Content uploaded by Lawrence W Zettler
Author content
All content in this area was uploaded by Lawrence W Zettler on May 18, 2020
Content may be subject to copyright.
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
136
THE GHOST ORCHID DEMYSTIFIED: BIOLOGY,
ECOLOGY, AND CONSERVATION OF DENDROPHYLAX
LINDENII IN FLORIDA AND CUBA
Lawrence W. Zettler1*, Michael E. Kane2, Ernesto B. Mújica3, Laura L. Corey1,
and Larry W. Richardson5
1Department of Biology, Illinois College, 1101 W. College Ave., Jacksonville, IL
62650, USA; 2Environmental Horticulture Department, University of Florida,
P.O. Box 110675, Gainesville, FL 32611, USA; 3Orquideario Soroa, Carretera a
Soroa Km. 8, Candelaria, Pinar del Rio, Cuba; 1Department of Biology, Illinois
College, 1101 W. College Ave., Jacksonville, IL 62650, USA; 5Richardson Nature,
6637 Merryport Lane, Naples, FL 34104, USA
*Author for correspondence: lwzettle@ic.edu
The ghost orchid, Dendrophylax lindenii, is a rare, well-
known, leafless epiphyte native to south Florida and
Cuba. Because this species is challenging to grow,
specimens are often illegally extracted from wild pop-
ulations (poaching), posing a serious threat to its long-
term survival. The orchid is also vulnerable to periodic
hurricanes and rise in sea-level from climate change.
Until recently, little was known about the biology,
ecology, propagation, and reintroduction of D. lindenii
– information that is crucial for effective conservation
strategies planned this century. This paper provides a
synopsis of what we currently know based on new infor-
mation collected over the past five years by specialists in
both countries. Studies in Guanahacabibes Peninsula
(Cuba) and the Florida Panther National Wildlife Ref-
uge (U.S.) have revealed that young ohost orchid roots
emerge beneath older roots at the apices of reduced
shoots, i.e., the orchid grows ‘upside down’ on host
trees. Floral color, structure, and fragrance indicate pol-
lination by large hawkmoths (Sphingidae), yet few flow-
ers produce capsules in a given year. Ghost orchids in
Florida were affixed to the bark of two host tree species,
whereas 18 tree species served this role in Cuba. Nearly
all host trees had corrugated or semi-corrugated bark,
and more than half of the orchids were oriented on
bark that faced north (NW, N, NE). Mycorrhizal fungi
isolated from D. lindenii roots in Florida yielded strains
assignable to Ceratobasidium, one of which facilitated
in vitro seed germination. Of 26 seedlings observed in
Cuba, all were estimated to have germinated 100 days
previously, coinciding with peak hurricane frequency
in the region. Based on the collective progress made,
we are cautiously optimistic that D. lindenii will persist
in natural areas in south Florida and Cuba in the years
ahead.
The ghost orchid, Dendrophylax lindenii (Lindl.)
Benth. ex Rolfe, is perhaps the most familiar
and sought-after of all orchids in the Western
Hemisphere. As a leafless epiphyte restricted
to small, unwelcoming forests in south Florida
and Cuba, the species is easily overlooked for
most of the year until it yields a striking floral
display consisting of one or two long-lasting,
fragrant, white flowers that appear to hover in
mid-air in its dimly lit habitat (Fig. 1). These
flowers are attached to a long, thin stem that
originates from the center of a root mass radi-
ating outward along and within groves of host
tree bark (Fig. 2). Anthesis typically occurs
during the summer months (May-August),
coinciding with excessive heat and humidi-
ty, hordes of biting mosquitoes, and tropical
cyclone activity. Only the most determined
orchid enthusiasts are willing to suffer these
hardships to experience the species first-hand
in its natural habitat. Because of its appeal, rar-
ity, and reputation for being difficult to grow
under greenhouse conditions, D. lindenii re-
mains vulnerable to poaching, as exemplified
by Susan Orlean’s (1998) best-selling book,
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
137
The Orchid Thief. Other threats to the species’
long-term survival include phytophagous pests
(Zettler et al. 2012), hydrological changes to
its low-lying wetland habitat (Langdon 1979;
Coile and Garland 2003), and periodic hurri-
canes (Wiegand et al. 2013) in addition to cli-
mate change. In Cuba, Raventós et al. (2015)
projected that D. lindenii could become extinct
on the Guanahacabibes Peninsula within 25
years if the annual probability of disturbances
(e.g., hurricanes) exceeds 14%.
Until only recently, surprisingly little was
known about the biology and ecology of D.
lindenii from a physiological context (e.g., seed
germination requirements). Given this or-
chid’s rarity and vulnerability, obtaining such
fundamental information becomes crucial
for the development of effective conservation
strategies planned this century. In light of cli-
mate change and sea-level rise in particular,
it is essential that methods be perfected for
artificial propagation of D. lindenii from seed,
leading to reintroduction of seedlings into
suitable habitats on higher ground.
During the past decade, studies carried out
in Florida and Cuba have led to exciting dis-
coveries involving D. lindenii and the biotic
agents on which the orchid depends on for
survival. Much of this new information stems
from collaborative efforts between research-
ers in both the United States and Cuba, who
aspire to save this species from extinction.
This paper represents a general summary of
what we currently know about the ghost or-
chid, including new information gleaned by
recent published studies and studies that are
currently underway. We conclude by offering
recommendations for future work to facilitate
long-term conservation of the species in this
‘age of extinction’.
