dans la foret dense humide d'Afrique occidentale

La superficie examinee communautes de montagne par exemple les


Cette etude est centree sur l'ecologie des epiphytes Polystachya dalzielii et P. leonensis (Fig. 2 1 ).
vasculaires dans la partie liberienne des Monts Nim­ Biologie de quelques epiphytes
ba et superficies adjacentes (Fig. 2). La flore Les orchidees peuvent fleurir chaque mois de l'annee
d'ensemble de cette meme region a ete decrite par mais on remarque deux pointes de floraison (Fig.
Schnell ( I 952) et Adam ( 1 97 1 ). 49). Le genre Bulbophyllum a une pointe de floraison
Dans l'aire examinee, la haute foret (dont la strate a la fin de la saison pluvieuse tandis que le genre
arborescente superieure est a 30 m ou plus au-dessus Polystachya presente une pointe correspondante au
du sol) occupe de petites surfaces. Ce sont les forets debut de la saison seche ou de la saison pluvieuse.
a differents etages de regeneration et les surfaces
Parmi les insectes frequentant les fleurs d'orchidees
agraires qui dominent le paysage. La partie la plus on a pu observer trois especes avec pollinies
haute de la region est couverte d'une foret vierge adherente : Polistes marginalis (Tab.) et Polybiodes
Parinari excelsa. tabida (Tab.) de Bublophyllum schinzianum et
Euchromia lethe (F.) (Fig. 5 2) de Diaphananthe
Composition de la flare epiphytique pellucida.
L'etude porte sur 1 5 3 especes au total (39
pteridophytes, I 0 I orchidees et 1 3 autres epiphytes Formes de vie et hydrometrie
vasculaires). Des enregistrements additionnels ont La resistance des epiphytes a la secheresse se
considerablement augmente la zone de distribution reflechit dans differentes adaptations xeromorphi­
de quelques especes (Fig. 22-24) et 32 especes sont ques (Fig. 5 9). Les epiphytes peuvent etre divisees en
nouvelles au Liberia : plantes tolerantes a la secheresse et plantes evitant la
B. saltatorium secheresse. C e dernier groupe est decidu et evite ainsi
Elaplwglossum isabelense
E. salicifolium B. schimperanum que l'eau ne se perde par la transpiration des feuilles
Lycopodium mildbraedii Cyrtorchis aschersonii durant la saison seche tres rigoureuse. Les feuilles
Pyrrosia mechowii Diaphananthe densiflora des plantes tolerant la secheresse persistent pendant
Xiphopteris oosora Eurychone rotschildiana la saison seche. Oes.valeurs tres elevees de transpira­
Podangis dactyloceras
Ancistrochilus recurvus
tion ont ete obtenues pour un certain nombre de
Polystachya saccata
A ngraecopsis elliptica P. subulata plantes evitant la secheresse par ex. Polystachya
A ngraecum classensii P. tenuissima leonensis et Polystachya dalzielii, provenant des
Bulbophyllum bifarium Rangae'ris brachyceras parties les plus hautes et les plus humides de la zone,
B. josephii R hipidoglossum paucifolium et Graphorchis lurida et Polystachya puberula
B. lucifugum Stolzia repens
B. lupulinum
d'altitudes plus basses. Des valeurs elevees ont ete
Tridactyle crassifolia
B. magnibracteatum T. tridactylites ainsi obtenues pour A splenium geppii et Begonia
B. pavimentatum T. tridentata rubro-marginata (Tableau 1 2).
B. phaeopogon C'est parmi les orchidees que l'on a obtenu les
B. rhizophorae Begonia rubro-marginata valeurs les plus faibles de transpiration, spe_c ialement
parmi celles qui ont une croissance monopidiale par
La tlore epiphytique comprend des elements de la
ex. Bolusiella talbotii, Calyptrochilum christyanum,
foret dense humide de basse et moyenne altitudes par
Chamaeangis vesicata et Neprhangis filiformis
exemple les Cyrtorchis monteiroae et Angraecum
(Tableau I 3).
birrimense (Fig. 1 9), de la foret decidue dans la
mosaique foret-savane par exemple les A nsellia Repartition des epiphytes sur les arbres
africana et Rangae'ris rhipsalisocia (Fig. 20) et des Methodes. Sur 463 phorophytes (arbres hotes) la

A cta phytogeogr. suec. 59


122 Dick Johansson

repartition des epiphytes a ete enregistree. Pour Colonisation et effets sur les phorophytes
simplifier l'enregistrement le phorophyte a ete divise Les differentes methodes indiquent que ce sont par­
en cinq sections (Fig. 76). On a applique trois ticulierement les phorophytes les plus grands qui
methodes differentes de denombrement : les obser­ sont de preference colonises par les epiphytes. Les
vations proches, les observations a distance et les pteridophytes semblent etre les epiphytes pionniers
observations occasionnelles. L'observation proche tandis que les orchidees sont les dernieres a s'etablir
implique des investigations sur des arbres abattus, (Fig. 8 7). Certains arbres ay ant une riche flore
!'observation a distance est une etude du tronc et epiphytique par exemple Mitragyna ciliata et
d'une branche complete (ou plus rarement un arbre Parinari excelsa, souffrent apparemment
entier) d'un arbre sur pied a l'aide de jumelles ou "d'epiphytosis" (Ruinen 1 95 3). Les effets des
d'un telescope sur trepied. orchidees sans feuille sur les phorophytes peuvent
L'observation occasionnelle est reduite aux in­ aussi indiquer un parasitisme partiel.
vestigations obtenues a partir seulement d'une partie
limitee du phorophyte. Les resultats de cette derniere Influences de l'environnement sur les epiphytes
methode ne sont utilises que pour completer ceux des Climat. Le rapport pteridophytes epiphyti­
deux autres. ques/orchidees se modifie avec !'altitude. A la base
On a egalement examine le substrat et l'eclaire­ de la zone (500-700 m) la proportion
ment au lieu de croissance de !'epiphyte individuelle. pteridophytes/orchidees est approximativement de
Le substrat a ete divise en trois classes : ecorce, 1 :3 mais, a 1 000- 1 300 m, elle est de 1 : 1 . Des
ecorce avec depots faibles d'humus, et depots impor­ variations similaires ont egalement ete observees sur
tants d'humus. La lumiere a aussi ete divisee en trois les hautes montagnes d' Afrique orientale.
classes : plein soleil, demi-obscurite et obscurite. Dans les milieux secs, ainsi qu'on l'a observe en
Tanzanie, le nombre des epiphytes decroit rapide­
ment (specialement les pteridophytes) mais l'abon­
Occurence des epiphytes dance de certaines orchidees epiphytiques peut sur­
Dans la haute foret 50,4 % des arbres hauts de dix prendre. D ans les montagnes meridionales de Tan­
metres ou plus portaient des epiphytes, par contre zanie on a observe que tous les arbres (� 4 m) cVune
dans une foret "secondaire" on descendait a 1 4,8 %. terre boisee Brachystegia portaient l'orohidee I
Il etait done evident que particulierement les Tridactyle tricuspis.
orchidees etaient rares dans une foret "secondaire".
Certaines especes d'arbres peuvent abriter un grand Substrat. La valeur pH d'echantillons d'ecorce
nombre d'especes d'epiphytes par ex. Heritiera uti/is provenant du meme phorophyte presentait d'assez
(30 spp.) Mitragyna ciliata (28 spp.) (Tableau 42). grandes variations. C 'est dans les parties exterieures
Aussi le nombre des individus ("stands") d'epiphytes des grandes branches et la partie superieure du tronc
sur certains arbres etait-il tres grand, par exemple un principal que l'on trouvait les valeurs les plus basses.
seul Heritiera uti/is ne portait pas moins de 1 85 7 On n'a cependant pas pu trouvef de corrClation per­
epiphytes et un Parinari excelsa 1 1 7 1 . tinente entre la valeur pH de l'ecorce et la flore
epiphytique. Les valeurs pH des depots d'humus
etaient beaucoup plus elevees que celles de l'ecorce.
Groupements epiphytiques La majorite de l'espece pteridophyte "preferait"
On dit qu'il y a groupement epiphytique quand trois les depots mineurs d'humus tandis que les orchidees
especes au minimum d'epiphytes forment une unite poussaient surtout sur l'ecorce et les autres epiphytes
et quand la distance entre deux de ces epiphytes et la vasculaires dans les grands depots d'humus.
troisieme ne depasse pas 0,5 m. On distingue sept
groupements epiphytiques differents au-dessous de Correlation entre l'abondance d'epiphytes et l'espece
1 000 m et trois autres de plus a des altitudes de phorophyte
superieures (Tableau 30). Les grands specimens d' Heritiera uti/is, Lophira
Les fougeres forment la base de sept des dix alata, Mitragyna ciliata et Parinari excelsa sont
groupements. Il semble que ce soit leur capacite de riches en epiphytes tandis que par exemple
formation de substrat qui attire les autres epiphytes. A nthocleista nobilis Fagara tessmannii, Terminalia

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 123

ivorensis et T. superba sont frequemment plus ou La distribution des epiphytes dans le phorophyte
moins depourvus d'epiphytes. On n'a pas observe de Chacune des cinq sections du phorophyte est
correlation specifique entre }'epiphyte et le dominee par un groupe d'epiphytes (Fig. 1 1 5).
phorophyte. La plupart des especes d'epiphytes ont une dis­
tribution plutot limitee sur le phorophyte, c'est-a-dire
Lumiere. L'intensite de la lumiere au sol dans une dans une ou deux sections. La majorite des especes
haute foret a la base de la zone etait de 0,36 % de la poussent dans la couronne. On distingue six modeles
lumiere exterieure mais dans une foret Parinari a differents de distribution sur les grandes branches
1 300 m d'altitude elle etait de 6,65 % (Tableau 42). (Fig. 1 1 9).
Dans les parties de base et les parties centrales des La distribution des epiphytes ne depend pas seule­
grandes branches d'un arbre Parinari excelsa a 1 300 ment de !'evaporation (Pessin 1 925) mais elle est
m la lumiere etait reduite de 70-85 % par rapport a egalement influencee par la preference marquee
celle de la demi-obscurite (Tableau 43). La variation d'une certaine sorte de substrat et d'eclairement. Les
quotidienne dans l'intensite de la lumiere au lieu de pteridophytes presentent une distribution qui va a
croissance de certains epiphytes a revele des l'encontre d'un accroissement possible de !'evapora­
differences tres significatives entre les differentes es­ tion. La distribution des orchidees semble toutefois
peces (Fig. I 06- 1 08). En general, on trouve les etre le resultat d'un equilibre entre le desir d'une
pteridophytes dans la demi-obscurite tandis que les grande intensite de lumiere et la capacite de resister a
orchidees sont plus egalement reparties entre une plus grande evaporation d'air qui s'ensuivrait.
!'habitat en demi-obscurite et )'habitat en plein soleil.

