Carpology of Orchids

2006·3 MB·

by  Rasmussen, Finn N

Selbyana 27(1): 44-53. 2006.
CARPOLOGY OF ORCHIDS
FINN N. RASMUSSEN* AND Bo JOHANSEN
Institute of Biology, University of Copenhagen, Gothersgade 140, DK-1123 Copenhagen K,
Denmark. Email: [email protected]
ABSTRACT. Orchid fruits have been largely neglected in orchid research. More variable than generally
imagined, orchid fruits have a basic structure that remains inadequately described and understood. Yet
structure and function of orchid fruits are important for many aspects of orchid conservation and may
contribute significantly to the understanding of orchid phylogeny. Carpological studies of orchids are part
of the ongoing research of the monocot phylogeny research group at the University of Copenhagen. A
model of the orchidaceous ovary explaining the pattern of six valves is presented here, and the "subpla
cental cushions" in the ovary are shown to be cords of pollen tubes.
Key words: Orchidaceae, carpels, carpology, endocarpic trichomes, fruits, intra-ovarian trichomes
INTRODUCTION for examination in transmitted light microscopy.
Fruits of Dendrobium thyrsiflorum Rchb.f. ex
It is because of the flowers that we find or Andre (live plants collected by G. Seidenfaden,
chids interesting in the first place. The immense GT 8338 X GT 8500, Thailand; see Johansen
variation seen in orchid flowers and pollination 1990) were cross-pollinated and collected 2, 4,
mechanisms captivates the mind of those inter 6, and 8 weeks after pollination. The developing
ested in natural history and is hence the ultimate fruits were fixed, embedded, and stained as de
reason for the wish to conserve orchids. scribed above. Fruits of cross-pollinated flowers
The next stage, the fruit, has been much ne of D. secundum Lindl. (Live plants collected by
glected in orchid research. Orchid fruits, how G. Seidenfaden, GT 9454 X GT 9248, Thailand;
ever, are much more variable than generally see Johansen 1990) were collected 2, 4, 6,8, and
imagined, and their basic structure is still not 12 weeks after pollination, fixed in a mixture of
completely clarified, despite the obvious fact PFAIGA (2 % paraformaldehyde and 2% glu
that structure and function of orchid fruits are taraldehyde in 0.1 M phosphate buffer), dehy
no less important than flowers. drated, and critical point dried for SEM accord
The primary reason for this state of affairs is, ing to standard procedures.
of course, that we recognize species by the flow Transmitted light microscopy images were
ers. Most orchids collected in nature have been digitized directly from the Leica DM RXA 2
flowering specimens, and non-flowering speci microscope using a Hamamatsu C-2400 (ORCA
mens collected for cultivation rarely produce II) camera. SEM observations were made using
fruits in captivity. Orchid taxonomy and identi a JEOL ISM P15 Scanning Electron Micro
fication has become a study of flowers. Even scope.
very striking fruit features are ignored, because
they are neither present in the herbarium nor in
the greenhouse; and when seen in nature, they RESULTS: STRUCTURE OF ORCIDD FRUITS
cannot be identified.
Orchids belong to the lily-like monocots in
the broad sense and may constitute the sister
MATERIALS AND METHODS group to the remainder of the clade that has be
come known as Asparagales (The designation
Ovaries of a large number of orchids flower
"Orchidales" has priority, but the principle of
ing or fruiting after artificial pollination in the
priority is not mandatory for ordinal names).
Botanical Garden, University of Copenhagen,
The flowers in this group are generally 3-mer
were hand sectioned and inspected in stereomi
ous. In Orchidaceae, the flowers are epigynous,
croscope and with transmitted light microscopy
and the ovary is thus inferior to perianth-seg
at low magnification. Ovaries of ca. 50 orchid
ments and stamens. Epigyny in lily-like plants
species collected at anthesis in the greenhouses
usually is regarded as being a result of congen
or in nature were fixed in FAA and embedded
ital fusion of the basal parts of the perianth and
in Paraplast@. Then 10 fLm thick sections were
the stamens, but also may be imagined as a re
deparafinized and stained in SafraniniFast Green
sult of a cup-shaped receptacle that envelops the
ovary-a view advanced by Pfitzer (1886) and
* Corresponding author. accepted by Dressler (1981, 1993).
44
RASMUSSEN & JOHANSEN: ORCHID FRUITS 45
One would expect orchid fruits to be con scribed a cross section of a fruit of Vanilla plan
structed like the fruits seen in Arnaryllidaceae, ifolia as showing three carpels with laminal pla
Hypoxidaceae, and other families with inferior centation along their midnerves, which would
capsules made of three carpels, but they are re place the carpels opposite the petals and the in
ally much different and have puzzled botanists ner stamen whorl and bring them out of align
since Robert Brown and Lindley. ment with the stigma lobes.