Classification
The generic name, Dendrophylax Rchb.f., is
derived from the combination of two Greek
words, dendron= tree, and phylax = guardian,
possibly attributed to the orchid’s large root
mass that wraps around the host tree as if to
protect it from outside danger. According to
the World Checklist of Selected Plant Fami-
lies (www.wcsp.science.kew.org), the genus is
composed of a small number (15) of species
confined to Mexico, Central America, Flori-
da, and the Caribbean. Although the accepted
name for the ghost orchid of south Florida and
Cuba is Dendrophylax lindenii (Lindl.) Benth.
ex Rolfe, the literature lists many different
synonyms for the same species. These include:
Angraecum lindenii Lindl., Aeranthes lindenii
(Lindl.) Rchb.f., Polyrrhiza lindenii (Lindl.)
Cogn., and Polyradicion lindenii (Lindl.) Garay.
The synonymy is not limited to the binomi-
al, as the ghost orchid is also known by two
other common names, ‘palm-polly’ and ‘frog
orchid’. The latter name was aptly described
by Correll (1950): “When one looks up into
the gloom of a thickly leaved tree and sees the
extraordinary flowers of this little orchid for
the first time, one is instantly impressed with
its likeness to a thin flat snow-white frog sus-
pended in mid-air – caught, as if it were, in the
middle of a leap from one branch to another”.
All members of the genus Dendrophylax belong
to Epidendroideae – a mainly tropical group
and by far the largest of the orchid subfami-
lies (Pridgeon et al. 1999, 2005, 2009, 2014).
Within this subfamily, D. lindenii is nestled in
the tribe Vandeae and subtribe Angraecinae,
which includes well-known African genera
Aeranthes and Angraecum as well as Campylocen-
trum of south Florida (Pridgeon et al. 2014).
The white flower, long nectar spur, and
sweet-smelling evening fragrance (Sadler et al.
2011) of the species fit the typical profile of
well-known, moth-pollinated angraecoids of
Africa and Madagascar, a group that includes
the so-called Darwin’s orchid, Angraecum ses-
quipedale Thouars). Thus, D. lindenii almost
seems like a misplaced African orchid that
somehow “leaped” into the West Indies and
south Florida from the east.
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
138
Fig. 1. Dendrophylax lindenii in bloom within the dimly lit hab-
itat of the Florida Panther National Wildlife Refuge. (Photo:
Larry W. Richardson).
Physical description
One of the most striking aspects of the ghost
orchid is the large, appealing floral display
associated with a leafless root mass. For this
reason, D. lindenii and other leafless showy
orchids garner considerable interest in judg-
ing contests. Among the 28,000+ accepted
orchid species worldwide according to the
Worldwide Checklist of Selected Plant Fam-
ilies (Govaerts, this volume), fewer than 300
exist as leafless epiphytes, all spanning one
tribe (Vandae) within the subfamily Epiden-
droideae (Carlsward et al. 2006; Stewart et
al. 2006). In south Florida and Guanahaca-
bibes Peninsula, two other leafless species
occur naturally with D. lindenii: Dendrophylax
porrectus (Rchb.f.) Carlsward & Whitten and
Campylocentrum pachyrrhizum (Rchb.f.) Rolfe.
Roots of D. lindenii and C. pachyrrhizum are
comparable in size and may be confused with
one another when not in flower, but D. lin-
denii can be distinguished by the presence of
white linear patches (pneumathodes) that run
parallel along the length of the root. Roots of
D. lindenii are also rounded or ovoid in cross
section whereas those of C. pachyrrhizum are
generally flat or compressed.
Although leafless most of its life, protocorms
(young seedlings) of D. lindenii possess a dorsal
crest (Veyret 1974; Pridgeon et al. 1999) that
is considered by Nishimura (1981) to be the
first leaf. As the protocorm matures leading
to a fully photosynthetic root system, leaves
that subsequently form remain reduced as
scales (Hoang et al. 2017). A fascinating dis-
covery was recently revealed in situ by Hoang
et al. (2017), but noted prior by Prem Subrah-
manyam, that new D. lindenii roots emerge
from beneath older roots at the apices of
reduced shoots (Fig. 2). In other words, the
plant actually grows “upside down” with the
apical shoot positioned between the host
tree’s bark surface and the root mass oriented
to the outside. Thus, when a plant is viewed
on the surface of a tree, the oldest roots are
those closest to the observer, and the newest
roots can be seen emerging from underneath
in contact with the bark surface. On mature
plants, the inflorescence likewise originates
from the underside of the plant. This is the
first time such a growth pattern has been doc-
umented in Orchidaceae (Hoang et al. 2017).
Why this occurs is unknown, but it is con-
ceivable that this feature affords the delicate
shoot protection from herbivory and/or des-
iccation. Moreover, this orientation may also
allow young roots to bond quickly to the bark
substrate as they begin to grow rather than
growing up and over the older roots.
Flower
It is the flower that distinguishes D. lindenii
from all other orchid species native to the
United States and Canada. Its large size,
lily-white color, and two long, twisting lobes
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
139
Fig. 2. Close-up view of a mature plant growing on its host
tree in Cuba. The image clearly shows a young root emerging
from the center, beneath older roots that are oriented to the
outside. The unique upside down growth pattern was first de-
scribed by Hoang et al. (2017). (Photo: Lawrence W. Zettler).
against a dark backdrop of its swampy habitat
are both appealing and easily recognizable. For
this reason, D. lindenii serves as a visual sym-
bol for the ca. 250 species of orchids native to
the U.S. and Canada, evidenced by its use in
logos (e.g., Orchid Society of Coral Gables),
jewelry, and even on the sides of trucks for
a national moving company. Those fortunate
enough to smell the sweet-smelling flowers to-
wards nightfall will find the fragrance equally
appealing. When compared to its angraecoid
relatives in Madagascar and Africa, D. lindenii
flowers might be considered average or certain-
ly not unique given its large size, white color,
and long nectar spur. It thus serves as a kind
of “ambassador” for Old World epiphytes,
many of which are also faced with extinction.