A cta phytogeogr. suec. 59


Literature

Adam, 1.G. 1 96 6 - La vegetation du Mont Nimba au Bernhard-Reversat, F., Huttel, C., & Lemee, G. 1 9 72 -
Liberia et sa protection. - Notes africaines, I. F.A.N. Some aspects of the seasonal ecologic periodicity and
1 1 2. plant activity in an evergreen rain forest in the Ivory
1 969 - Etude comparee de quelques forc�ts ouest­ Coast. - In Golley, P.M. & Golley, F.B. (ed.)
africaines. - Bull. I.F.A.N. 3 1 : 3 4 1 -409. 1 97 2 :2 1 7-2 1 9.
1 9 70 - Etat actuel de la vegetation des monts Nimba Blum, G. 1 93 3 - Osmotische Untersuchungen in Java. I.
au Liberia et en Guinee. - Adansonia 1 0 : 1 9 3 -2 1 1 . - Ber. Schweiz. Bot. Ges. 42 : 5 5 0-680.
1 9 7 1 - Flore descriptive des Monts Nimba. - Mem. Bond, G. 1 9 5 9 - The incidence and importance of
M us. N at. Hist. N at. B, 20, 22. biological fixation of nitrogen. - Adv. of Sci. 1 5 :3 82-
Adamson, A.M. 1 943 - Termites and the fertility of soil. 3 86.
- Trop. Agric. Trin. 20(6) : 107- 1 1 2. Boyer, Y. 1 964 - Contribution a la etude de
Ake Assi, L. 1 963 - Contribution a la etude floristique de l'ecophysiologie de deux fougeres epiphytiques :
la Cote d'Ivoire et des territoires limitrophes. 2, Platycerium stemaria et Platycerium angolense. -
Monocotyledones et Pteridophytes. - Lechevalier, Ann. Sci. Nat. 1 2(5): 8 7-228. Paris.
Paris. Bovey, R.W., & Diaz-Colon, J.D. 1 969 - Occurrence of
Allee, W.C. 1 926 - Measurement of environmental fac­ plant growth inhibitors in tropical and subtropical
tors in the tropical rain forest of Panama. - Ecology vegetation. - Physiol. Plant. 22:253-259.
7 :273-302. Burgeff, H. 1 936. - Die Samenkeimung der Orchideen. ­
Alien, S.E., C arlisle, A., White, E.1., & Evans, C.C. 1 968 Fisher, Jena.
- The plant nutrient content of rainwater. - 1. Ecol. - 1 95 9 - Mycorrhiza of orchids. - In Withner, C .L.
5 6 :497-504. (ed.) 1 95 9 : 36 1 -3 9 5 .
Alston, A.H.G. 1 95 9 - The Ferns and Fern-Allies of Cachan, P. 1 963 - Signification ecologique des variations
West Tropical Africa. - Crown agents, London. microclimatiques verticales dans la foret sempervirante
Ames, 0. 1 922 - Observations on the capacity of orchids de basse Cote d'Ivoire. - Annales Fac. Sci. Univ.
to survive in the struggle for existence. - Orch. Rev. Dakar 8 :89- 1 5 5.
30:229-234. C achan, P. & Duval , J. 1 963 - Variations microclimati­
Anonymus. 1 968 - How old is an orchid plant? - Amer. ques verticales et saisonnier dans la forc�t semper­
Orch. Soc. Bull. 3 7 :405 . virante de basse Cote d'Ivoire. - Ibid. 5 -98.
Arditti, 1. 1 966 - Flower induction in orchids. - Orch. C arter, G.S. 1 934 - Reports of the C ambridge expedition
Rev. 74 :208-2 1 7. to British Guiana, 1 9 33. Illumination in the rain forest
- 1 96 7 - Factors affecting the germination of orchid at ground level. - J. Linn. Soc. (Zool.) 3 8 : 5 79-589.
seeds. - Bot. Rev. 33 : 1 -97. Childers, N.F., Cibes, H.R., & Hernandez-Medina, E.
Arditti, 1. & Dueker, J. 1 968 - Photosynthesis by various 19 5 9 - V anilla - The orchid of commerce. - In
organs of orchid plants. - A mer. Orch. Soc. Bull. 3 7 : Withner, C .L. (ed.) 1 95 9 :447-508.
862-866. Coe, M . & Curry-Lindahl, K . 1 96 5 - Ecology of a
Arditti, 1., Healy, P., &· Ernst, R. 1 972 - The role of Mountain : First report on Liberian Nimba. - Oryx
mycorrhiza in nutrient uptake of orchids 11. - Ibid. (1ourn. of the fauna preserv. soc.) 8(3).
4 1 :503-5 1 0. Cook-Melville, T. 1 926 - Epiphytic orchids as an
Ashton, P.S. 1 969 - Speciation among tropical forest epiphytic pest on citrus trees. - 1ourn. Dept. Agr.
trees. - Bioi. J. Linn. Soc. 1 : 1 5 5 - 1 96. Puerto Rico 1 0 : 5 -9.
Atlas of Tanzania, 1 967 - Ministry of lands, settlement & Cole, N.H.A. 1 968 - The vegetation of Sierra Leone. -
water development. Dar es Salaam. Njala, Sierra Leone.
Aubreville, A. 1 93 8 - La foret coloniale; Les forets de Coombe, D.E., & Hadfield, W. 1 962 - An analysis of the
l'Afrique occidentale fran�aise. - Ann. A cad. Sci. growth of Musanga cecropioides. - 1. Ecol. 50 :22 1 -
colon. 9 : 1 -245. Paris. 234.
Bainbridge, R., Evans, G.C., & Rackham, 0. (ed.) 1 968 - Cooper, G . P . & Record, S.J. 1 93 1 - The evergreen
Light as an ecological factor. - The British ecological forests of Liberia. - Yale Univ. Forestry Bull. 3 1 .
society. Blackwell Scientific Publications, Oxford and Coutinho, L.M. 1 964 - Untersuchungen tiber die Lage
Edinburgh. des Lichtskompensationspunkts einiger Pflanzen zu
Barkman, 1.1. 1 95 8 - Phytosociology and ecology of verschidenen Tageszeiten mit besonderer
cryptogamic epiphytes. - Van Gorcum & Co, Assen. Berticksichtigung des 'de-Saussure-Effektes' bei