Halle (1977, 1980, 1981, 1986), the most im
How Many Carpels? portant recent contributor to orchid carpology,
regarded the carpel number and ovary structure
FIGURE 1 shows a cross section of a typical as still open for interpretation. He pointed out
orchid fruit as viewed by Francis Bauer in 1801. that the distinctive pattern with six "valves" or
This and other beautiful drawings of orchid "elements" is still in need of an explanation. We
flower details were published in Bauer & Lind offer one here.
ley (1830-1838). The ovary is composed of six
distinct parts, unlike any other fruits we know Split-Carpel Model
from the monocotyledons; and at maturity, the
capsule of most species will open along dehis The orchidaceous ovary can be described
cence lines separating these six elements. Lind within the framework of the standard liliiflorous
ley assumed that the ovary was made up of six gynoecium. No extranumerary carpels, recepta
carpels, three fertile ones carrying luminal V cular cups, shifted whorls, or unusual ovulifer
shaped placentae and three narrower, sterile car ous structures are needed, if this ovary is rec
pels, carrying the stigmata (Bauer & Lindley ognized as inferior and the tepal bases as in
1830-1838; Lindley 1847). Robert Brown volved in the structure. After studying a great
(1831) preferred a less complicated interpreta number of orchid ovaries, we suggest that the
tion: orchids have only three carpels that alter hexamerous pattern seen in most orchid fruits is
nate with the inner whorl of stamens. This in produced by the bases of the sepals and petals,
terpretation places the midline of each carpel di which are less amalgamated into the inferior
rectly below a stigma lobe, and the placentae are ovary than in, for example, Amaryllidaceae
marginal, as in comparable families. This view (Rasmussen et al. in press). The "sterile valve"
was accepted by later authorities such as Payer ("jugum" in Venverloo et al. 2003) corresponds
(1857), van Tieghem (1875), and Pfitzer (1886), to a sepal base and may in many cases appear
but others regarded the question as unresolved to split the carpel opposite to it. In some cases,
(Prillieux 1857, Eichler 1875, and Arber 1925 however, it is evidently an element adnate to the
who ascribed Lindley's 6-carpel model to Saun carpel, sometimes even with a separate vascular
ders 1923; see FIGURE 2). strand (FIGURES 3, 4). The broader "fertile
Garay (1960) suggested that uni-locular or valve" is then made up of a petal base and two
chid ovaries comprise just one carpel, a view carpel-halves, carrying one marginal placenta
apparently adopted by Arditti (1992). Barbosa from each (FIGURE 5). In most species, the carpel
Rodrigues is cited by Halle (1980) for having walls thus will split along the midline at matu
advanced a model with no less than 12 carpels rity, which is indeed the most common mode of
in 1883. Osterberg (1883) and Swamy (1948), dehiscence in capsule-fruited liliiflorous mono
unaware of Osterberg'S work, tried to elucidate cots; but the sepal base will most often remain
the problem by studies of vascularization (FIG as a separate "sterile valve." All six "valves"
URE 4), but inferences from this approach are will usually remain fused near the apex of the
limited by the fact that vascular bundles are very capsule, rendering the characteristic lantern-like
plastic. Anastomoses and new side branches impression of a dehisced orchid capsule. The
may arise as a function of distance to other bun stamen bases are integrated to the extent that no
dles and depth of the tissue (Schmid 1972). trace is left in the inferior ovary, which is to be
Although prominent orchidologists (Schlech expected considering their strong integration in
ter 1927, Dressler 1981) accepted Brown's mod the gynostemium.
el, dissent continues to be expressed. A very dif This interpretation of the orchidaceous ovary,
ferent view was published by Vermeulen (1966), in accordance with general monocot floral mor
who suggested that the three placentiferous el phology, is compatible with all phenomena we
ements might not be carpels at all but ovule have seen so far in orchid ovaries and fruits.
bearing stem-like organs. He even introduced a
special term, "stachyospermy," for this way of Observations of Variation
producing a fruit. Veyret (1981, 1988) argued
for the 6-carpel view after analysing fruits of Despite this simple ground plan, orchid fruits
Sobralia and Elleanthus. Odoux et al. (2003) de- are extremely variable in size, shape, texture, or-
46 SELBYANA Volume 27(1) 2006
1 2 Old vlew:One.$arpe~1Tl8r&,nal ovules
Tnid.ib. . .... "~>< stlgma.above
.. ' . -, . . '. '. !'lidTib
~I
;>Q) Q;~::O
", _, .... ,-' c""1:5j J<lII
m\drlh·-'·', --', .' -'
SaundetSS "'-ew: OnE"taTpel.ovules marginal.