Although the floral characteristics (e.g., white
color, long nectar spur, sweet-smelling fra-
grance) clearly point to a sphingid moth pol-
linator with a long proboscis (Faegri and van
der Pijl 1979; Dressler 1981; Grant 1983),
there are no known reports of a pollinator ac-
tually being captured with attached pollinia.
In south Florida, the only moth with a probos-
cis long enough to probe deep into the nectar
spur is the giant sphinx moth, Cocytius antaeus
Drury, the larvae of which feed on pond ap-
ple (Annona glabra L.) – a host tree of D. lin-
denii. In Cuba, Raventós et al. (2015) observed
hawkmoth pollination of D. lindenii, but the
moth species was not identified. Sadler et al.
(2011) devised a technique used to sample the
blend of volatile chemicals emitted by D. lin-
denii flowers in situ and determined that the
compounds were typical for other moth-pol-
linated flowers. Upon mixing the chemicals
artificially in the laboratory in the proper pro-
portions, the scent closely mimicked the fra-
grance of D. lindenii (L. Zettler, pers. observ.).
At least one perfume company has taken steps
to use the blend in marketing, and plans are
underway for a percentage of perfume sales to
be diverted to orchid conservation. It is also
conceivable that artificial scent could be used
in natural habitats to attract pollinators into
small D. lindenii populations where few flow-
ers occur. Doing so could potentially improve
fruit set through cross pollination, leading to
additional seedlings spread in such areas.
Distribution and current status
The ghost orchid was first discovered in Cuba
in 1844 by the Belgian botanist, Jean Jules
Linden, and described two years later by John
Lindley as Angraecum lindenii (Luer 1972). In
1880 – 36 years after its discovery in Cuba -
D. lindenii was collected for the first time in
the United States from Palm Hammock near
Cape Romano in Collier County, Florida,
by A. H. Curtiss (Correll 1950). According
to Luer (1972), the distribution of D. lindenii
also stretched southeast into the Bahamas,
but there are no known records from this lo-
cation (Ackerman and Hagsater 2014; J. D.
Ackerman, pers. comm.) despite the presence
of potentially suitable habitats in the archipel-
ago. Thus, this species appears to be restrict-
ed to Cuba and south Florida in populations
separated ca. 600 km by the Straits of Florida.
For reasons still not well understood, the dis-
tribution of D. lindenii is limited to tropical
(mixed) semi-deciduous hardwood forests in
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
140
select, low-lying areas. In south Florida’s Big
Cypress Basin Eco-region, D. lindenii frequents
cypress domes and strand swamps where it typ-
ically attaches to the bark of woody trees root-
ed in standing water. It is assumed that the
high relative humidity levels within these for-
ested wetlands serve to insulate D. lindenii and
other cold-sensitive epiphytes from occasional
subfreezing temperatures. In Cuba where le-
thal sub-freezing temperatures do not occur,
the largest known population occurs on the
Guanahcacabibes Peninsula on the western
tip of the island. Unlike south Florida, most
of the orchids in western Cuba are affixed to
the bark of host trees rooted on crack reef
limestone (Acevedo 1992) with little stand-
ing water. Both locations pose a challenge for
researchers seeking to study D. lindenii over a
wide area because of the difficult terrain cou-
pled with the usual hardships associated with
tropical field work (e.g., heat, humidity, mos-
quitoes). In south Florida, wading waist-high
through stagnant water occupied by alligators
and venomous snakes is the most effective
way to document and study D. lindenii at close
range (Fig. 3).
Although there are no venomous snakes in
Cuba, the jagged, sharp rocks in Guanahaca-
bibes Peninsula have resulted in twisted an-
kles, bruises, and deep cuts among research-
ers. With its leafless form and gray-green
roots that visibly blend in with mosses and
lichens on the host tree, D. lindenii is easily
overlooked when not in flower, especially giv-
en the obstacles and harsh conditions. Thus,
there is guarded optimism that more individ-
uals and new populations may be discovered
in the coming years in other suitable habitats
throughout the region. Nevertheless, D. lin-
denii is by no means common and appears to
be selective of where it naturally occurs.
Current estimates for numbers of the species
throughout the region remain unclear. In
south Florida, the total estimated number of
individuals is ca. 2000, most within the Faka-
hatchee Strand State Park and adjacent Flor-
ida Panther National Wildlife Refuge, locat-
ed in remote eastern Collier County. During
July and August of 2017, a total of 157 new
plants were documented in the Florida Pan-
ther NWR, raising the total of number at the
site to 400 individuals (E. Mujica, unpubl.
data). In Guanahacabibes National Park, ca.
250 D. lindenii have been counted as of 2017,
and 100+ additional individuals are thought
to exist in more remote regions of the park.
Elsewhere in Cuba, two individual orchids are
known to occur in Zapata Swamp, but prob-
ably more await discovery. There are also re-
cords of D. lindenii in Granma province, but
that population is assumed to be small and
vulnerable to human activity. Efforts are now
underway to document additional plants in
these sites in the coming years.
Habitat
When populations of D. lindenii in south Flor-
ida and Cuba are visited by the same observ-
er, the difference between the two habitats is
striking. For rare orchids such as D. lindenii,
this observation seems to counter a prevail-
Fig. 3. Dr. Ernesto Mújica is shown in waist-deep water in the
Florida Panther National Wildlife Refuge adjacent to a plant
of D. lindenii in flower. The presence of alligators, venomous
snakes, mosquitoes, coupled with high heat and humidity,
pose a challenge to those seeking to study the orchid first-
hand. (Photo: Larry W. Richardson).