A cia phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 1 25

Sukkulenten. - Beitr. zur Phytologie (Walter­ Focke, W.O. 1 893 - tfber epiphytische Gewiichse. Nach
Festschr.) 1 0 1 - 1 08 . Briefen von Fritz Muller. - Abh. naturwiss. Ver.
Curtis, C . H. 1 93 3 - How orchid seeds are dispersed. - Bremen 1 2 :5 62.
Orch. Rev. 4 1 : 1 1 2- 1 1 6. Foggie, A. 1 947 - Some ecological observations on a
- 1 943 - Pseudo-bulbs. - Ibid. 5 1 : 1 3 7. tropical forest type on Gold Coast. - J. Ecol. 3 4 :88.
Curtis, J.T. 1 93 9 - The relation of specificity of orchid Frei, J.K. 1 973 a - Orchid ecology in a cloud forest in the
mycorrhizal fungi to the problem of symbiosis. - mountains of Oaxaca, Mexico. - Amer. Orch. Soc.
Amer. J. Bot. 26 :390-399. Bull. 4 2 :307-3 1 4.
1 946 - Nutrient supply of epiphytic orchids in the 1 973 b - Effect of bark substrate on germination and
mountains of Haiti. - Ecology 27 :264-266. early growth of Encyclia tampensis seeds. - Ibid.
1 94 7 - Ecological observations on the orchids of 4 2 : 70 1 -708.
Haiti. - Amer. Orch. Soc. Bull. 1 6 :262-269. Freise, F. 1 936 - Das Binnenklima von Urwiildern in
19 5 2 - Outline for ecological life history studies of subtropischen Brasilien. - Petermanns Mitt. 8 2 : 30 1 -
vascular epiphyte plants. - Ecology 3 3 :5 50-5 5 8 . 307.
Curry-Lindahl, K. 1 969 - Report t o t h e government of Furman, T.E. 1 95 9 - Structural connections between
Liberia on conservation, management and utilization epiphytes and host-plants. - Proc. 9 :th lnt. Bot.
of wildlife resources. -,--- I.U.C.N. Morges, Switzerland. Congr. 2 : 1 27. Toronto.
(Stencil). Garay, L.A. 1 972 - On the origin of Orchidaceae. 11. - J.
Davies, T.A.W. & Richards, P.W. 1 93 3 , 1 934 - The Arnold. Arbor. 5 3 :202-2 1 5.
vegetation of Moraballi Creek, British Guiana. An Gessner, F. 1 9 56 - W asserausha1t der Epiphyten und
ecological study of a limited area of tropical rain Lianen. - In Ruhland, W. (ed.) 1 95 6 :9 1 5 -9 50.
forest. Parts I and 11. -J. Ecol. 2 1 : 3 50-3 84 ; 22 : 1 06- Goebel, K. 1 888 - Morphologische und Biologische Stu­
1 55 . dien. I. Ueber Epiphytische Fame und Muscineen. -
Dirmhirn, I . 1 96 1 - Light intensity at different levels. - Ann. J ard. Bot. Buitenz. 7 : 1 - 7 3.
Trans. Roy. Ent. Soc. London 1 1 3(2) :270-27 5 . 1 922 - Erdwurzeln mit Velamen. - Flora (N.F.) 1 5 : 1 -
Dudgeon, W. 1 923 - Successions o f epiphytes in the 26.
Quercus incana forest at Land our, Western Golley, P.M. & Golley, F.B. (ed.) 1 972 - Tropical
Himalayas. - Journ. Ind. Bot. Soc. 3 :2 70. ecology with an emphasis on organic productivity. -
Dunn, H.A. 1 94 1 - The major orchids of Panama. The Papers from a symp. on trop. ecology, in New Dehli
physiography of their habitats. - Amer. Orch. Soc. 1 97 1 . Athens, Ga.
Bull. 9 :25 1 -267. Gore, A.J.P. 1 968 - The supply of six elements by rain to
Du Rietz, G.E. 1 9 32 - Zur Vegetationsokologie der os­ an upland peat area. - J. Ecol. 5 6 :483-495.
tschwedischen Kustenfelsen. - Beih. Bot. Centralbl. Gorgla, T.A. 1 96 9 - Inventory results, East Nimba
49 :6 1 - 1 1 2. National Forest and parts adjacent. -_Report to Lam­
1 945 - Om fattigbark- och rikbarksamhiillen. - eo J.V. Op. Co. Yekepa, Liberia. (Stencil).
Svensk Bot. Tidskr. 3 9 : 147- 1 50. Greendale, D.J. & Nye, P.H. 1 964 - Organic matter and
Dycus, A.M. & Knudson, L. 1 95 7 - The role of the nutrient cycles under moist tropical forests. - Proc.
velamen of the aerial roots of orchids. - Bot. Gaz. 1 0 :th Int. Bot. Congr. :248. Edinburgh.
1 1 9 :78-87. Grubb, P.J., Lloyd, J.R., Pennington, T.D. & Whitmore,
Eaton, J.S., Linkens, G.E. & Bormann, H.F. 1 97 3 - T.C . 1 963 - A comparison of montane and lowland
Throughfall and stemflow chemistry in a northern rain forests in Ecuador. - J. Ecol. 5 1 :567-60 1 .
hardwood forest. - J. Ecol. 6 1 :495-508. Guillaumet, J. 1 967 - Recherches sur la vegetation e t la
Eggeling, W.J. 1 947 - Observations on the ecology of the flore de la region du Bas C avally (Cote d'Ivoire). -
Budongo rain forest, Uganda. - Ibid. 3 4 :20-87. O.R.S.T.O.M., Paris-Orsay.
Eidmann, H. 1 94 1 - Meine Forschungsreise nach Gusinde, M. & Lauscher, F. 1 94 1 - Meterologische
Spanisch Guinea. - D. Biologie 1 0 : 1 - 1 3. Beobachtungen im Kongo-Urwald. - Sitzungs. Akad.
Erickson, L.C. 1 95 7 - Respiration and photosynthesis in Wiss. Wien. 1 50 :2 1 8-347.
Cattleya roots. - Amer. Orch. Soc. Bull. 26 :40 1 -402. Gardlund, T. 1 96 7 - Lamco i Liberia. - A.-W.,
Evans, G.C . 1 939 - Ecological studies on the rain forest Stockholm.
of Southern Nigeria. 11. The atmospheric environmen­ Hall, B.J. & Bowling, J.C. 1 96 9 - Field key to the
tal conditions. - J. Ecol. 27 :43 6-482. epiphytic orchids of Ghana based on characters of
1 968 - Model and measurement in the study of shoots and infructescences. - Adansonia 9( 1 ) : 1 39-
woodland light climates. - In Bainbridge, R., et al. 1 55 .
(ed.) 1 96 8 : 5 3-76. Harley, G.W. 1 94 1 - Native african medicine, with
Evans, G.C., Whitmore, T.C . & Wong 1 960 - The dis­ special reference to its practice in the Mano tribe of
tribution of light reaching the ground vegetation in a Liberia. - Cambridge, Mass.
tropical forest. - J. Ecol. 48 : 1 93 -204. Harley, W.J. 1 95 5 - The ferns of Liberia. - Contrib.
Exell, A.W. & Launert, E. (ed.) 1 9 70 - Flora Gray Herb. Harvard Univ. 1 77 :5 8- 1 0 1 .
Zambesiaca. X, Pteridophyta. - Crown agents, Lon­ 1 95 6 - Plants of the tree trunks i n Liberia. - Amer.
don. Fern J. 46 :65-96.

A cta phytogeogr. suec. 59


126 Dick Johansson

- 1 9 5 9 - Handbook of Liberian ferns. - Ganta, Liberia. Hutchinson, J. & D alziel, J . M . 1 95 4 - Flora of West
Harris, J .A. 1 9 1 8 - On the osmotic concentration of the Tropical Africa. 2 :nd ed. Vol. 1 , part I. Revised by
tissue fluids of phanerogamic epiphytes. - Amer. J . Keay, R . W .J . � Crown agents, London .
B ot. 5 :490- 506. 1 9 5 8 - Ditto . Vol. I, part 2. Revised by Key, R . W . J .
1 9 34 - The physio-chemical properties of plant saps in - Ibid.
relation to phytogeography. - M inneapolis. 1 96 3 - Ditto. Vol. 2. Edited by Hepper, F.N. - Ibid.
<2.
H arvais, G. & H adley, G. 1 9 7 - The relation between 1 96 8 - Ditto. Vol. 3 , part 1 . Revision edited by
host and endophyte in orchid m ycorrhiza. - N ew Hepper, F . N . - Ibid.
P hytol. 66 : 205-2 1 5 . Iwanoff, L. 1 9 28 Zur M ethodik der
H awkes, A.D. 1 965 - Encyclopaedia of cultivated Transpirationsbestimmung am Standort. - B er.
orchids. - Faber & Faber, London. Deutsch. Bot. Gesell. 43 :306-3 1 0.
Heagarty, C . P. 1 9 5 5 - O bservations on the germination Jaeger, P., H alle, N . & Adam, J . G. 1 9 6 8 - Contribution a
of orchid seeds . - Amer. Orch. Soc. B ull. 24 :45 7-464. la etude des Orchidees des Monts Loma (Sierra
Hilitzer, A . 1 9 25 - La vegetation epiphyte de l a B oheme. Leone). - Adansonia 8 :2 6 5 - 3 1 0.
- P ubl. Fac. Se. Univ. Charles, Prague 4 1 : 1 -200. Jaeger, P. & Adam, J . G . 1 9 72 - Contribution a !'etude de
Holdridge, L . R . 1 9 70 - A system for representing struc­ la vegetation des Monts Loma (Sierra Leone). -
ture i n tropical forest associations. - In Odum, H .T. & Compte rendu des seances de la societe de
Pigeon, R . F . (ed . ) 1 9 70, C h apter B - 1 2. biogeographie 422-424 : 7 7 - 1 03 .
Holtum, R . E . 1 960 - The ecology of tropical epiphytic Jarvi s, P . G. & J arvis, M . S . 1 96 3 - The water relations of
orchids. - Proc. Third World Orch. C onference. Lon­ tree seedlings. I V Some aspects of the tissue water
don. relations and drought resistance. - P hysiol. Plant.
Hopkins, B. 1 9 70 - Vegetation of the Olokemeji forest 1 6 : 5 0 1 -5 1 6.
reserve, N igeria. VI, The plants of the forest site with Johansson, D. 1 9 74 - Rhipidoglossum paucifolium, a
special reference to their seasonal growth. - J . Ecol. new African species of Orchidaceae. - B ot. N ot.
5 8 : 7 65 - 7 9 3 . 1 2 7 ( 1 ).
Hosokawa, T. 1 943 - Studies o n t h e life-forms o f Johnson, P . L. & Atwood, D . M . 1 9 70 - Aerial sensing
vascular epiphytes a n d t h e epiphyte flora o f Ponape, and photographic study of the El Verde rain forest. -
M icronesia. - Trans. N at. Hist. Soc. Taiwan 3 3 : 3 5 - In Odum, H.T. & Pigeon, R . F . (ed.) 1 9 70, C h apter B
5 5 , 7 1 -8 9 , 1 1 3 - 1 4 1 . 5.
1 94 9 . Studies of the life-form of vascular epiphytes and Jones, E . 1 9 60 - Contribution o f rainwater to the nutrient
the spectrum of their life-forms. - J . J ap. B ot. 24 :4 1 - economy of Northern N igeria. - Nature 1 8 8 :4 3 2.
45. K amerling, Z. 1 9 1 2 - De verdampin·g van epiphyte
1 9 5 0 - Epiphyte-quotient. - B ot. M a g . Tokyo 63 : 1 8- Orchideen. N atuurk. Tijd schr. N ed.-Ind.
20. 7 1 :5 4-72.�
1 9 5 2 a - A plant-sociological study in the mossy Karsten, G. 1 8 95 - M orphologische und biologi sche
forests of Micronesian i sl ands. - Mem. F ac . Sci. Untersuchungen iiber einige Epiphytenformen der
Kyushu Univ. E ( 1 ) :6 5 - 8 2 . Molukken. - Ann. J ard. B ot. B uitenz. 1 2 : 1 1 7- 1 9 5 .
1 9 25 - Uber mantelformige organe bei Epiphyten und
1 9 5 2 b - A synchrological study in t h e swampforests
Wurzelklettern. ·- Flora 1 8- 1 9 ( N . F.) :3 00- 3 1 I.
in the Micronesian Islands . - I bid. : 1 0 1 - 1 23 .
K eay, R . W.J. 1 9 5 9 - Explanatory notes to vegetation
1 9 5 3 - O n the nomenclature of aerosynusiae. - P roc.
map of Afric a published on behalf of A . E.T.F.A.T. ­
7 :th. lnt. B ot. Congr. :69 1 -694. Stockholm.
Oxford Univ. Press, London.
1 95 4 a - On the structure and composition of the
Kerr, A . D . 1 9 72 - The leafless Orchids. - Amer. Orch.
C ampnosperma forests in P alau, M ironesia. - Mem.
Soc. Bull. 4 1 : 3 07-309.
Fac. Se. Kyushu Uniy. E( 1 ) : 1 99-2 1 8 .
K linge, H. 1 963 - Vber Epiphytenhumus aus El
1 9 54 b - On the C ampnosperma forests of Yap,
Salvador. - Pedobiologia 2 (2) : 1 02- 1 07 .
Ponape and K usaie in M icronesia. - I bid. : 2 1 9-243.
Knapp, R. 1 9 73 - D ie Vegetation von Afrika. - Fisher,
1 95 4 c - On the C ampnosperma forests of Kusaie in Stuttgart.
M icronesia, with special reference to the community Knudson, L. 1 9 22 - Non-symbiotic germination of
units of epiphytes. - Vegetatio 5 - 6 : 3 5 1 -3 60. orchid seeds. - B ot. Gaz. 73 : 1 - 2 5 .
1 95 5 - On the vascular epiphyte comm unities in Kochenderfer, J. & Lee, R . 1 9 73 - Indexes t o transpira­
tropical rainforests of M icronesia. - Mem. Fac. Sci. tion by forest trees. - Oecologogica Plantarum 8 : 1 7 5 -
Kyushu Univ. E(2) : 3 1 -44. 1 84.
1 95 7 - Outline of the mangrove and strand forests of Kolkwitz, R . 1 9 3 2 - Urwald und epiphyten. - Ber.
the M icronesian i sl ands. - Ibid. (2) : 1 0 1 - 1 1 8 . Deutsch. Bot. Gesell . 50 : 1 1 0- 1 1 6 .
1 9 6 8 - Ecological studies of tropical epiphytes in K unkel, G. 1 9 62 a - Uber die F arne Liberias ihre
forest ecosystem. - In M isra, R. & Gopal, B. (ed.) V orkommen und Verbreitung. - Ber. Schweiz. B ot.
·
1 968 :482-50 1 . Gesell. 72.
Hunt, P . F . 1 9 6 7 - Size of the orchid family. - Orch. Rev. 1 9 62 b - Aufzeichnungen iiber die Vegetation des Mt.
75 :229. Wutivi-Gebietes (Nordwest-Liberia). - Wildenowia