but tatpel seml'solld.~ placenta! appIOIIcn mIdrib
PLATE A. Cross sections and interpretations of orchid ovaries. Flowers are shown as resupinated with la
bellum lowermost in all sections. FIGURE 1. Bletia purpurea (Lam.) DC, three sterile valves alternate with three
placentiferous valves (Bauer 1801 in Bauer & Lindley 1830: pI. IX as B. verecunda). Tissue corresponding to
the median carpel and bases of adjoining lateral petals is shadowed here. Scale bar = 1 mm. FIGURE 2. Two
interpretations of the orchidaceous ovary rendered by Arber (1925: fig. eli): "Old view" (Brown 1831) and
"Saunders's view" (Lindley 1830). FIGURE 3. Gavilea glandulifera (Poepp.) M.N.Correa. The sterile valve (top)
is seen on the outer side of the carpel, the vascular strands are not fused. Scale bar = 500 fJ-. (Collected by
RASMUSSEN & JOHANSEN: ORCHID FRUITS 47
namentation, way of dehiscence, number of and began in the last part of the 20th century.
seeds, and seed dispersal mechanisms. Prillieux The orchid paintings are treated in Schweinfurth
(1857) recognized seven different dehiscence et al. (1963, 1969) and in Fernandez Perez
types and observed that particular modes of de (1985); three more volumes are planned. The
hiscence may characterize genera. The criteria fruits, however, are not as detailed as those of
were number of valves (3 + 3, 3, 2, or 1) and Beer (1863), and the text provides little discus
apically fused vs. free and recurving valves. sion of fruit morphology.
Prillieux did not distinguish the case where the Orchid fruit observations have been rare and
three placentiferous valves recurve, as the sterile scattered for a long time; only the works on New
ones remain apically conjoined; and he was not Caledonian orchids by Halle (1977, 1980, 1981,
aware that some orchid fruits do not dehisce at 1986) contain numerous illustrations of orchid
all. fruit details and penetrating discussions of their
The most beautiful contribution to orchid car structure and biology. Silhouettes of fruit cross
pology was published by Beer in 1863. The folio sections ("carpograms") were used in Halle
work "Beitdige zur Morphologie und Biologie (1981) together with formula-like data lines for
der Familie der Orchideen" contains 12 litho presenting a large amount of fruit data. Reeve
graphic color plates of orchid seeds and fruits, and Woods (1989) included cross sections of the
depicting 238 different fruits of mainly neotrop often spectacularly winged fruits in most of their
ical orchids along with descriptions of fruit de drawings of species of Dendrobium sect. Oxy
velopment observed in culture. Beer even sug glossum. Venverloo et al. (2003) showed cross
gested a classification of the orchid family into sections of several species in a paper on post
six main groups characterized by a combination pollination phenomena in Phaiaenopsis spp.
of floral and fruit characters. This scheme, very Often overlooked are the non-dehiscent, more
much at variance with established classifica or less fleshy fruits of some orchids. Examples
tions, was cOlllpletely ignored by later authors. are found in Apostasia, Selenipedium, Cyrtosia.
Le Maout and Decaisne (1868) illustrated the Vanilla, and Palmorchis (Dressler 1981, 1993;
dehiscence modes described by Prillieux (1857). Kocyan & Endress 2001). In Neuwiedia, some
Their drawings were included in Pfitzer (1882) species typically have 6-valvate capsules with
and in his treatment of Orchidaceae in Die Na caducous fertile valves, but others have genuine,
turlichen Pflanzenfamilien (Pfitzer 1889). gooseberry-like berries (K.A. Kristiansen pers.
Horowitz (1901) undertook a detailed ana comm., Kocyan & Endress 2001, Rasmussen et
tomical study (supervised by Pfitzer) of 38 or al. in press).
chid fruits. He concluded that the main differ
ence between the sterile and the fertile valves is
that the cells of the former do not get much larg Endocarpic Trichomes
er during maturation, while the cells of the fer
tile valves undergo considerable lateral growth. Hairline structures inside the ovary of orchids
He also suggested a classification of orchid fruits seem to have been first recognized by du Petit
based on number and distribution of vascular Thouars (1822). They have been observed and
bundles, and recognized six different ways of discussed in some detail by Blume (1848), Beer
dehiscence. (1857), Prillieux (1857), Horowitz (1901), Mal
Fruits of a number of Colombian orchids are guth (1901), and Halle (1986), but often with
present on several of the more than 200 beautiful incomplete references to other treatments. Halle,
orchid plates painted from 1783-1816 for a for example, was unfamiliar with the compre
grand folio report in the Real Expedicion Botan hensive work of Malguth. In a recent survey of
ica. Publication of the report was much delayed intra-ovarian trichomes in monocotyledons (Ru-
N. Jacobsen 1979-3085, Argentina, Prov. Rio Negro). FIGURE 4. Phaius tankervilleae (Aiton) Blume, "Vasco
gram" by Osterberg (1883: fig. 34 as P. grandifolius). Median strands from the carpels continue into the style;
strands of the sterile valves become midnerves of the sepals, with the one from the median sterile valve branching
and also supplying the anther. FIGURE 5. The "split carpel model" of the orchidaceous ovary. Left: Ovary with
incomplete splitting of the carpel, as in FIGURE 3. Right: Typical orchid ovary with three sterile valves (sepal
bases) and three fertile valves, each consisting of two half-carpels and a petal base. FIGURE 6. Endocarpic
trichomes emanate from the midline (eventually sutures) of the carpels of Microperafasciculata (Lindley) Garay
(top) and Trachoma subluteum (Rupp) Garay (middle, bottom). "Subplacental cushions" (coussinet sous-pla
centaires) are seen next to the placentae (sc) in T. subluteum (details from Halle 1986: pI. 9, © Publications
Scientifiques du Museum national d'Histoire naturelle, Paris).