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
141
ing assumption that rare orchids are restrict-
ed in distribution because they have specific
biotic and/or abiotic needs. In 2015, we es-
tablished a joint research collaboration with
the common goal of conserving D. lindenii in
both countries. The initial aim of this ongo-
ing, 4-year project was to provide a detailed
description of ghost orchid habitats in south
Florida and Cuba, with the Florida Panther
National Wildlife Refuge and Guanahaca-
bibes National Park serving as study sites, re-
spectively. Population size and fecundity (e.g.,
seedling recruitment) were identified as prior-
ities of study to determine if the existing habi-
tats are stable, growing or in decline. We also
wanted to know if, and to what extent, these
habitats might be linked to specific mycorrhi-
zal fungi that D. lindenii relies on for survival.
After three years of collecting data, a more
complete picture is beginning to emerge that
is described in more detail by Mujica et al.
(2018). In Cuba, D. lindenii has been found
on 18 host tree species, primarily Maba crassin-
ervis (Krug. & Urb.) Urb., Erythroxylum areola-
tum L., and Comocladia dentata Jacq. In south
Florida, however, just two species served as
primary host trees, Fraxinus caroliniana Mill.
(pop ash) and A. glabra (pond apple). One
similarity shared by most of the host trees was
corrugated or semi-corrugated bark, which we
hypothesize serves as a physical anchor point
for wind-blown seeds. Compared to smooth
bark surfaces, corrugated bark would also con-
ceivably retain moisture for longer periods as
well as organic matter (detritus) that might
support a mycoflora conducive to seed ger-
mination needs. Another revelation was D.
lindenii’s orientation on host trees; more than
half of all orchids documented in both coun-
tries had a north-facing component (NW, N,
NE). We suspect that, like corrugated bark,
north-facing surfaces would retain more mois-
ture, as these surfaces would be shielded from
direct sunlight for longer periods of time each
day. Both areas had comparable mean annual
rainfall amounts (1300-1400 mm/year) and
temperatures (ca. 25 °C), except orchids in
Cuba which would be sheltered from occa-
sional sub-freezing temperatures. Except for
tropical cyclone activity and afternoon thun-
derstorms during the rainy season, the forests
in both areas were marked by minimal breez-
es. Thus, D. lindenii habitats are cloaked by
humid, stagnant air for most of the year.
Seedling recruitment and “senile” popu-
lations
According to Rasmussen et al. (2015), orchid
populations are characterized as “senile” when
opportunities for local seedling recruitment
Fig. 4. One of the 26 seedlings photographed in Cuba, five
months after they were initially discovered on corrugated
bark of a single host tree. The dorsal crest is still visible (lower
right) even after the seedling has initiated three roots that are
shown growing within the root crevices. Pneumathodes are
visible on roots at this early stage of development, and their
presence confirms that the seedlings are indeed those of D.
lindenii. A plastic tack, used for marking seedlings on host
trees during surveying, is seen at the left of the image. (Photo:
Lawrence W. Zettler).
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
142
have all but disappeared. Thus, our surveys
in south Florida and Cuba examined seedling
recruitment to determine if D. lindenii is in-
trinsically endangered or if it merely lacks the
components needed for seedling recruitment
in these areas (e.g., specific mycorrhizal fungi).
Thus far, our surveys have revealed that seed-
ling recruitment is more widespread in Cuba,
given that a higher percentage (ca. 30%) of
the orchids documented were in the seed-
ling stage. One host tree, in fact, harbored 26
seedlings that appeared to be ca. 100 days old
based on the presence of the dorsal crest (Fig.
4) on seedlings of comparable age reared in
vitro (e.g., Hoang et al. 2017). In the Florida
Panther NWR, however, seedlings were docu-
mented at only one of five sites. Considering
that most host trees of D. lindenii in Florida
were fully mature, we suspect that these pop-
ulations, with one exception, may indeed be
senile.
Mycorrhizal fungi
Yukawa et al. (2009) proposed that orchids in
Vandeae and especially those in the Angraeci-
nae (e.g., D. lindenii), are linked to mycorrhizal
fungi in Ceratobasidiaceae. Chomicki et al.
(2014) went a step further by stating, “The
type and fungal structures observed in Dendro-
phylax and the phylogenetic signal in Angrae-
cinae altogether suggest that D. lindenii forms
mycorrhiza with a Basidiomycete fungus of
the Ceratobasidiaceae”.
Our ongoing efforts in the Florida Panther
NWR supports their view. For example, three
different fungal endophytes (so far) have been
recovered from D. lindenii roots in Florida,
and all have been identified as belonging to
the genus Ceratobasidium Rogers confirmed
by molecular methods (DNA sequencing of
the rRNA-ITS locus). One of these strains,
in fact, facilitated in vitro seed germination of
D. lindenii, confirming its mycorrhizal nature
(Hoang et al. 2017). In a separate study carried
out by Lynnaun Johnson and Greg Mueller
at Northwestern University and Chicago Bo-
tanic Garden, respectively, preliminary results
indicate that D. lindenii has a high specificity
for Ceratobasidium spp. (Johnson and Mueller
2017), further supporting this view. Swarts et
al. (2010) provided evidence that orchid rari-
ty is linked to high mycorrhizal specialization.
Thus, it is conceivable that D. lindenii may be
limited in distribution because it is limited by
its mycorrhizal fungi, namely specific strains
of Ceratobasidium. Plans are underway to iso-
late additional endophytes in south Florida
and to expand this work into Cuba, assuming
proper collecting and import/export permits
are secured.
Corrugated bark and Ceratobasidium
Why D. lindenii appears to associate with Cer-
atobasidium in south Florida may, in part, be
linked to both nutrition and corrugated bark.