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 12 7

3 : 1 5 1 - 1 68 . Oliver, W . R . B . 1 9 30 - N ew Zealand epiphytes. - J. Ecol.


1 96 5 - The trees o f Liberia. - German Forest. Miss. 1 8 : 1 -50.
to Liberia, Report 3 . BL V , Mi.inchen. Omura, M . 1 95 3 - Syn.ecological studies on t h e epiphyte
1 96 6 - Einige Temperaturmessungen in liberianischen communities of lichens in the temperate forests of the
Regenwald. - Zeitschrift fi.ir Meteorologie 1 9 :5 1 - 5 9 . I de and Titib u mountain regions in J apan. - Mem.
Kup, A . P . (ed. ) 1 96 7 - Adam Afzelius Sierra Leone J our­ Fac . Sci. K yushu Univ. E 1 (3) : 1 47- 1 5 7 .
nal 1 7 9 5 - 1 796. - Studia Ethnographica Upsaliensia Oye, P. van, 1 92 1 - Influence des facteur climatiques sur
27. l a repartition des epiphytes a l a surface des troncs
Lamotte, M., Leclerc, J . C . , Richard-Molard, J . , Rougeri, d'arbres a J ava. - Rev. Gen. B ot. 3 3 : 1 6 1 - 1 76 .
G. & Porteres, P . 1 95 5 - La chaine du N im ba. Essai 1 924 a - S u r l'ecologie d e s epiphytes a l a surface des
geographique. - Mem. I .F . A . N . 4 3 . troncs d'arbres a J ava. - I bid. 3 6 : 1 -30, 68-83.
Lebrun, J . 1 93 7 - Observations s u r l e s epiphytes de l a 1 924 b - Ecologie d e s epiphytes d e s troncs d'arbres au
foret equatoriale congolaise. - S o c . Scientif. Bruxelles Congo Beige. - Ibid. 3 6: 48 1 -498.
5 7 :3 1 -3 8 . Paulian, R . 1 945 - La voute de l a foret tropicale, milieu
Longman, H . A . & White, C .T. 1 9 1 7 - The flora o f a sin­ biologique. - Rev. Se. 8 3 :28 1 -286.
gle tree. - Proc. Roy. Soc. Queensland 2 9 : 64-69. Pennfound, W . T . & Deiler, F.G. 1 94 7 - On t h e ecology
M ayer, K .1 95 1 - Forest resources of Liberia. - U.S. of the Spanish moss. - Ecology 2 8 :455-458.
Dept. Agricult. Bull.67, Washington, D . C . Pessin, L.J. 1 92 5 - An ecological study of t h e pqlypody
Miehe, H . 1 9 1 1 Javanische Studien. Z u r Frage der
- fern Polypodium polypodioides as an epiphyte in
mikrobiologische Vorgange im Humus e1mger Mississippi. - Ecology 6 : 1 7-3 8 .
humussammelender Epiphyten. - A bh . Sachs. Ges . Picado, C . 1 9 1 3 - L e s Bromeliacees epiphytes, con­
Wiss. (Math. -Phys.) 3 2 : 3 76- 3 9 8 . siderees comme milieu biologique. - Bull. sci. Fr. Belg.
Mildbraed, J . 1 922 - Wissenschaftliche Ergebnisse der Ser. 7, 4 7 :2 1 5 - 3 60.
zweiten deutschen Zentral-Afrika-Expedition 1 9 1 0- Piers, F . 1 95 9 - Orchids of East Africa. - Nairobi.
1 9 1 1 unter Fi.ihrung Adolf Friedrichs, Herzog zu - 1 96 8 -Orchids of East Africa. 2 :nd ed. - C r amer,
Mecklenburg. - Leipzig. Lehre
.
M ilne-Redhead, E. & Polhill, R.M. (ed.) 1 96 8 - Flora of Pijl, L. van der, & Dodson, C . H. 1 96 9 - Orchid flowers.
Tropical East Africa. Orchidaceae, Part I, by Their pollination and evolution. - University of M i ami
S ummerhayes, V.S. - C rown agents, London. Press. Coral Gables, Fa.
Misra, R. & Gopal, B. (ed.) 1 96 8 - Proceedings of the Pittendrigh, C .S. 1 94 8 - The Broemliad-Anopheles­
symposium on recent advances in tropical ecology, IL M al aria complex in Trinidad. I The B romeliad flora. -
- V aranasi India. Evolution 2 :5 8-89.
Moir, W.W.G. 1 963 - Aerangis ellisii. - Amer. Orch. Polak , B . 1 9 33 - E e n tocht in h e t zandsteen gebid bij
Soc. Bull. 3 2 : 803-806. M andor (West Borneo). - Trop. N atuur 22 :23-28.
Moreau, R . E. 1 93 5 - Synecological study of U s ambara, Richards, P .W. 1 93 6 - Ecological observations on the
Tanganyika territory. - J . Ecol. 2 3 : 1 -43. rain forest of Mt Dulit Sarawak. - J. Ecol. 24, Part 1 :
Moreau, W . M . & Moreau, R . E . 1 943 - An introduction 1 - 3 7 , Part 1 1 : 340-360.
to the epiphytic orchids of East Africa. - East. Africa 1 93 9 - Ecological studies on the rain forest of
N at. Hist. Soc. N airobi. Southern Nigeria. I . The structure and floristic com­
Morris, B. 1 968 - The epiphytic orchids of the Shire position of the primary forest. - J. Ecol.2 7 : 1 -6 1 .
Highlands, M al awi . - Proc. Linn. Soc. 1 79 ( 1 ): 5 1 -66. 1 964 - The tropical rain forest. - C ambridge Univ.
- 1 9 70 - Th e epiphytic orchids of M al awi. - The Socie­ Press, London.
ty of M al awi, Blantyre. Richards, P . W . , Tansley, A . G. & Watt, A . S . 1 940 - The
Newman, I. 1 954 - Locating strata in tropical rain forest. recordings of structure, life form and flora of tropical
- J . Ecol. 4 2 : 2 1 8-2 1 9. forest communities as a basis for their classification. ­
Nihoul, E., Schelpe, E.A. & Hunt, P . F . 1 969 - A J. Ecol.
28 :224-23 9 .
provisional checklist of the orchids in the C ongo­ Robbins, R . G. 1 95 9 - The u s e of t h e profile diagram i n
Kinshasa. - Amer. Orch. Soc. Bull. 3 8 :5 78-5 84. rain forest ecology. - J . Bioi. Sci. 2 (2) :5 3-63.
Ochsner, F. 1 928 - Studien i.iber die Epiphyten­ Romberger, J . A . & Mikola, P . (ed. ) 1 9 64 - International
Vegetation der Schweiz. - J ahrb. St. Gall. Naturw. review of forestry research. - New York.
Gesell .
63 : 1 - 1 08 . Rufelt, H. 1 963 - Rapid changes in transpiration in
Odum, H . T . 1 970 a - R ain forest structure and mineral plants. - Nature 1 9 7 :985-986.
cycling homeostasis. - In Odum, H.T. & Pigeon, R . F . Ruhl and, W. (ed.) 1 95 6 - Encyclopedi a of plant
(ed.) 1 9 70, Ch apter H 1 : 3 -5 2. physiology. - Springer, Berlin.
1 9 70 b - Holes in leaves and the grazing control Ruinen, J. 1 95 3 - Epiphytosis. A second view on
mechanism. - In Odum, H.T. & Pigeon, R . F . (ed .) epiphytis m . - Ann. B ogoriensis 1 : 1 0 1 - 157.
1 9 70, C h apter I 6 :69-80. 1 9 5 6 - Occurrence o f B eijerinckia species i n the
Odum, H.T. & Pigeon, R . F. (ed . ) 1 970 - A tropical rain 'phyllosphere'. - Nature 1 77 : 1 22- 1 2 3 .
forest. Book 1 -3 . - National technical Inform. Service, 1 96 1 - The phyllosphere. I An ecological neglected
Springfield, V a. m ilieu. - Plant and Soil. 1 5 :8 1 - 1 09 .

