Challenges and 
opportunities in 
orchid ecology and 
conservation
Edited by  
Pavel Kindlmann, Tiiu Kull and Dennis Whigham
Published in  
Frontiers in Ecology and Evolution 
Frontiers in Environmental Science
August 2023
FRONTIERS EBOOK COPYRIGHT STATEMENT About Frontiers
The copyright in the text of individual Frontiers is more than just an open access publisher of scholarly articles: it is 
articles in this ebook is the property 
of their respective authors or their a pioneering approach to the world of academia, radically improving the way 
respective institutions or funders. scholarly research is managed. The grand vision of Frontiers is a world where 
The copyright in graphics and images all people have an equal opportunity to seek, share and generate knowledge. 
within each article may be subject 
to copyright of other parties. In both Frontiers provides immediate and permanent online open access to all its 
cases this is subject to a license publications, but this alone is not enough to realize our grand goals.
granted to Frontiers. 
The compilation of articles constituting 
this ebook is the property of Frontiers. Frontiers journal series
Each article within this ebook, and the The Frontiers journal series is a multi-tier and interdisciplinary set of open-
ebook itself, are published under the access, online journals, promising a paradigm shift from the current review, 
most recent version of the Creative 
Commons CC-BY licence. The version selection and dissemination processes in academic publishing. All Frontiers 
current at the date of publication of journals are driven by researchers for researchers; therefore, they constitute 
this ebook is CC-BY 4.0. If the CC-BY a service to the scholarly community. At the same time, the Frontiers journal 
licence is updated, the licence granted 
by Frontiers is automatically updated series operates on a revolutionary invention, the tiered publishing system, 
to the new version. initially addressing specific communities of scholars, and gradually climbing 
When exercising any right under  up to broader public understanding, thus serving the interests of the lay 
the CC-BY licence, Frontiers must be society, too.
attributed as the original publisher  
of the article or ebook, as applicable. 
Authors have the responsibility of Dedication to quality
ensuring that any graphics or other 
materials which are the property of Each Frontiers article is a landmark of the highest quality, thanks to genuinely 
others may be included in the CC-BY collaborative interactions between authors and review editors, who include 
licence, but this should be checked some of the world’s best academicians. Research must be certified by peers 
before relying on the CC-BY licence 
to reproduce those materials. Any before entering a stream of knowledge that may eventually reach the public 
copyright notices relating to those - and shape society; therefore, Frontiers only applies the most rigorous 
materials must be complied with. 
and unbiased reviews. Frontiers revolutionizes research publishing by freely 
Copyright and source delivering the most outstanding research, evaluated with no bias from both 
acknowledgement notices may not  the academic and social point of view. By applying the most advanced 
be removed and must be displayed 
in any copy, derivative work or partial information technologies, Frontiers is catapulting scholarly publishing into  
copy which includes the elements  a new generation.
in question. 
All copyright, and all rights therein,  
are protected by national and What are Frontiers Research Topics? 
international copyright laws. The 
above represents a summary only. Frontiers Research Topics are very popular trademarks of the Frontiers 
For further information please read journals series: they are collections of at least ten articles, all centered  
Frontiers’ Conditions for Website Use 
and Copyright Statement, and the on a particular subject. With their unique mix of varied contributions from  
applicable CC-BY licence. Original Research to Review Articles, Frontiers Research Topics unify the 
most influential researchers, the latest key findings and historical advances  
in a hot research area.
ISSN 1664-8714 
ISBN 978-2-8325-2853-2 
DOI 10.3389/978-2-8325-2853-2
Find out more on how to host your own Frontiers Research Topic or 
contribute to one as an author by contacting the Frontiers editorial office: 
frontiersin.org/about/contact
Frontiers in Ecology and Evolution 1 frontiersin.org
August 2023
Challenges and opportunities in 
orchid ecology and conservation
Topic editors
Pavel Kindlmann — Charles University, Czechia
Tiiu Kull — Estonian University of Life Sciences, Estonia
Dennis Whigham — Smithsonian Institution, United States
Citation
Kindlmann, P., Kull, T., Whigham, D., eds. (2023). Challenges and opportunities in 
orchid ecology and conservation. Lausanne: Frontiers Media SA. 