For example, orchid mycorrhizal fungi in the
Rhizoctonia complex produce enzymes for de-
composition of cellulose (Zelmer et al. 1996)
that are used in the breakdown of leaf litter
into simple sugars. Orchids that digest these
fungi in their root systems therefore receive
a supplemental supply of carbon (=mycotro-
phy), augmenting photosynthesis. Fungi in the
genus Ceratobasidium, however, also produce
polyphenoloxidases that are involved with
the breakdown of woody debris (Rasmussen
1995). Thus, it is conceivable that Ceratobasid-
ium strains may be targeted by D. lindenii to
gain a selective advantage for colonizing cor-
rugated bark surfaces where woody debris and
moisture accumulate, the latter being more
prevalent on the north sides of the host tree.
As the orchid matures, roots in moist crevices
would have regular access to Ceratobasidium,
supplementing photosynthesis with mycotro-
phy which would be expected for a leafless or-
chid. Indeed, all three of the Ceratobasidium
strains recovered from D. lindenii in Florida
were obtained from the roots of mature plants
(Hoang et al. 2017; Mujica et al. 2018). More-
over, Ceratobasidium strains have also been
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
143
recovered from Campylocentrum pachyrrhizum
in Florida (Radcliffe et al. 2015), indicating
that this may be a more widespread phenom-
enon among leafless orchids in Angraecinae.
Additional studies are needed to determine
whether or not other bark epiphytes (Dressler
1981) associate with Ceratobasidium fungi,
which might explain, in part, why these or-
chids evolved independently in three different
regions of the globe (Dressler 1993). Studies
that examine a possible link between mosses
and the bark mycoflora would also be worth
pursuing.
Hurricanes, climate change, and other
threats
In the Caribbean Basin, hurricanes play a
dominant role in the abundance and distri-
bution of many species (Tanner et al. 1991;
Waide and Lugo 1992; Willig et al. 2011) and
pose a serious threat to epiphytes even when
trees are not toppled (Tremblay 2008). Muji-
ca et al. (2013) assessed population recovery
of two epiphytic orchids in Cuba, including
D. lindenii, following a severe hurricane (Ivan,
category 5) that passed over Guanahacabibes
Peninsula in 2004. They concluded that veg-
etative parts of D. lindenii did not recover and
that lower rates of fruit set may have been
caused by the hurricane’s negative impact on
the community of pollinators. Raventós et al.
(2015) proposed that the lack of storage or-
gans (e.g., pseudobulbs) in leafless epiphytes
renders these plants more vulnerable to sto-
chastic disturbances (hurricanes), and for this
reason D. lindenii is thought to rely more heav-
ily on seedling recruitment for recovery. Most
orchids depend heavily on pollinators and
mycorrhizal fungi for seedling recruitment,
but these biotic agents appear to be especially
important for the perpetuation of D. lindenii
in the region following hurricanes.
As detrimental as hurricanes may be, D. lin-
denii may have adapted its life cycle around
the timing of moderate stochastic disturbanc-
es to its reproductive advantage. For example,
the recent discovery of 26 small D. lindenii
seedlings on one host tree in Guanahacabibes
Peninsula in the month of January indicates
that seeds germinated ca. 100 days prior. The
identity and age of these seedlings was con-
firmed by the presence of the dorsal crest and
small root length, which closely matched seed-
ling developmental stages observed in vitro by
Hoang et al. (2017). If true, seed release from
capsules and subsequent germination would
have occurred in Guanahacabibes during late
August or early September, when hurricane
frequency nears its peak in the region (http://
www.nhc.noaa.gov/climo/). This time frame
also agrees with the May-June fruiting times
observed for D. lindenii in Cuba (Mujica et
al., unpubl.). Taken together, we speculate
that D. lindenii is pollinated by moths before
the onset of hurricane season and that seed
dispersal is assisted by strong winds during
the mid/late summer months when tropical
storms are most frequent. The tiny, dust-like
seeds then become lodged in the moist crevic-
es of corrugated bark harboring a saprophytic
mycoflora. Seed germination would then be
assisted by Ceratobasidium present in the bark,
at a time when cyclone activity begins to sub-
side. Our ongoing surveys in Cuba and south
Florida are expected to test this hypothesis in
the years ahead.
The scenario presented above assumes that D.
lindenii and other orchids in the Caribbean
Basin are adapted to a normal set of climatic
conditions over a long period of time. In light
of ongoing climate change, however, the fre-
quency and severity of hurricanes is expected
to increase this century (Mujica et al. 2013),
posing a serious threat to the long-term sur-
vival of D. lindenii in the region. In addition,
the species in both south Florida and Cuba
occupies habitats in low-lying areas vulner-
able to sea-level rise (3.2 mm/year; https://
climate.nasa.gov), and new suitable habitats
on higher ground will be needed to accommo-
date it. Thus, they will need to colonize new
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
144
Fig. 5. One of many laboratory-grown plants that were suc-
cessfully established on burlap mesh in a greenhouse at the
University of Florida. This specimen is a three-year-old plant-
let that flowered under greenhouse conditions and indicates
a high frequency of flowering in the past few years. (Photo:
Mike Kane).
areas via spontaneous seedlings generated on
their own, through human intervention (=as-
sisted migration) or both. The likelihood of
the former being successful is doubtful unless
fruit set can be significantly increased, per-
haps aided by artificial hand pollination. The
approach most likely to succeed will consist of
large-scale artificial propagation using mature
seed derived from cross-pollination between
genetically diverse individuals, followed by
seedling reintroduction into new habitats.