A cta phytogeogr. suec. 59


128 Dick Johansson

Rupp, H.M.R. 1 96 9 - The orchids of New South Wales. Stewart, J. & C ampbell, B. 1 9 70 - Orchids of tropical
- National herb., Sydney. Africa. - Allen & Co., London.
Rutter, A.J. & Whitehead, F.H. (ed.) 1 9 63 - The water Stacker, 0. 1 929 - Eine Feldmethode zur Bestimmung
relations of plants. - British Ecological Soc. Symp. der momentanen Transpirations- und
Ill. Blackwell, Oxford and Edinburgh. Evaporationsgrosse. I, Il. - Ber. Deutsch. Bot. Gesell.
Sanford, W.W. 1 96 7 - Orchids of West Africa. - Amer. 4 7 : 1 26- 1 29, & 1 30- 1 3 6 .
Orch. Soc. Bull. 3 6 :963-969. Sulit, M.D. 1 9 5 0 - Field observations on trees hosts of
1 96 8 a - The Genus Aerangis. - Ibid. 37: 490-4 93. orchids in the Philippines. - Philip. Orch. Rev. 3 : 3 - 8 .
1 96 8 b - Distribution of epiphytic orchids in semi­ Summerhayes, V.S. 1 96 8 - Orchidaceae, Part 1 . - In
deciduous tropical forest in Southern Nigeria. - J. Milne-Redhead, E. & Polhill, R.M. (ed.) 1 9 68.
Ecol. 5 6 : 6 9 7- 705 . Swanson, R.H. & Lee, R. 1 96 6 - Measurement of water
1 96 9 - The distribution of epiphytic orchids in Nigeria movement from and through shrubs and trees. - J.
in relation to each other and to geographic location Forest. 64 : 1 8 7- 1 90.
and climate, type of vegetation and tree species. - Teuscher, H. 1 9 7 2 - Microcoelia guyoniana and other
Bioi. J. Linn. Soc. 1 : 24 7 - 2 8 5 . leafless epiphytic orchids. - Amer. Orch. Soc. Bull.
1 97 1 - The flowering time of West African orchids. - 4 1 :497-5 0 1 .
Bot. J. Linn. Soc. 64 : 1 6 3 - 1 8 1 . Thornton, I. 1 9 6 5 - Nutrient content o f rainwater i n the
1 9 7 2 - Epiphytic orchids as characterizers of vegeta­ Gambia. - Nature 205 : 1 0 2 5 .
tion in West Africa. - A paper presented at the World Thorold, C .A. 1 9 5 2 - The epiphytes of Theobroma cacao
Orchid Conference at Mendelin, Colombia 1 9 7 2. in Nigeria in relation to the incidence of Black Pod dis­
Sanford, W.W. & Adanlawo, I. 1 9 7 3 - Velamen and ex­ eases (Phytophthora palmivora). - J. Ecol. 40 : 1 25 -
odermis characters of West African epiphytic orchids 1 42.
in relation to taxonomic grouping and habitat Tixier, P. 1 9 66 - Flore et vegetation orophiles de l'Asie
tolerance. - Bot. J. Linn. Soc. 66 : 307-3 2 1 . tropicale. - Soc. d 'edition d'enseignement superieur,
Scaetta, H . 1 93 7 - L a genese climatique des sots mon­ Paris.
tagnards de 1' Afrique Centrale. Les formations Tournay, R. 1 9 5 5 - Orchidaceae africanae. 1 . Les
vegetates qui en caracterisent les stades de degrada­ orchidees congolaises : caracteres vegetatifs et floraux,
tion. - Inst. Roy. Colonial Beige, Bruxelles. types biologiques, clef de determination des genres
Schelpe, E.A.C .L.E. 1 9 70 - Pteridophyta. - In Exell, sympodiaux, apen;u les genres. - Bull. Soc. Roy. Bot.
A.W. & Launert, E. (ed.) 1 9 70. Belg. 8 7 : 5 7 - 7 9.
Schimper, A.F.W. 1 88 8 - Die epiphytische Vegetation Tukey, H.G., Jr. 1 9 70 - Leaching of rnetablites from
Amerikas. - Bot. Mitt. Trop. 2. foliage and its implication. - In Odum, H.T. & Pigeon,
1 903 - Plant geography upon a physiological basis. R.F. (ed.) 1 9 70, Chapter H 1 1 .
Oxford. Vanderyst, R.P.H. 1 9 2 2 - Nouvelle contribution a !'etude
1 9 3 5 - Pflantzengeographie auf Physiologischen des Ficus epiphytes sur l'Elaeis. - Rev. Zool. Afric.
Grundlage. - Revised by Faber, A.F., Jena. 1 0( 2 ) .
Schnell, R. 1 9 5 0 - La foret dense. - Lechevalier, P aris­ Wallach, A. 1 9 3 9 - Beitrage zur Kenntnis der
VI. Wasseraufnahme durch die Luftwurzeln tropischer
1 9 5 2 - Vegetation et flore de la region montagneuse Orchideen. - Zeitschr. f. Bot. 3 3 : 443-468.
du Nimba. - Mem. I.F.A.N. 22, D akar. Waiter, H. 1 9 7 1 - Ecology of tropical and subtropical
1 9 70 - Phytogeographie des payes tropicaux. I -11. - vegetation. - Oliver & Boyd, Edinburgh.
Gauthier-Vitlars, Paris. Visser, S.A. 1 9 64 - Origin of nitrates in tropical
Sernander, R. 1 9 1 2 - Studier ofver lafvarnes biologi. I . rainwater. - Nature 20 1 : 3 5 .
Nitrofila lafvar. - Syensk Bot. Tidskr. 6 : 803 - 8 8 3 . Voorhoeve, A.G. 1 9 65 - Liberian high forest trees. -
Sharp, A.J. 1 95 7 - Vascular epiphytes in the Great Pudoc, W ageningen.
Smoky mountains. - Ecology 3 8 : 6 3 5 - 6 54. Voth, P.H. 1 9 3 9 - C onduction of rainfall by plant stem in
Sjogren, E. 1 9 6 1 - Epiphytische Moosvegetation in a tropical rain forest. - Bot. Gaz. 1 0 1 : 3 2 8-3 40.
Laubwaldern der Insel Oland (Schweden). - Acta Voute, A.D. 1 9 46 - Regulation of the density of the in­
Phytogeogr. Suec. 44. Uppsala. sect populations in virgin-forests and cultivated woods.
Skye, E. 1 9 68 - Lichens and air pollution. - Ibid. 5 2. - Arch. Neerl. Zool. 7 :4 3 5 -4 70.
Smith, J.J. 1 9 2 5 � Ephemeral orchids. - Ann. Jard. Bot. 1 9 64 - Harmonious control of forest insects. - In
Buitenz. 3 5 : 5 5 -70. Romberger, J.A. & Mikola, P. Vol. I 1 9 6 4 : 3 25 -3 8 3 .
Spanner, L. 1 9 3 9 - Untersuchungen iiber den Warme Went, F.A.F.C. 1 8 9 5 - Ueber Haft- und Nahrwurzeln bei
und Wasserhaushalt von Myrmecodia und Hyd­ Kletterpflanzen und Epiphyten. - Ann. Jard. Bot.
nophytum. - Jahrb. Wiss. Bot. 8 8 : 24 2-283. Buitenz. 1 2 : 1 -7 2 .
Steenis, C .G.J.J., van 1 93 8 - Recent progress and Went, F.W. 1 9 3 1 - Over de sociologie der Epiphyten. ­
prospects in the study of the M alaysian flora. - Handel . 6 Ned. Ind. Naturw. Congr. : 3 8 1 -3 8 7. Ban­
Chron. Bot. 4 : 3 92-3 9 7. doeng.
Stewart, J. 1 9 73 - More about Microcoelias. - Amer. 1 9 40 - Soziologie der Epiphyten eines tropischen
Orch. Soc. Bull. 42 :80 1 -805. Urwaldes. - Ann. J ard. Bot. Buitenz. 5 0 : 1 -9 8 .