48 SELBYANA Volume 27(1) 2006
RASMUSSEN & JOHANSEN: ORCHID FRUITS 49
dall et al. 1998), orchids were not mentioned at veloping ovary or even for transport of nutrition,
all. as they in some cases have numerous dots that
Beer (1857) had already noted that endocarpic he interpreted as pores. He also suggested they
trichomes seemed to be found only in epiphytic might provide the mature seeds with glutinous
genera. Halle (1986) found that they are a char material enabling them to stick on branches,
acteristic of the subtribe Sarcanthinae (=Aeri which would be an interesting alternative to the
dinae). Freudenstein and Rasmussen (1999) seed-stickiness described by Barthlott and Zie
scored them as present in 41 genera, including gler (1980).
all "Aeridinae," in most of the genera formerly Halle (1986) also observed and depicted pe
classified as "vandoids" (Vandoideae in Dress culiar, seemingly inexplicable intraovarian struc
ler 1981), and more surprisingly in the terrestrial tures that he called "coussinets sous-placentai
orchidoid genus Prasophyllum. re," subplacental cushions, in the comers of the
Considerable variation has been observed in placentae (FIGURE 6). Viewed in three dimen
appearance and texture of the endocarpic tri sions over an interval of time they are, as we
chomes. They may be up to 4 cm long, curled, shall see, natural transient phenomena in con
and hose-like, sometimes cylindrical (Halle nection with fertilization.
1986), or laterally conjoined forming mat-like
structures (Malguth in Horowitz 1901). In some
DISCUSSION: WHAT GOES ON IN AN
fruits, they are developed as stiff brushes that
ORCHID OVARY?
remain on the margins of one set of valves after
dehiscence (Malguth 1901, Halle 1989), perhaps The awesome number of ovules and later
acting as sieves that bring a bryophyte peristome seeds in orchid ovaries has long fascinated bi
to mind rather than elaters. We suspect that the ologists. Darwin (1862) estimated the number of
resiniferous papillae seen in the fruits of Vanilla seeds in Cephalanthera damasonium (as C.
(Berg 1865, Roux 1954, Odoux et al. 2003) may grandiflora) and Orchis maculata at 6020 and
be homologous structures. 6200, respectively, describing in detail how he
Knowledge remains incomplete regarding the arrived at his number. He also related a count of
distribution, homologies, range of variation, and 371,250 seeds in a Gongora sp. (as Acropera).
structural details of endocarpic trichomes; and Arditti and Ghani (2000) cited counts of 4 mil
we continue to observe and collect data for a lion seeds in Cycnoches, but such extremes are
more comprehensive survey. To date, no obser in fact rare. The 1.5-2 cm long fruit of Dendro
vational data have elucidated their ecological bium secundum contains ca. 15,000 seeds. Even
significance. Prillieux (1857) and particularly small capsules will often hold a few thousand
Halle (1986) emphasized the resemblance to the seeds, which is still far more than in non-orchi
elaters seen in the sporangia of liverworts and daceous plants. As for many other unique orchid
some pteridophytes, and concluded that their characters, this is viewed as a consequence of
function is most probably also similar. This the mycotrophic germination (Benzing 1987,
seems plausible in most cases, where they are Rasmussen 1995). The chance of germination
strongly hygroscopic one-celled trichomes for anyone seed is very low, so a huge surplus
emerging from the lines of dehiscence in the of very mobile seeds is necessary simply for
capsules (near the median nerve of the carpels), population maintenance. On the other hand, my
eventually becoming detached, and sometimes cotrophy dispenses largely with the need for
taking up more volume than the seeds. Malguth stored nutrition, allowing seeds to become very
(1901) speculated that they also might function small but still capable of germination. An ex
as regulators of the environment inside the de- tremely high number of ovules per fruit neces-
PLATE B. Stages in the development of Dendrobium placentae, pollen tubes, and ovules. FIGURE 7. Schematic
placentiferous and sterile valve in an orchid ovary. Note the wavy margin (placenta) of the carpel and the sutural
papillae where the carpel will split at dehiscence of the fruit. (Halle 1977: pI. 2, fig. 6, © Publications Scienti
fiques du Museum national d'Histoire natureUe, Paris). FIGURE 8. Dendrobium thyrsiflorum Rchb.f. ex Andre 2
weeks after pollination. Two sets of placental lobes (pi) almost fill the lumen of the ovary; "subplacental
cushions" (sc) are dense masses of pollen tubes very distinct at the bases of the lobes. Scale bar = 500 fLm.