doi: 10.3389/978-2-8325-2853-2
Frontiers in Ecology and Evolution 2 frontiersin.org
August 2023
Table of 05 Editorial: Challenges and opportunities in orchid ecology and conservation
contents Pavel Kindlmann, Tiiu Kull and Dennis Whigham
08 Range Size and Niche Breadth as Predictors of 
Climate-Induced Habitat Change in Epipactis (Orchidaceae)
Alexandra Evans and Hans Jacquemyn
19 Orchid diversity along an altitudinal gradient in the central 
Balkans
Vladan Djordjević, Spyros Tsiftsis, Pavel Kindlmann and 
Vladimir Stevanović
34 Dominant Dendrobium officinale mycorrhizal partners vary 
among habitats and strongly induce seed germination in vitro
Liyue Zhang, Kento Rammitsu, Kenshi Tetsuka, Tomohisa Yukawa and 
Yuki Ogura-Tsujita
46 Delimiting species in the taxonomically challenging orchid 
section Pseudophrys: Bayesian analyses of genetic and 
phenotypic data
Nina Joffard, Bruno Buatois, Véronique Arnal, Errol Véla, 
Claudine Montgelard and Bertrand Schatz
59 Shade and drought increase fungal contribution to partially 
mycoheterotrophic terrestrial orchids Goodyera pubescens 
and Tipularia discolor
Melissa K. McCormick, Kerry L. Good, Thomas J. Mozdzer and 
Dennis F. Whigham
70 Host tree species effects on long-term persistence of 
epiphytic orchid populations
Adriana Ramírez-Martínez, Tamara Ticktin and Demetria Mondragon
82 Will Greenland be the last refuge for the continental 
European small-white orchid?Niche modeling of future 
distribution of Pseudorchis albida
Marta Kolanowska, Sławomir Nowak and Agnieszka Rewicz
96 Taxonomic revision of Sobralia section Racemosae Brieger 
(Sobralieae, Orchidaceae)
Przemyslaw Baranow, Dariusz Szlachetko and Pavel Kindlmann
123 The effect of habitat transformation on a twig epiphytic 
orchid: Evidence from population dynamics
Nhora Helena Ospina-Calderón, Raymond L. Tremblay, 
Alba Marina Torres and Nicola S. Flanagan
Frontiers in Ecology and Evolution 3 frontiersin.org
August 2023
135 Floral and genetic divergence across environmental gradients 
is moderated by inter-population gene flow in Platanthera 
dilatata (Orchidaceae)
Lisa E. Wallace and Marlin L. Bowles
151 Diversity and specificity of orchid mycorrhizal fungi in a 
leafless epiphytic orchid, Dendrophylax lindenii and the 
potential role of fungi in shaping its fine-scale distribution
Lynnaun J. A. N. Johnson, Michael E. Kane, Lawrence W. Zettler and 
Gregory M. Mueller
Frontiers in Ecology and Evolution 4 frontiersin.org
TYPE Editorial
PUBLISHED 14 June 2023
DOI 10.3389/fevo.2023.1226614
OPEN ACCESS
EDITED AND REVIEWED BY
Marco Girardello,
Joint Research Centre, Italy
*CORRESPONDENCE
Pavel Kindlmann
pavel.kindlmann@centrum.cz
RECEIVED 21 May 2023
ACCEPTED 05 June 2023
PUBLISHED 14 June 2023
CITATION
Kindlmann P, Kull T and Whigham D (2023)
Editorial: Challenges and opportunities
in orchid ecology and conservation.
Front. Ecol. Evol. 11:1226614.
doi: 10.3389/fevo.2023.1226614
COPYRIGHT
© 2023 Kindlmann, Kull and Whigham. This
is an open-access article distributed under
the terms of the Creative Commons
Attribution License (CC BY). The use,
distribution or reproduction in other
forums is permitted, provided the original
author(s) and the copyright owner(s) are
credited and that the original publication in
this journal is cited, in accordance with
accepted academic practice. No use,
distribution or reproduction is permitted
which does not comply with these terms.
Frontiers in Ecology and EvolutionEditorial: Challenges and
opportunities in orchid
ecology and conservation
Pavel Kindlmann1*, Tiiu Kull2 and Dennis Whigham3
1Institute for Environmental Studies, Charles University, Prague, Czechia, 2Department of Biodiversity
and Nature Tourism, Estonian University of Life Sciences, Tartu, Estonia, 3Smithsonian Environmental
Research Center, Edgewater, MD, United States
KEYWORDS
orchids, conservation, ecology, survival, managementEditorial on the Research Topic
Challenges and opportunities in orchid ecology and conservationUnderstanding diversity patterns and how they are affected by global change are topics
of active discussion in biodiversity research. In response to species declines, it is important
to not only understand patterns of diversity but also develop a knowledge base for use in
species conservation. We still do not know, for example, the abiotic and biotic
requirements for population persistence for most species.
Orchid ecology and conservation are the subjects of this Research Topic. We focus on
orchids because the family has the most species and more than 50% of the species that have
been assessed fall into one or more risk categories. Given the large number of orchid
species, relatively few have been studied in detail. As a result, it is difficult to determine the
best approach for conserving species. Given the increasing threats to orchids globally, the
editors chose to focus on orchid ecology and conservation and the contributing authors
have provided a range of relevant topics.Orchid-fungal interactions are the focus of
three papers
Most orchids are mixotrophic, indicating that they obtain resources from fungal
interactions as well as photosynthesis. Orchid responses to changes in environmental
conditions have rarely been investigated, especially in terms of orchid-fungal interactions.
McCormick et al. experimentally manipulated light and soil moisture for two terrestrial
species and used isotopes to compare changes in carbon and nitrogen. They found that
reductions in light and soil moisture increased the dependence of both species on fungal
carbon and nitrogen.