A cta phytogeogr. suec. 59


Ecology of vascular epiphytes in West African rain forest 1 29

Whitmore, T.C. 1 9 6 8 - A study of light conditions in Koopowitz, H. 1 964 - Aerangis- an extrapolation from
forests in Ecuador with some suggestions for further ecology to cultivation. - Amer. Orch. Soc. Bull.
studies in tropical forests. - In Bainbridge et al. (ed.) 3 3 : 1 026- 1 02 8 .
1 96 8 : 2 3 5 - 24 7 . Kunkel, G. 1 9 63 - Die einfliisse von Kultur und Siedlung
Wiesner, J. 1 89 7 - Zur Physiologie von Taeniophyllum auf das Waldbild Liberias. - Zeitschrift fiir
Zollingeri. - S.B. Akad. Wiss. Wien, Math.-Naturw. Wirtschaftsgeogr. 7 :4 5 - 5 1 .
C l . 1 06 : 7 7- 9 8 . Leakey, C . L.A. 1 968 a - The orchids of Uganda. -
Willis, A.J. & Jetferies, R.L. ( 1 9 6 3) - Investigations of Amer. Orch. Soc. Bull. 3 7 : 8 8 3- 8 8 6.
the water relations of sand-dune plants under natural 1 96 8 b - The orchids of Uganda. The major
conditions. - In Rutter, A. & Whitehead, F. (ed.) angraecoid genera. - Ibid. 3 7 : 9 7 8 - 9 8 3 .
1 963. 1 9 6 8 c - The orchids of Uganda. Various species of
Willis, J.A., Yemm, E.W. & B alasubramaniam, S. 1 9 63 - horticultural merit. - Ibid. 3 7 : 1 05 2 - 1 0 5 5 .
Transpiration phenomena in detached leaves. - Morton, J.K. 1 9 6 1 - West african lilies and orchids. -
Nature 1 99 : 2 6 5 - 2 6 6 . Longmans, Green & Co ... London.
WiJlis, J. C . 1 96 6 - A dictionary of flowering plants & Onochie, C .F.A. 1 9 6 2 - The use of field characters in the
ferns. - Seventh ed., revised by Airy Shaw, H.K., tree identification. - C omptes Rendus, IV Reunion
C ambridge Univ. Press, London. pleniere A.E.T.F.A.T. 1 9 6 1 . Lisboa.
Withner, C . L. (ed.) 1 9 5 9 - The orchids, a scientific sur­ Piers, F. 1 96 8 - Orchids of Africa. Angraecum distichum
vey. - Ronald Press, New York. and A. aporoides. - Amer. Orch. Soc. Bull. 3 7 :9 9 8-
- 1 9 5 9 - Orchid physiology. - In Withner (ed.) 999.
1 9 5 9 : 3 1 5 - 3 60. Richards, P.W. 1 9 5 7 - Ecological notes on West African
Yarranton, G.A. 1 96 7 - An instrument for measuring the vegetation. - J. Ecol. 45 : 5 6 3 - 5 7 7 .
microrelief of the bark. - C an. J. Bot. 4 5 (8) : 1 1 73 - 1 9 6 3 - Ecological notes on West African vegetation.
1 1 78. Ill. The upland forest of Cameroon Mountain. - J.
Ecol. 5 1 : 5 29-544.
Sanford, W.W. 1 9 70 a - Plectrelminthus caudatus. -
Literature not cited in the text but related to the sub­ Amer. Orch. Soc. Bull. 3 9 : 224-226.
1 9 70 b - Practical conservation of orchids in Nigeria.
ject
- Nig. J. Se. 4 :49-5 7 .
Briscoe, M.S. 1 95 2 - The relation of insects and insect­ 1 9 70 c - Orchids of the Bamboutos and Manengouba
borne diseases to the vegetation and environment in highlands in East Cameroun. - Amer. Orch. Soc. Bull.
Liberia. - Ecology 3 3 : 1 8 7-2 1 4. 3 9 : 6 8 1 -6 8 5 .
Bryant, M.T. 1 9 66 Some of our palms. - The Liberian
-
1 9 70 d - A selection o f West African orchids. - Ibid.
naturalist 4( I ) : 9 - 1 3 . Monrovia. (Stencil). 3 9 : 8 90-8 9 5 .
- 1 96 6 Trees
- roadsidP.s . - Ibid. 1971 - The orchid flora o f Equatorial Guinea i n rela­
4(2) :9-20.
tion to that of West Africa. - Mitt. Bot. 1 0 : 2 7 8-298.
Hopkins, B. 1 96 5 Forest and savanna. - Heinemann,
-
Schelpe, E.A.C .L.E. - 1 9 66 - An introduction of the
London.
South African orchids. - MacDonald, London.
Harrison, E.R. 1 9 72 - Epiphytic orchids of Southern Strong, R.H. (ed.) 1 9 30 - The African Republic of
Africa. - The Natal branch of the Wildlife Preserv. Liberia and the Belgian Congo. I, 11. - C ambridge,
and Conserv. Soc. of South Africa. Durban. Mass.
Jaeger, P. & Summerhayes, V.S. 1 94 9 - Note sur Warner, R. 1 94 5 - Vegetation and agriculture of Liberia
quelques Orchidees recoltees dans les Monts Loma and adjacent West Africa. - Iowa A cad. Sci. 5 2 : 1 7 1 -
(Sierra Leone) et les contrees limitrophes. - Kew. Bull. 1 89 .
3 :4 7 5 - 4 8 3 .
Kloppers, J. 1 9 7 1 - Die Suid-afrikaanse epifiete. - S.
African Orch. J. March 1 9 7 1 : 20-23.