FIGURE 9. Dendrobium thyrsiflorum 8 weeks after fertilization. The young ovules fill most of the lumen of the
ovary. Scale bar = 0.5 mm. FIGURE 10. Dendrobium secundum Lindl. SEM of placenta 8 weeks after pollination.
The pollen tubes have nearly reached the developing ovules. Scale bar = 100 fLm. FIGURES 11, 12. Dendrobium
secundum 12 weeks after pollination, with fertilization imminent. Scale bar = 200 fLm. Insert: detail of ovules
with two integuments clearly seen. Scale bar = 50 fLm.
50 SELBYANA Volume 27 (l) 2006
sitates an equally high number of pollen being ture and dehisce in the next dry season a year
transferred to the stigma, leading to all the well after pollination (Johansen 1990).
known specializations of the orchid flower. The time required for the fruit to ripen is
As in all liliiflorous monocots, the ovules will known from a number of cultivated orchid spe
develop along the margins of the carpels, where cies (45 days to 20 months; see surveys in Wirth
the placentae, which in orchids usually are un & Withner 1959, Lucke 1971, Arditti 1992). In
dulate, optimize the usage of available space in addition, more complete event tables worked out
the fruit (FIGURE 7). Development of the ovary for a limited number of species record the fol
into a fruit and formation of ovules, however, lowing events: fruit age in days after pollination,
will only start if the fruit is pollinated (Brong physical fruit size, and such stages as the begin
niart 1831; Hildebrand 1863; Swamy 1949a, ning of cell divisions in the ovary wall, differ
1949b; Veyret 1965). Pollination events and entiation of archesporial cell, megasporogenesis,
hence fruit set is often low in orchids, especially megagametogenesis, fertilization, seed coat for
in epiphytic orchids (e.g., Ackerman & Montal mation, seed disconnection from placenta, and
vo 1990, Arag6n & Ackerman 2004), but the fruit dehiscence. See comparable surveys in
effect of this seems counterbalanced by the high Veyret (1965), Arditti (1992), and a detailed ta
number of ovules per fruit. For every seed, a ble for Epidendrum ibaguense in Taylor et al.
pollen grain must germinate on the stigma and 1982. Much of the development is initiated and
produce a pollen tube that will grow all the way regulated by the primary pollination event at the
down the stylar canal into the fruit lumen, cross stigmatic surface (O'Neill 1997). The sequence
the placenta, and enter the micropyle of an of these events appears consistent, but the timing
ovule. Observations on several species of Den differs and is probably adapted to the ecological
drobium show that in cross section single tubes conditions at the preferred habitat. Structural and
may be 2-10 p,m in diam. depending on species. developmental constraints are likely to be in
The tubes collapse and flatten behind the sperm volved as well, but very little is known yet. Not
cells, but they nevertheless form a conspicuous surprisingly, almost all information on the on
cord because of their sheer number. Orchid pol togeny of orchid fruits is concerned with ovules
len contains little or no reserve nutrients, but and seeds; little data concerns the ovary walls
pollen tube growth is promoted by mucilage se and placentae; and hardly any data are available
creted by the stigma and stylar canal (Johansen on papillae and trichomes.
1990). After passing through the stylar canal
they will spread and continue along the corners
CONCLUSIONS: THE CARPOLOGY
of the placentae. Viewed in cross section, they
CHALLENGE
form the "subplacental cushions" (FIGURE 8).
When the tube apices with the sperm cells fan The ability to set fruit and successfully dis
out over the placentae, the cushions become less perse seeds is critical for maintaining a viable
prominent and the ovules, having developed in popUlation and indispensable for spreading and
the meantime, begin to appear in cross sections forming new populations and hence for the long
of the fruit (FIGURES 9, 10). Observations on term survival of a species. Much research has
fruit development in Dendrobium farmeri indi focused on pollination; also ecological factors
cate that when the pollen tubes reach their tar such as pollen load, availability of resources,
gets, the ovary has increased in size by a factor fruit predation, and controlling fruit set and re
of ten (Johansen 1990) and is stuffed with production efficacy have been investigated (e.g.,
ovules ready for fertilization (FIGURES 11, 12). Montalvo & Ackerman 1987, Zimmerman &
The size of the fruit also depends on the amount Aide 1989, Snow & Whigham 1989, Ackerman
of pollen applied to the stigma. When 1, lh, lA, & Montalvo 1990, Ackerman & Zimmerman
% pollinarium were used for pollination in Den 1994, Ackerman 1998, Melendez-Ackerman et
drobium secundum, the fruit size decreased with al. 2000). The structure, physiology, and me
decreasing amounts of pollen applied, indicating chanical function of the fruit, however, are still
that the number of pollen cells in the pollinarium very little known. Apparently this part of the
roughly corresponds to the number of ovules de plant is just as specialized and uniquely adapted
veloped, and hence little or no pollen competi as other orchidaceous structures that attract
tion takes place in this species. much more attention. A detailed study of orchid
Time from pollination to fertilization in orchid fruits might offer some genuine surprises.