In south Florida, another potential threat to
D. lindenii linked to nature being out of bal-
ance looms on the horizon – one that orig-
inated in Asia. The emerald ash tree borer,
Agrilus planipennis Fairmaire, is an exotic, me-
tallic green beetle that has been implicated in
the widespread die-off of ash trees (Fraxinus)
in central and northeast North America. As
our ongoing surveys have revealed, ca. 70%
of all ghost orchids documented in the Flor-
ida Panther NWF were affixed to pop ash (F.
caroliniana). Although it remains unknown
whether pop ash is susceptible to the beetle,
land managers in south Florida should be
prepared for a worst-case scenario that could
include the rapid die-off of D. lindenii’s prima-
ry host tree should the beetle spread south.
Thus far, the insect has been documented as
far south as Atlanta, Georgia, and the pros-
pect of it entering south Florida habitats is
real. The spread of the beetle to Cuba seems
unlikely given that the orchid populations are
separated by ca. 600 km by the Florida Straits.
Even if the insect somehow entered Cuba, it
would be expected to have little or no impact
on D. lindenii in Guanahacabibes where host
tree diversity is much higher and where Fraxi-
nus species do not occur.
Propagation – a glimmer of hope
Considering the current and projected threats
to D. lindenii mentioned previously, habi-
tat preservation alone does not appear to be
sufficient for this orchid’s long-term surviv-
al. According to Swarts and Dixon (2009),
orchid conservation in this age of extinction
will hinge on three actions: 1) design and
management of natural reserves, taking the
specialized needs of orchids into account; 2)
establishment of ex situ seed and fungal banks
for orchids under immediate threat; and 3)
development of techniques for orchid resto-
ration. Recently, significant breakthroughs in
propagation of D. lindenii have been made by
Mike Kane and his students at the University
of Florida which may now meet the needs of
the third action above. For example, Hoang
Nguyen’s doctoral research provided a frame-
work for in vitro seed germination of D. lin-
denii for the first time, and this accomplish-
ment was particularly noteworthy because it
described, in detail, the procedure by which
thousands of seedlings could be generated,
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
145
with and without the use of mycorrhizal fungi
(Hoang 2017). During in vitro culture, he care-
fully documented embryo development and
seedling morphology, thereby providing clues
into the unique inverted growth pattern on
trees. His work also confirmed experimentally
that a strain of Ceratobasidium was indeed a
mycorrhizal symbiont of D. lindenii, not merely
a benign fungal inhabitant of the root cortical
region. Clonal propagation via protocorm-like
bodies (PLBs) was also carried out, and this
yielded a sizable number of seedlings that were
used in pre-hardening and acclimatization ex-
periments. Surprisingly, seedlings that were
pre-hardened exhibited stress, whereas those
without pre-hardening did not. Intermittent
misting had a positive effect on establishment
and survival, and seedlings grown in this man-
ner flowered in a greenhouse three years after
germination (Fig. 5), demonstrating that culti-
vation of D. lindenii was indeed possible.
Following Hoang’s in vitro and ex vitro research
in Gainesville, experiments then shifted to
reintroduction attempts in south Florida. In
2015, for example, dozens of D. lindenii seed-
lings grown on burlap mesh in a greenhouse
were introduced into an existing ghost orchid
site in the Florida Panther NWR where they
were affixed to the bark host tree. After four
months, 87% of the seedlings survived, and
several subsequently flowered a year later.
The same reintroduction protocol used at the
Florida Panther NWR was then applied ex situ
when 80 D. lindenii seedlings were placed on
young pop ash and pond apple trees in the
Naples Botanical Garden along a boardwalk
in public view. After two years, survival rates
remained high, and several of these seedlings
flowered, generating media attention and
serving as an effective educational tool. Like-
wise, seedlings were also placed in a public
display at the Florida Museum of Natural His-
tory’s Butterfly Garden in Gainesville where
they continue to garner interest. Additional
studies aimed at understanding and perfect-
ing the acclimatization needs of D. lindenii
are being carried out by another graduate
student, Jameson Coopman, in Mike Kane’s
lab. Collectively, these efforts offer hope that
long-term conservation through reliable prop-
agation of D. lindenii may now be possible.
Future work
Much has been learned about D. lindenii
during the last five years, and more informa-
tion is expected to follow, improving the like-
lihood that this North American “signature”
orchid will persevere in this age of extinction.
The ongoing surveys of D. lindenii populations
in south Florida and Cuba, spearheaded by
Ernesto Mújica, provide a fundamental un-
derstanding of the orchid’s specific habitat re-
quirements, and this information is expected
to benefit propagation and reintroduction ef-
forts throughout the region. The surveys will
also provide land managers with more confi-
dence in making predictions about a popula-
tion’s status (i.e., whether it is stable, senile
or growing). Collectively, these surveys are on
pace to fulfill the first action mentioned by
Swarts and Dixon (2009) above.
Important breakthroughs have been made in
Mike Kane’s lab that now make it possible for
D. lindenii to be mass-propagated from seed,
benefiting orchid restoration efforts in both
countries. Hoang Nguyen’s graduate work has
led to the much-awaited (published) protocol
for in vitro seed germination of D. lindenii, and
Jameson Coopman’s research involving ac-
climatization is expected to improve seedling
survival rates ex vitro. Their work should now
make it possible for D. lindenii to be successful-
ly reintroduced into existing and new habitats
and into ex situ collections.
More information is needed at understand-
ing seedling recruitment in situ and the biot-
ic agents that facilitate the process. Research
by Zettler and Corey at Illinois College and
others (Lynnaun Johnson, Greg Mueller) pro-
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
146
vides new clues as to the types of mycorrhizal
fungi utilized by D. lindenii in south Florida,
but comparative studies are still needed in
Cuba. Moreover, we know little about the
natural pollinators of D. lindenii other than
circumstantial evidence that implicates large
sphingid moths.