A cta phytogeogr. suec. 59


A C TA P H Y T O G E O G R A P H I C A S U E C I C A

1. E. A lmquist, Upplands vegetation och flora. (Vegeta­ mossflora i Uppsalatrakten. (Summary : Moss flora
tion and flora of Uppland.) 1 9 29. 4 0 : - ( 2 7 :-). I S B N and moss vegetation in the neighbourhood of Uppsa­
9 1 - 7 2 1 0-00 1 -X . la.) 1 9 4 5 . 2 7 : - ( 1 8 :-). I SB N 9 1 - 72 1 0-0 1 9-2.
2. S . Thunmark, D e r See Fiolen und seine Vegetation. 20. N. A lbertson, Gsterplana hed. Ett alvaromrade pa
1 9 3 1 . 2 1 :- ( 1 4 :-). I S B N 9 1 -7 2 1 0-002-8. Ki nnekulle. (Zusammenf. : G sterplana hed. Ein Alvar­
3 . G.E. Du R ietz, Life-forms of terrestrial flowering gebiet au f dem Kinnekulle.) 1 94 6 . 2 7 :- ( 1 8 : - ). I S B N
plants. I. i 9 3 1 . 1 5 :- ( I 0 :-) . I S B N 9 1 -7 2 1 0-003-6. 9 1 - 72 1 0-020-6.
4 . B. Lindquist, Om den vildvaxande skogs almens raser 2 1 . H. Sjdrs, Myrvegetation i B ergslagen. (Summary :
och deras utbredning i Nordvasteuropa. (Summary : Mire vegetation in Bergslagen, Sweden.) 1 94 8 . 3 6 :­
The races of spontaneous Ulmus glabra Huds. and (24 :-). I S B N 9 1 - 72 1 0-02 1 -4 .
their d istribution in N W . Europe.) 1 9 3 2. 1 2 :- ( 8 : -). 22. S . A hlner, Utbredningstyper bland nordiska barr­
ISBN 9 1 - 7 2 1 0-004-4. trad slavar. (Zusammenf. : Verbreitungstypen unter
5. H. Osvald, Vegetation of the Pacific coast bogs of fennos kandi schen N adelbaumflechten.) 1 94 8 . 3 0 : ­
N orth America. 1 9 3 3 . 1 2 :- ( 8 :-). I S B N 9 1 - 7 2 1 0- (20: -). I S B N 9 1 - 72 1 0-022-2.
005 -2. 23. E. Julin, Vessers· udde. M ark och vegetation i en igen­
6. G. Samuelsson, Die Verbreitung der hoheren Was ser­ vaxande lovang vid Bj arka-Saby. (Zusammenf. : Ves­
pflanzen in N ordeuropa. 1 9 3 4 . 2 1 :- ( 1 4 :-). I S B N 9 1 - sers udde. Boden und Vegetation in einer verwachsen­
72 1 0-006-0. den Laubwiese bei Bj arka-S aby in Gstergotland, Siid­
7 . G. Degelius, Das ozeanische Element der Strauch­ schweden.) 1 94 8 . 2 1 :- ( 1 4 :-). I S B N 9 1 - 7 2 1 0-023 -0.
und Laubflechtenflora von Skandinavien. 1 9 3 5 . 3 6 :­ 24. M. Fries, Den nordi ska utbredningen av Lactuc a
(24 :-). I S B N 9 1 - 7 2 1 0-007-9. alpina, A conitum septentrionale, R anunculus platani­
8 . R . Serna nder, Granskar och F i b y urskog. En studie folius och Polygonatum verticillatum. (Z usammenf. :
over stormluckornas och marbuskarnas betydelse i Die nordis che Verbreitung von Lactuca alpina. . . )
den svenska granskogens regeneration. (Summary : 1 94 9 . 1 5 : - ( 1 0 : -) . I S B N 9 1 - 7 2 1 0-024-9.
The primitive forests of Granskar and Fiby. A study 25. 0. Gjcerevoll, Sn0leievegetasjonen i Oviksfjellene.
of the part played by storm-gaps and dwarf trees in the ,(Summary : The snow-bed vegetation of Mts Oviks­
regeneration of the Swedish spruce forest.) 1 9 3 6 . 2 1 :­ f] allen, J amtland, Sweden.) 1 9 49. 1 8 : - ( 1 2 :-). I S B N
( 1 4 :-). I S B N 9 1 - 7 2 1 0-008- 7 . 9 1 - 7 2 1 0-02 5 - 7 .
9 . R . Sterner, Flora der I n sel Gland. D i e Areale d e r Ge­ 26. H . Osvald, Notes on the vegetation o f British and Irish
fasspflanzen Glands nebst Bemerkungen zu ihrer mosses. 1 94 9 . 1 2 :- ( 8 :-). ISBN 9 1 - 7 2 1 0-026-5 .
Oekologie und Soziologie. 1 9 3 8 . 2 1 :- ( 1 4 :-). I S B N 27. S. Selander, Floristic phytogeography of South­
9 1 - 72 1 0-009- 5 . Western Lule Lappmark (Swedish Lapland) I. 1 9 50.
1 0. B . L indquist, D al by Soderskog. E n sklmsk lovskog i 2 1 : - ( 1 4 :-). ISBN 9 1 - 7 2 1 0-02 7 - 3 .
forntid och nutid. (Zusammenf. : Ein Laubwald in 28. S . Selander, Floristic phytogeography o f South­
S chonen in der Vergangenheit und Gegenwart.) 1 9 3 8 . Western Lule Lappmark (Swedish Lapland) 11. Karl­
2 7 :- ( 1 8 :-). I S B N 9 1 - 7 2 1 0-0 1 0-9. vaxtfloran i sydvastra Lule Lappmark. (Summary :
1 1 . N. Stalberg, Lake Vattern. Outlines of its natural Vascular flora.) 1 9 05. 1 8 : - ( 1 2 :-). I S B N 9 1 - 7 2 1 0-
history, especially its vegetation. 1 9 3 9 . 1 2 :- ( 8 :-). 028- 1 .
I S B N 9 1 - 7 2 1 0-0 1 1 - 7. 29. M. Fries, Pollen analytiska vittnesbord o m senkvartar
1 2 . G.E. Du Rietz, A . G. Hannerz, G. Lohammar, R. vegetationsutveckling, sarskilt skogshi storia, i nord­
. Santesson & M. Wcern, Zur Kenntnis der Vegetation vastra Gotal and. (Z usamm enf. : Pollenanalytische
des Sees Takern. 1 9 3 9 . 1 2 :- ( 8 :-). I S B N 9 1 - 7 2 1 0- Zeugnisse der spatquartaren Vegetationsentwicklung,
0 1 2-5 .. . hauptsachlich der Waldgeschichte, im nordwestlichen
1 3 . Viixtgeografiska studier tilliignade Car! Skoltsberg Gotaland, Sii d schweden.) 1 9 5 1 . 2 1 :- ( 1 4 :-). I S B N
pa sextioarsdagen 1 / 1 2 1 940. ( Geobotanical studies 9 1 - 7 2 1 0-02 9-X.
dedicated to C. Skottsberg.) 1 9 40. 3 6 : - (24 :-). I S B N 30. M. Wcern, Rocky-shore algae in the Gregrund Archi­
9 1 - 7 2 1 0-0 1 3 - 3 . pelago. 1 9 5 2 . 3 3 : - (22 :-). I S B N 9 1 - 72 1 0-030- 3 .
1 4. N . Hylander, De svenska formerna a v Mentha gentilis 3 1. 0. R une, Plant life on serpentines a n d rel �ted rocks in
L. coli. (Zusammenf. : D ie schwedischen Formen der the North of Sweden . 1 9 5 3 . 1 8 :- ( 1 2 :-). I S B N 9 1 -
Mentha gentilis L . sensu cot!.) 1 9 4 1 . 1 2 : - ( 8 :-). I SB N 7 2 I 0-0 3 1 - I .
9 1 - 7 2 1 0-0 1 4- 1 . 3 2. P . Kaaret, Was servegetation d e r Seen Orlangen und
1 5 . T. E. Hasselrot, Till kannedomen om nagra nordiska Trehorni ngen . 1 9 5 3 . 1 2 :- ( 8 :-). I S B N 9 1 - 7 2 1 0-03 2-
umbilicariaceers utbredning. (Zusammenf. : Zur X.
Kenntnis der Verbreitung einiger U mbilicariaceen in 3 3 . T. E. Hasselrot, Nordliga lavar i Syd- och Mellan­
Fennoscandia.) 1 9 4 1 . 1 2 :- ( 8 :-). I S B N 9 1 - 72 1 0- sverige. (Nordliche Flechten in Sii d - und M ittelschwe­
0 1 5-X. den . ) 1 9 5 3 . 2 2 :- ( 1 4 : -). I S B N 9 1 - 7 2 1 0-03 3-8.
· 1 6. G . Samuelsson, D i e Verbreitung d e r Alchemilla-Arten 34. H. Sjdrs, Slatterangar i Grangiirde fi nnmark. (Sum­
aus der Vulgaris-Gruppe in N ordeuropa. 1 9 4 3 . 1 8 :­ mary : Meadows in Grangarde Finnmark, S W . D alar­
( 1 2 :-). ISBN 9 1 - 7 2 1 0-0 1 6-8. n a, Sweden . ) 1 9 5 4 . 1 8 :- ( 1 2 : -). I S B N 9 1 - 7 2 1 0-034-
I 7 . Th. A rwidsson, Studien iiber d i e Gefasspflanzen in 6.
den Hochgebirgen der Pite Lappmark. 1 9 4 3 . 2 7 :­ 35. S. Ki/ander, Karlvaxternas ovre gran ser pa fj all i syd­
( 1 8 :-). I S B N 9 1 - 7 2 1 0-0 1 7-6. vastra J amtl and samt an grans an de delar av H arjeda­
1 8 . N. Dahlbeck, Strandwiesen am siidostlichen Gre­ len och Norge. [ S u m m ary : Upper limits of vascular
sund. (Summary : Salt marshes on the S . E. coast of plants on mountains in Southwestern J a mtland and
Gresund.) 1 9 4 5 . 1 8 : - ( 1 2 :-). I S B N 9 1 - 7 2 1 0-0 1 8-4 . . adja cent parts of Harjed alen (Sweden) and N orway . ]
1 9 . E . von Krusenstjerna, Bladmossvegetation och blad- 1 9 5 5 . 2 2 : - ( 1 4 :-). I S B N 9 1 - 7 2 1 0-03 5 -4.
Distributor: Svenska Viixtgeografis k a Siillskapet, Box 5 5 9, S- 7 5 1 22 Uppsala, S weden