fruits has been measured from 4 days to 10 The bottleneck for this research is availability
months (Wirth & Withner 1959, Veyret 1965). of study material. Orchid fruits should be stud
In Dendrobium chrysotoxum, development of ied in-situ at accessible sites to investigate not
fruit stops shortly after fertilization (ca. 80 days only incompatibility, fruit set, and timing but
after pollination). The fruit will eventually ma- also dehiscence and release of seeds. For this
RASMUSSEN & JOHANSEN: ORCHID FRUITS 51
research, fruits should be collected and fixed for Arditti, J. 1992. Fundamentals of Orchid Biology. Wi
structural studies. In collections of cultivated or ley, New York.
chids, fruits can be produced and fixed at inter Arditti, J. and A.K.A. Ghani. 2000. Numerical and
physical properties of orchid seeds and their bio
vals to study timing of developmental processes
logical implications. New Phytol. 45: 367-421.
and function of endocarpic structures. As polli
Barbosa Rodrigues, J. 1883. Structure des Orchidees.
nation triggers the development of ovules in
Tipographie nationale, Rio de Janeiro.
most orchids, and as many thousands of ovules
Barthlott, W. and B. Ziegler. 1980. Uber ausziehbare
are developed simultaneously, orchids are ideal helicale Zellwandverdickungen als Haft-Apparat
plants for studying the molecular background of der Samenschalen von Chiloschista lunifera (Or
ovule and gametophyte development in angio chidaceae). Ber. Deutsch. Bot. Ges. 93: 391-403.
sperms. For these reasons, we currently are us Bauer, E and J. Lindley. 1830-1838. Illustrations of
ing various molecular techniques to study Orchidaceous Plants. James Ridgway and Sons,
MADS-box gene expression during ovule de London.
velopment in orchids. Fruit character data (Skip Beer, J.G. 1857. Uber das Vorkommen eines Schleu
derorganes in den Friichten verschiedener Orchi
per et al. 2006) mapped on phylogenies may
deen. Sitzungsber. Kaiser!. Akad. Wiss., Math.
help clarify where and how these attributes
Naturwiss. C!., 24: 23-28 + 3 figs.
evolved. The aerodynamics of seed shedding
---. 1863. Beitrage zur Morphologie und Biologie
and dispersal may be studied experimentally.
der Familie der Orchideen. Carl Gerold's Sohn,
An understanding of the structure and biology Wien.
of orchid fruits may offer crucial insights, ap Benzing, D.H. 1987. Major patterns and processes in
plicable when addressing conservation ques orchid evolution: a critical synthesis. Pp. 34-77
tions. in J. Arditti, ed. Orchid Biology-Reviews and
Perspectives, IV. Cornell University Press, Ithaca,
N.Y.
ACKNOWLEDGMENTS
Berg. O. 1865. Anatomischer Atlas zur pharmazeu
We thank Lis M. Frederiksen and Kate Jensen tischen Waarenkunde. Rudolph Gaertner, Berlin.
for technical assistance. Martin Skipper kindly Brongniart, A.T. 1831. Observations sur Ie mode de
fecondation des Orchidees et des Cistinees. Ann.
provided the photographs for FIGURES 8 and 9.
Sci. Nat. (Paris). 24: 113-130.
N. Halle and The Museum national d'Histoire
Brown, R. 1831. Observations on the Organs and
naturelle, Paris, permitted reprint of details from
Mode of Fecundation in Orchidere and Asclepi
illustrations by Halle (1977, 1986). Material for
adere. Richard Taylor, London.
this research is to a large extent from the live Darwin, C. 1862. On the Various Contrivances by
collections cultivated in the Botanical Garden, which British and Foreign Orchids Are Fertilised
University of Copenhagen. The research on evo by Insects. John Murray, London.
lution and structure of monocots at the Depart Dressler, R.L. 1981. The Orchids-Natural History
ment of Evolutionary Biology, Biological Insti and Classification. Harvard University Press,
tute, University of Copenhagen, is supported by Cambridge, Mass.
the Danish Natural Science Research Council, ---. 1993. Phylogeny and Classification of the Or
chid Family. Dioscorides Press, Portland, Ore.
grants no. 21-02-0361 and 51-00-0271.
du Petit-Thouars, L.M.A.A. 1822. Histoire particuliere
des plantes Orchidees receuillies sur les trois i1es
LITERATURE CITED Australes d' Afrique, de France, de Bourbon et de
Madagascar. A. Bertrand, Treuttel & Wurtz, Paris.