To satisfy the third component for successful
orchid conservation outlined by Swarts and
Dixon (2009), seed and fungal banks need to
be established for D. lindenii in both countries.
In the United States, this process is current-
ly underway coordinated through the North
American Orchid Conservation Center (NA-
OCC; Whigham and Zettler 2016). Beginning
in 2018, seed and mycorrhizal fungi from D.
lindenii and other rare orchids of western Cuba
will be safeguarded in storage at Soroa Orchid
Botanical Garden in Pinar del Rio. This facili-
ty will also oversee the seed propagation of D.
lindenii and other endangered orchids in that
country for eventual reintroduction.
As this paper has shown, two difficult obsta-
cles have been cleared in the battle to save the
ghost orchid from extinction – one linked to
demystifying D. lindenii’s biology and ecology,
and the other to overcoming political interfer-
ence between the United States and Cuba. Al-
though much work remains, we are cautiously
optimistic that D. lindenii will persist in nat-
ural areas in south Florida and Cuba in the
years ahead.
Acknowledgments
This joint research effort between specialists
in the United States and Cuba was made
possible by the hard work, persistence, and
diplomatic talent of Steven M. Gardner, De-
partment of World Languages and Cultures
at Illinois College, to whom we are grateful.
We express sincere gratitude to the mem-
bers of the Naples (FL) Orchid Society for
their long-standing support of ghost orchid
research in the region, with special thanks to
Kit Kitchen-Maran, La Raw Maran, Bill Over-
ton, and Connie and Lake Sims. Support
from the Illinois Orchid Society, Prairie State
Orchid Society, Illinois College’s Tillery Fac-
ulty-Student Research Collaboration Fund,
and Native Orchid Conference, Inc. (Fred
Case Grant) also made this research possible
by funding Illinois College students in Cuba.
We thank the following Ilinois College stu-
dents, current and former, for their valiant
field research efforts: Eve Bahler, Adam Herd-
man, Justin Mably, Connor Melton, Shan-
non Skarha, and Jack Waggoner. Exceptional
graduate-level research by Hoang Nguyen and
Jameson Cooper at the University of Flori-
da served as an inspiration to everyone. We
warmly thank our colleagues in Cuba for their
dedication to our work: José Bocourt (Soroa
Orchid Garden), Elaine González (Soroa Or-
chid Garden), and José L. Camejo (Parque
National Guanahacabibes). In south Florida,
the long-standing support by staff at the Flori-
da Panther National Wildlife Refuge is much
appreciated, especially Mark Danaher, Kevin
Godsea, and Ben Nottingham. Special thanks
to Andy Stice (Illinois College) for field sup-
port in both countries.
Literature cited
Acevedo, M. G. 1992. Geografía física de Cuba. Tomo II. Edi-
torial Pueblo y Educacon, La Habana.
Ackerman, J. D. and Hagsater, E. 2014. Orchid flora of the
Greater Antilles. New York Botanical Garden Press, New York.
Carlsward, B. S., Whitten, W. M., Williams, N. H., and Byte-
bier, B. 2006. Molecular phylogenetics of Vandeae (Orchida-
ceae) and evolution of leaflessness. American Journal of Botany
93: 770-786.
Chomicki, G., Bidel, L. P. R., and Jay-Allemand, C. 2014. Ex-
odermis structure controls fungal invasion in the leafless ep-
iphytic orchid Dendrophylax lindenii (Lindl.) Benth. ex Rolfe.
Flora 209: 88-94.
Coile, N. C. and Garland, M. 2003. Notes on Florida’s en-
dangered and threatened plants. Florida Department of Ag-
riculture and Consumer Services Division of Plant Industry,
45. Gainesville, Florida.
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
147
Correll, D. S. 1950. Native orchids of North America north of
Mexico. Stanford University Press.
Dressler, R. L. 1981. The orchids: natural history and classifica-
tion. Harvard University Press.
Dressler, R. L. 1993. Phylogeny and classification of the orchid
family. Cambridge University Press.
Faegri, K. and van der Pijl, L. 1979. The principles of pollination
ecology. 3rd ed. Pergamon Press, Oxford, UK.
Grant, V. 1983. The systematic and geographical distribution
of hawkmoth flowers in the temperate North American flora.
Botanical Gazette 144: 439-449.
Hoang, N. H. 2016. In vitro physiology of recalcitrant tissue cul-
tured plants in the Nymphaeaceae, Alismataceae, and Orchidaceae.
Ph.D. dissertation. University of Florida, Gainesville, Florida.
Hoang, N. H, Kane, M. E., Radcliffe, E. N., Zettler, L. W.,
and Richardson, L. W. 2017. Comparative seed germination
and seedling development of the ghost orchid, Dendrophylax
lindenii (Orchidaceae), and molecular identification of its
mycorrhizal fungi from south Florida. Annals of Botany 119:
379-393.
Johnson, L. J. A. N., Kane, M. E., and Mueller, G. M. 2017.
Role of mycorrhizal fungi in the nutrition and fine scale dis-
tribution patterns of epiphytic orchids. Proceedings of the
9th International Conference on Mycorrhizae, Prague, Czech
Republic.
Langdon, K. R. 1979. The ghost orchid, Polyrrhiza lindenii,
and endangered species in Florida. Circular 56. Florida De-
partment of Agriculture and Consumer Services, Division of
Plant Industry, Gainesville, Florida.
Luer, C. A. 1972. The native orchids of Florida. New York Bo-
tanical Garden, New York.
Mújica, E. B., Raventós, J., González, E., and Bonet, A. 2013.