36. N. Quennerstedt, D i atomeern a i Limgans sjovegeta­ 48. E. Sjogren, Epilithische und eJ?igiiische Moosvegetati­
tion. ( S u m m ary : Diatoms i n the lake vegetation of the on in Laubwiildern der Insel Oland, Schweden. (Sum­
Langan drainage area, J iimtland, Sweden.) 1 9 5 5 . 2 2 :- · m ary : Epilithic and epigeic m o s s vegetation i n
( 1 4 :-). I S B N 9 1 - 7 2 1 0-03 6-2. deciduo us woods on t h e island of Gland, Sweden.)
3 7 . M.-B. Florin, Plankton of fresh and brackish waters in 1 9 64. 40:- (27 :-). I S B N 9 1 - 7 2 1 0-048-6 ( I S B N 9 1 -
the Sodertiilje area. 1 9 5 7 . 20:- ( 1 4 :-). ISBN 9 1 - 72 1 0-448- 1 ).
7 2 1 0-03 7-0: 49. 0. Hedberg, Features of afroalpine plant ecology.
3 8 . M.-B. Flo rin, l n sjostudier i Mellansverige. M ik ro­ Resume franc;ais. 1 9 64. 3 6 :- (24 : - ). I S B N 9 1 -7 2 1 0-
vegetation och pollenregn i vikar av Gstersjobiickenet 04 9 -4 (IS B N 9 1 - 7 2. 1 0-449-X).
och insjoar fran preboreal tid till nutid. (Summ ary : 50. The Pla n t Cover of S weden. A study dedicated to G.
Lake studies in Central Sweden. Microvegetation and Einar D u Rietz on his 70th birthday by his pupils.
pollen rain in inlets of the B altic basin and in lakes 1 9 6 5 . 7 2 : - (4 8 :- ). I S B N 9 1 - 7 2 1 0-050-8.
from Preboreal time to the present d ay.) 1 9 5 7 . 1 0 :­ 5 I. T. Flensburg, Desmids and other benthic algae of L ake
(6 :-). I S B N 9 1 - 7 2 1 0-03 8-9. Kiivsjon and Store Mo sse, SW S weden. 1 9 6 7 . 3 9 :­
3 9 . M. Fries, Vegetationsutveckling och odlingshistoria i (26 :-). I S B N 9 1 - 7 2 1 0-05 1 -6 ( I S B N 9 1 - 7 2 1 0-45 1 - 1 ).
V arnhem strakten. En pollenan alytisk undersokning i
5 2 . E. Skye, Lichens and air pollution. A study of crypto­
gamic epiphytes and environment i n the Stockholm
V iistergotl and. [Z usammenf. : V egetationsentwick ­
lung und Siedlungsgeschichte im Gebiet von V arnhem . region. 1 9 68. 4 2 : - ( 2 8 : -). I S B N 9 1 - 72 1 0-05 2-4
Eine pollen analytische Untersuchung aus V ii stergot­ ( I S B N 9 1 - 72 1 0-45 2-X).
land (Slidschweden).] 1 9 5 8 . 1 6 :- ( 1 0 : -). I S B N 9 1 - 5 3 . Jim Lundqvist, Plant cover and environment of steep
7 2 1 0-03 9 - 7 . hill- sides in Pite Lappm ark. (Resume : La couverture
4 0 . Bengt Pettersson, Dynamik o c h konstans i Gotl ands vegeta!e et !'habitat des fl ancs escarpes des collines de
flora och vegetation. (Resume : Dynamik und Kon­ Pite Lappm ark .) 1 9 6 8 . 4 8 :- ( 3 2 : -). I S B N 9 1 - 72 1 0-
stanz in der Flora und Vegetation von Gotland, 0 5 3 - 2 ( I S B N 9 1 - 7 2 1 0-45 3 -8).
S chweden .) 1 9 5 8 . 40:- ( 2 7 :-). I S B N 9 1 - 7 2 1 0-040-0. 5 4. Conservation of Vegetatioil in Africa South of the
4 1 . E. Uggla, Skogsbrandfiilt i M uddus n ationalpark. Sahara. Proc. of sixth plen. meeting of AETF AT. Ed.
( S u m m ary : Forest fire areas i n M uddus N ation al by Inga and Olov Hedberg. 1 96 8 . 60:- ( 40:-) , cloth
Park, Northern Sweden.) 1 9 5 8 . 2 1 :- ( 1 4 :-). I S B N 7 0 : - (47 :-). I S B N 9 1 - 7 2 1 0-05 4-0 ( I S B N 9 1 - 7 2 1 0-
9 1 - 72 1 0-04 1 - 9. 454-6).
42. K. Thomasson, N ahuel H u api. Plankton of some lakes 55. L . -K. Ko1�(gsson, The Holocene hi story of the Great
in an Argentine N ational P ark, with notes on terrestri­ Alvar of Ol and. 1 9 6 8 . 5 4 :- ( 3 6 :-). I S B N 9 1 - 7 2 1 0-
al vegetation. 1 9 5 9 . 2 1 :- ( 1 4 : -). I SB N 9 1 - 7 2 1 0-042- 05 5 - 9 ( I S B N 9 1 - 7 2 1 0-45 5 -4).
7. 56. H.P. Hallberg, Vegetation auf den Schalenablage­
rungen in Bohusliin, Schweden . ( S u m m ary : Vegetati­
4 3 . V. Gillner, Vegetations- und Standortsuntersuchungen
on on shell deposits in B ohusliin, Sweden.) 1 9 7 1 . 4 8 :­
in den Strandwiesen der schwedischen Westkliste.
(3 2 : - ). I S B N 9 1 - 7 2 1 0-05 6 - 7 (ISBN 9 1 - 7 2 1 0-45 6-2).
1 9 60. 3 3 :- (22 :-). I S B N 9 1 - 7 2 1 0-04 3 - 5 .
5 7 . S. Fransson, Myrvegetation i sydviistra V iir mland.
4 4 . E . Sjogren, Epiphytische Moosvegetation in Laub­
(Sum m ary : Mire vegetation i n south-western V iirm­
wiildern der I n sel Gl and, Schweden. (Summ ary :
land, Sweden .) 1 9 72. 4 2 :- ( 2 8 :-). ISBN 9 1 - 7 2 1 0-
Epiphytic moss communities in deciduous woods on
05 7-5 ( I S B N 9 1 - 72 1 0-45 7-0).
the i sland of Gland, S weden. ) 1 9 6 1 . 24 : - ( 1 6 : -).
I S B N 9 1 - 72 1 0-044-3 (ISB N 9 1 - 7 2 1 0-444-9). 58. G. Wa!lin, Lovskogsvegetation i Sjuhiiradsbygden.
45. G. Wistrand, Studier i Pite Lappmarks kiirlviixtflora, ( S u m m ary : Deciduous woodlands in Sjuhiiradsbyg­
den, V iistergotland, south-western Sweden.) 1 9 7 3 .
med siirskild hiinsyn till skogsl andet och de isolerade
42 :--(2 8 : -) I S B N 9 1 - 72 1 0-05 8 - 3 ( I S B N 9 1 - 7 2 1 0-.
fj iillen. (Z usammenf. : Studien U ber die Gefiisspflan­
zenflora der Pite Lappmark mit besonderer Berlick ­ 45 8-9).
sichti gung d e s Waldl andes u n d der i solierten niederen 5 9. D . Johansson, Ecology of vascular epiphytes in West
Fjel de.) 1 9 6 2 . 3 6 : - (24 :- ). I S B N 9 1 - 7 2 1 0-04 5 - 1 African rain forest. (Resume : Ecologie des epiphytes
( I S B N 9 1 - 72 1 0-4 4 5 - 7). vascul aires dans la foret dense h um ide d'Afrique oc­
46. R. Ivarsson, Lovvegetationen i Mollosunds sock en. cidentale.) 1 9 74. 42 :-. I S B N 9 1 - 7 2 1 0-05 9 - 1 ( IS B N
(Zusammenf. : Die Laubvegetation i m K irch spiel 9 1 - 7 2 1 0-4 5 9-7).
Mollosund, Bo husliin, Schweden.) 1 9 6 2. 3 0 : - (20:- ). Limited n umbers of cloth-bound copies of Acta 44, 45,
I S B N 9 1 - 7 2 1 0-046-X (ISBN 9 1 - 72 1 0-446-5).
46 , 4 8 , 49, 5 1 , 52, 5 3 , 5 5 , 5 6 , 5 7, 58, 59 are available
4 7. K. Thomasson, Araucanian Lakes. Plankton studtes in
through the Society at an additional cost of 1 0 :- per
North Patagonia, with notes on terrestrial vegetation.
1 9 6 3 . 3 6 : - ( 24 : - ) . I S B N 9 1 - 7 2 1 0-04 7-8. copy. I S B N nos. in brackets refer to cloth-bound copies.

V A XT E K O LO G I S K A STUDI ER

1 . S. Brakenhielm & T. fngeldg, Vegetationen i K ungs­ I S B N 9 1 - 72 1 0-802-9.


hamn - M o rga naturreservat med forslag till skotselplan. 3 . H. Sjors och medarb., Skyddsvarda m yrar i K oppar­
(Summ ary : Vegetation and proposed m an agement in bergs !an. (Sum mary : Mires considered for protection in
the K ungshamn- Morga N ature R eserve sout� of Upp­ K opparberg C o unty (Prov. Dal arna , C entral Sweden. )
sala.) 1 9 7 2 . 1 4 :-. ISBN 9 1 - 72 1 0-80 1 -0. 1 9 7 3 . 1 5 : -. I S B N 9 1 - 7 2 1 0-803 - 7 .
2. T. Ingelog & M. R isling, K ronparken vid Uppsala, 4. L . Karlsson, Autecology of cliff a n d scree plants i n
hi stori k och bestandsanalys av en 3 00-arig tall skog. Sarek N ational Park, northern Sweden. 1 9 7 3 . 1 5 :- .
(Sum mary : K ronparken, hi story and analysis of a 3 00-
I S B N 9 1 - 7 2 1 0-804 - 5 .
year-old pinewood near Uppsala, Sweden.) 1 9 7 3 . 1 5 :- .
S VEN SKA V AXT GE O GRAFI SKA SALLS KAPET
S O C I ETAS P H Y T O G E O G RA P H I C A S U E C A N A
A dress: Vaxtbiologiska institutionen, B o x 5 5 9, S-75 1 2 2 Uppsala, Sweden

Sallskapet bar till andamru att vacka och underhrula The object of the Society is to promote investigation in
intresse for vaxtgeografien i vidstracktaste mening, att flora and vegetation, . their history and their ecological
framja utforskandet av flora och vegetation i Sverige och background. Through publication of monographs, and
andra lander och att havda geobotanikens praktiska och other activities, the Society tries to stimulate geobotanical
vetenskapliga betydelse. research and its application to practical and scientific
Sallskapet anordnar sammankomster och exkursioner problems. Membership is open to all who have a personal
samt utger tva publikationsserier. Medlemskap kan erhal­ interest in the advancement of phytogeography.
las efter anmalan hos sekreteraren. Foreningar, bibliotek, Individu� members and subscribers (societies, insti­
laroanstalter och andra institutioner kan inga som abon­ tutes, libraries, etc.) receive the A cta Phytogeographica
nenter. Arsavgift 24 kr ( 1 8 : - for studerande); standigt Suecica for annual dues of 24 Skr plus postage. There are·
medlemskap 400 kr. additional fees in years when more than one volume are
Sa.IIskapet utger arligen Acta Phytogeographica Sueci­ issued. Permanent membership. 400 Skr. For membership
ca. Medlemmar och abonnenter erhaller arets Acta mot ple�se apply to the Secretary.
postfOrskott pa arsavgiften jamte porta och expeditions­ B ack volumes may be obtained by members and
kostnader. De kan ocksa forvarva aldre publikationer tili subscribers at reduced cost (see list of publications, prices
nedsatt pris (parentespriset i Acta-listan). Vissa ar utges in brackets).
extraband av Acta, som erhalls mot en tillaggsavgift. The Society also issues Viixtekologiska studier, which
Sallskapet utger ocksa den ickeperiodiska serien Vaxt­ appear irregularly and are available upon request or
ekologiska studier. Den kan f6rvarvas efter bestallning standing order.
eller genom staende abonnemang hos Sallskapet. Both series can be received by exchange for other
Bada serierna kan ocksa erhrulas i byte mot andra scientific publications. Please apply to the Institute Library
publikationer efter hanvandelse till Vaxtbiologiska institu­ (address as above).
tionens bibliotek.

ISBN 9 1 -7210-059- 1

You might also like