Ackerman, J.D. 1998. Evolutionary potential in or
Eichler, E.A. 1875. Bltithendiagramme. W. Engel
chids: patterns and strategies for conservation.
mann, Leipzig.
Selbyana 19(1): 8-14.
Fernandez Perez, A. 1985. Flora de la Real Expedicion
Ackerman, J.D and A.M. Montalvo. 1990. Short- and
Botanica del Nuevo Reino de Granada, Vol. IX.
long-term limitations to fruit production in a trop
Ediciones Cultura Hispanica, Madrid.
ical orchid. Ecology 7: 263-272.
Freudenstein, J.v. and EN. Rasmussen. 1999. What
Ackerman, J.D. and J.K. Zimmerman. 1994. Bottle
does morphology tell us about orchid relation
necks in the life histories of orchids: resources,
ships? A cladistic analysis. Amer. J. Bot. 86: 225-
pollination, population structure, and seedling es
tablishment. Pp. 125-129 in A.M. Pridgeon, ed. 248.
Proceedings of the 14th World Orchid Confer Garay, L.A. 1960. On the origin of the Orchidaceae.
ence. HMSO, Edinburgh. Bot. Mus. Leafl. 19: 57-95.
Aragon, S. and J.D. Ackerman. 2004. Does flower col Halle, N. 1977. Flore de Nouvelle Caledonie. Museum
or variation matter in deception pollinated Psy National d'Histoire Naturelle, Paris.
chilis monensis (Orchidaceae). Oecologia 138: ---. 1980. Filles de la terre ou filles de l'air? Pp.
405-413. 172-182 in Museum National d'Histoire Naturel
Arber, A. 1925. Monocotyledons, A Morphological Ie. Fernand Nathan, Paris.
Study. Cambridge University Press, Cambridge, ---. 1981. Donnees nouvelles et carpogrammes
U.K. d'orchidees canaques. Adansonia 20: 353-368.
52 SELBYANA Volume 27(1) 2006
---. 1986. Les elateres des Sarcanthinae et addi Rasmussen, H.N. 1995. Terrestrial Orchids-From
tions aux Orchidaceae de la Nouvelle-Caledonie. Seed to Mycotrophic Plant. Cambridge University
Bull. Mus. Natl. Hist. Nat., B, Adansonia 3: 215- Press, Cambridge, U.K
239. Reeve, TM. and P'J.B. Woods. 1989. A revision of
Hildebrand, E 1863. Die Fruchtbildung der Orchideen, Dendrohium section Oxyglossum (Orchidaceae).
ein Beweis flir die doppelte Wirkung des Pollen. Notes Roy. Bot. Gard. Edinburgh 46: 161-305.
Bot. Zeitung (Berlin) 21: 329-333, 337-345. Roux, G. 1954. Etudes morphologiques et histologiqes
Horowitz, A. 1901. Ueber den anatomischen Bau und dans Ie genre Vanilla. Pp. 45-93 in G. Bouriquet,
das Aufspringen der Orchideenfrtichte. Bot. Cen ed. Le Vanillier et 1a Vanille dans Ie Monde. Paul
tralbl.: 487-521 + 3 plates. Lechevalier, Paris.
Johansen, B. 1990. Incompatibility in Dendrobium Rudall, P.J., C.J. Prychid, and C. Jones. 1998. Intra
(Orchidaceae). Bot. J. Linn. Soc. 103: 165-196. ovarian trichomes in monocotyledons. Pp. 219-
Kocyan, A. and P.K Endress. 2001. Floral structure 230 in S.J. Owens and P.J. Rudall, eds. Reproduc
and development and systematic aspects of some tive Biology. Royal Botanic Gardens, Kew.
"lower" Asparagales. PI. Syst. Evol. 229: 187- Saunders, E.R. 1923. A reversionary character in the
216. stock (Matthiola incana) and its significance in
Le Maout, J. and J. Decaisne. 1868. Traite general de regard to the structure and evolution of the gy
Botanique descriptive et analytique. Didot freres, noecium in the Rhoeadales, the Orchidaceae, and
Paris. other families. Ann. Bot. (Oxford) 37: 451-482.
Lindley, J. 1847. The Vegetable Kingdom, 2nd ed. Schlechter, R. 1927. Die Orchideen. Paul PaTey, Ber-
Bradbury and Evans, London. lin. !
Lucke, E. 1971. Samenkeimung mediterraner Ophrys. Schmid, R. 1972. Floral bundle fusion and vascular
Die Orchidee 22: 62-65. conservatism. Taxon 21: 429-446:
MaJguth, R. 1901. "Bio10gische Eigenthtimlichkeiten Schweinfurth, c., A. Fernandez Perez, 4nd L.A. Garay.
der Frtichte epiphytischer Orchideen." Ph.D. diss., 1963. Flora de la Real Expedicion Botanica del
Breslau. Nuevo Reino de Granada, Vol. 'VII. Ediciones
Melendez-Ackerman, E., J.A Ackerman, and J.A. Ro Cultura Hispanica, Madrid.
drfgues-Robles. 2000. Reproduction in an orchid ---. 1969. Flora de la Real Expedicion Botanica
can be resource-limited over its lifetime. Biotro del Nuevo Reino de Granada, Vol. VIII. Ediciones
pica 32: 282-290. Cultura Hispanica, Madrid.