Long-term hurricane effects on populations of two epiphytic
orchid species from Guanahacabibes Peninsula, Cuba. Lank-
esteriana 13: 47-55.
Mújica, E. B., Mably, J. J., Skarha, S. M., Corey, L. L., Rich-
ardson, L. W., Danaher, M. W., González, E. H., and Zettler,
L. W. 2018. A comparison of ghost orchid (Dendrophylax lin-
denii) habitats in Florida and Cuba, with special reference to
seedling recruitment and mycorrhizal fungi. Botanical Journal
of the Linnean Society 186: 572-586.
Nishimura, G. 1981. Comparative morphology of Cattleya
and Phalaenopsis (Orchidaceae) seedlings. Botanical Gazette
142: 360-365.
Orlean, S. 1998. The orchid thief: a true story of beauty and obses-
sion. Random House, New York.
Pridgeon, A. M., Cribb, P. J., Chase, M. W., and Rasmus-
sen, F. N. (ed.). 1999. Genera orchidacearum. Vol. 1. General
introduction, Apostasioideae, Cypripedioideae. Oxford University
Press.
Pridgeon, A. M., Cribb, P. J., Chase, M. W., and Rasmussen,
F. N. (ed.). 2005. Genera orchidacearum. Vol. 4. Epidendroideae
(Part one). Oxford University Press.
Pridgeon, A. M., Cribb, P. J., Chase, M. W., and Rasmussen,
F. N. (ed.). 2009. Genera orchidacearum. Vol. 5. Epidendroideae
(Part two). Oxford University Press.
Pridgeon, A. M., Cribb, P. J., Chase, M. W., and Rasmussen,
F. N. (ed.). 2014. Genera orchidacearum. Vol. 6. Epidendroideae
(Part three). Oxford University Press.
Radcliffe, E. N., Bishop, K. M., Corey, L. L., and Zettler, L.
W. 2015. Mycorrhizal fungi from mature epiphytic orchids
and seedlings native to south Florida. Transactions of the Illi-
nois State Academy of Sciences (107th meeting), Supplement to
Volume 108, 49.
Rasmussen, H. N. 1995. Terrestrial orchids: from seed to my-
cotrophic plant. Cambridge University Press.
Rasmussen, H. N., Dixon, K. W., Jersáková, J., and Tĕšitelová,
T. 2015. Germination and seedling establishment in orchids:
a complex of requirements. Annals of Botany 116: 391-402.
Raventós, J., González, E., Mújica, E., and Doak, D. F. 2015.
Population viability analysis of the epiphytic ghost orchid
(Dendrophylax lindenii) in Cuba. Biotropica 47: 179-189.
Sadler, J. J., Smith, J. M., Zettler, L. W., Alborn, H. T., and
Richardson, L. W. 2011. Fragrance composition of Dendro-
phylax lindenii (Lindley) Bentham ex Rolfe (Orchidaceae),
using a novel technique applied in situ. European Journal of
Environmental Sciences 1: 137-141.
Stewart, J., Hermans, J., and Campbell, B. 2006. Angraecoid
orchids: species from the African region. Timber Press, Portland,
Oregon.
Swarts, N. D. and Dixon, K. W. 2009. Terrestrial orchid
conservation in the age of extinction. Annals of Botany 104:
543-556.
Tanner, E. V., Kapos, K., and Healey, J. R. 1991. Hurricane
effects on forest ecosystems in the Caribbean. Biotropica 23:
513-521.
Tremblay, R. 2008. Ecological correlates and short-term ef-
fects of relocation of a rare epiphytic orchid after Hurricane
CONFERENCE PAPERS / CONSERVATION SCIENCE / ZETTLER ET AL.
148
Georges. Endangered Species Research 5: 83-90.
Veyret, Y. 1974. Development of the embryo and young seed-
ling stages of orchids. In The orchids: scientific studies (ed. C. L.
Withner), pp. 223-266. John Wiley & Sons, New York.
Yukawa, T., Ogura-Tsujita, Y., Shefferson, R. P., and Yokoya-
ma, J. 2009. Mycorrhizal diversity in Apostasia (Orchidaceae)
indicates the origin and evolution of orchid mycorrhiza.
American Journal of Botany 96: 1997-2009.
Waide, R. B. and Lugo, A. E. 1992. A research perspective
on disturbance and recovery of a tropical montane forest. In
Tropical forests in transition (ed. J. G. Goldammer), pp.173-190.
Birkhauser Verlag, Basel, Switzerland.
Whigham, D. F. and Zettler, L. W. 2016. North American
Orchid Conservation Center (NAOCC) – getting organized
and moving forward. Abstract, 6th International Orchid
Conservation Congress, 16-22 May 2016, Kadoorie Farm and
Botanic Garden, Hong Kong.
Wiegand, T., Raventós, J., Mújica, E., González, E., and
Bonet, A. 2013. Spatio-temporal analysis of the effects of
hurricane Ivan on two contrasting epiphytic orchid species in
Guanahacabibes, Cuba. Biotropica 45: 441-449.
Willig, M. R., Bloch, C. P., Brokaw, N., Higgins, C., Thomp-
son, J., and Zimmerman, C. R. 2011. Cross-scale responses of
biodiversity to hurricane and anthropogenic disturbance in a
tropical forest. Ecosystems 10: 824-838.
Zelmer, C. D., Cuthbertson, L., and Currah, R. S. 1996. Fun-
gi associated with terrestrial orchid mycorrhizas, seeds and
protocorms. Mycoscience 37: 439-448.
Zettler, J. A., Zettler, L. W., and Richardson, L. W. 2012.
Pestiferous scale insects on native epiphytic orchids in south
Florida: a new threat posed by introduced species. Southeast-
ern Naturalist 11: 127-134.