Montalvo, A.M. and J.A Ackerman. 1987. Limitation Skipper, M., L.B. Johansen, KB. Pedersen, S. Fred
to fruit production in Ionopis utriculariodes. Bio eriksen, and B.B. Johansen. 2006. Cloning and
tropic a 19: 24-31. transcription analysis of an AdAMOUS- and
O'Neill, S.D. 1997. Pollination regulation of flower SEEDS TICK ortholog in the orchjd Dendrohium
development. Ann. Rev. PI. Physiol. & PI. Molec. thyrsifiorum (Reichb. f.). Gene 366: 266-274.
BioI. 48: 547-574. Snow, A.A. and D.E Whigham. 1989. :Cost of flower
Odoux, E., J. Ecoute, J-L Verdeil, and J-M. Brillouet. and fmit production in Tipularia discolor (Orchi
2003. Localization of I3-D-g1ucosidase activity and daceae). Ecology 70: 1286-1293. i
glucovanillin in vanilla bean (Vanilla planifolia Swamy, B.G.L. 1948. Vascular anatomy of orchid
Andrews). Ann. Bot. (Oxford) 92: 437-444. flowers. Bot. Mus. Leafl. 13: 61-95.
Osterberg, J.A. 1883. Bidrag till kannedomen af peri ---. 1949a. Embryological studies in the Orchi
carpiets anatomi och kiirlstrangfbrloppet i blom daceae. 1. Gametophytes. Amer. MidI. Nat. 41:
man hos orchideerna. Ofvers. Fl\rh. Kongl. Sven 184-201.
ska Vetensk.-Akad. 3: 47-62 + 3 plates. ---. 1949b. Embryological studies. in the Orchi
Payer, J.B. 1857. Traite d'organogenie vegetale com daceae. II. Embryogeny. Amer. MidI. Nat. 41:
paree 1a fleur. Victor Masson, Paris. 202-232.
Pfitzer, E. 1882. GrundzLige einer vergleichenden Mor Taylor, J.S., S.J. Blackman, and E.c.: Yeung. 1982.
phologie der Orchideen. Carl Winter, Heidelberg. Hormonal and stmctural aspects ot· fruit develop
---. 1886. Morphologishe studien tiber die Orchi ment in the orchid, Epidendrum. J, Exp. Bot. 33:
deenbltithe. Carl Winter, Heidelberg. 495-505.
---. 1889. Orchidaceae. Pp. 52-224 in A. Engler Tieghem, van Ph. 1875. Recherches sur:la structure du
and K Prantl, eds. Die NatUrlichen Pflanzenfam pistil et sur l'anatomie comparee d~ la fleure. Im
ilien. Engelman, Leipzig. primerie Nationale, Paris.
Prillieux, E. 1857. Observations sur la dehiscence du Venverloo, J., A. Schuitemann, and E.E de Vogel.
fmit des Orchidees. Bull. Soc. Bot. France 4: 803- 2003. Phalaenopsis (Orchidaceae-Aeridinae):
809. Fruit morphology and post-pollination phenome
Rasmussen, EN., S. Frederiksen, B. Johansen, L.B. na. J. flir den Orchideenfreund 10: 262-277.
J~rgensen, G. Petersen, and O. Seberg. In Press. Vermeulen, P. 1966. The system of the orchidales. Acta
Fleshy fruits in liliiflorous monocots. In J.T Co Bot. Neerl. 15: 224-253.
lumbus, E.A. Friar, J.M. Porter, L.M. Prince, and Veyret, Y. 1965. Embryogenie comparee et blastogenie
M.G. Simpson, eds. Monocots: Comparative Bi chez les Orchidaceae-Monandrae. O.R.S.TO.M.,
ology and Evolution, 2 vols. Rancho Santa Ana Paris.
Botanic Garden, Claremont, Calif. ---. 1981. Quelques aspects du pistil et de son
RASMUSSEN & JOHANSEN: ORCHID FRUITS 53
devenir chez quelques Sobraliinae (Orchidaceae) development in the Orchidaceae. pp. 155-188 in
de Guyane. Bull. Mus. Natl. Rist. Nat., B, Ad C.L. Withner, ed. The Orchids-A Scientific Sur
ansonia 3: 75-83. vey. Ronald Press, New York.
---. 1988. Les six carpelles des orchid€es. Orchi Zimmerman, J.K. and T.M. Aide. 1989. Patterns of
dophile (Asnieres) 82: 113-117. fruit production in a neotropical orchid. Amer. J.
Wirth, M. and C.L. Withner. 1959. Embryology and Bot. 76: 67-73.