Zhang et al. identified orchid mycorrhizal fungi (OMF) associated with Dendrobium
officinale, an orchid of medicinal value. Almost 84% of the OMF identified from plants at
six sites were in the Tulasnellaceae and Serendipitaceae families and the relative abundance
of the two fungi varied between plants that grew on rocks versus plants on trees. They
demonstrated that two of the fungi supported the germination and growth of Dendrobium,01 frontiersin.org
5
Kindlmann et al. 10.3389/fevo.2023.1226614providing evidence that there are differences among OMF in their
ability to support germination and growth. They suggested that
future research should focus on the use of in situ seed baiting as a
method for obtaining OMF from protocorms that are most likely to
support the early growth stages of orchids in nature.
Like terrestrial species, epiphytic orchids interact with
mycorrhiza. Johnson et al. identified the mycorrhiza associated
with the Ghost Orchid (Dendrophylax lindenii) and other epiphytic
orchids. They also compared the fungi on the bark of trees that had
the Ghost Orchid with bark from trees where the orchid did not
occur. They found that the fungus associated with Dendrophylax
was very specific and was a species of Ceratobasidium that was not
found in other epiphytes. Furthermore, they found that plants
grown in the lab had a lower abundance of Ceratobasidium than
plants that occurred naturally. Their results provide evidence that
the distribution of fungi influences the distribution of the
Ghost Orchid.Surprisingly, taxonomy had the
second-highest number of
contributions
Likely the result of the rapid development of powerful
computers and sophisticated genetic and molecular biology
methods, taxonomy is becoming a Cinderella in systematic
research, including orchids. An increased knowledge of orchid
identity is, however, necessary to support ecological and
conservation research.
Baranow et al. revised the Sobralia, section Racemosae, a large
and diverse genus that can be divided into four sections and some
informal species groups based mainly on inflorescence architecture.
The section Racemosae has species with an elongated inflorescence
with distinct internodes, but the species are often similar and easily
misidentified, especially with herbarium specimens. Baranow et al.
present species’ morphological characteristics, keys for identification,
ecological data, and distribution maps. They describe a new species,
Sobralia gambitana, and a neotype for S. hoppii Schltr. is proposed.
Tools that can integrate genetic and phenotypic data in
taxonomic studies have been recently developed and were used by
Joffard et al. to investigate species in the genus Pseudophrys. Using
an approach termed iBPP they identified four groups of species
rather than 12 and they merged two groups of species. They
demonstrated that phenotypic data are particularly informative in
section Pseudophrys, and the approach that they used improves
species identification. They recommended that an integrative
taxonomic approach holds great promise for conducting
taxonomic revisions in other orchid groups.Climate change, a globally important
topic, was the focus of two papers
Evans and Jacquemyn examined the impact of climate change
on 14 Epipactis species with a focus on species that are habitatFrontiers in Ecology and Evolution 02
6
specialists or generalists. Species with a wide distribution are more
capable of shifting habitats but only if they can fully expand into
habitats at the leading edge of their distributions. This study
provides valuable insights into how terrestrial orchid species with
differing niche breadths may respond to climate change.
Kolanowska et al. investigated the impact of climate change on
the future distribution of the small-white orchid (Pseudorchis
albida). The niche model that they used predicted that although
the number of suitable niches will increase significantly in
Greenland, suitable habitats will severely decline in continental
Europe. Importantly, their research indicated that global warming
might have an opposite effect on the pollinators of P. albida because
of insect habitat loss, but some pollinators are expected to remain
within the orchid’s potential geographical range, supporting its
long-term survival.The remaining four papers are
examples of topics that are relevant to
a more complete understanding of
orchid ecology and conservation
“Can orchids occur in landscapes that have been modified by
human activities”? That question is the topic addressed by Ospina-
Calderón et al. They studied the distribution of epiphytes in
undisturbed forests in the Andes and their distribution on shade
trees in coffee plantations and trees in a grassland matrix. They
collected data over 2 years and constructed demographic transition
matrices with transition probabilities calculated using the Bayesian
approach. Population growth rates were higher on trees in coffee
plantations compared with forests. Although the orchids also
occurred on trees in the grassland matrix, the authors suggested
that those populations represented a temporal phase that would not
be sustainable.
Wallace and Bowles explored the topic of genetic variation as a
function of gene flow in Spiranthes dilitata, a widespread species in
Alaska. They found evidence for small-scale genetic variation
associated with different habitats and differences in the ability of
pollinators to pollinate different morphotypes. This research
provided clear evidence that evolution in orchids can occur at
spatially small scales and can be influenced by pollinators.
Ramıŕez-Martıńez et al., like Wallace and Bowles, found that
differences in species performance can operate at small scales in
response to habitat conditions. They compared the population
dynamics of two epiphytic species in Mexico that occurred on
deciduous and semi-deciduous trees. It was demonstrated that in
years with normal rainfall, there were no differences in plant
performance, but during dry years, Alamania punicea was more
vulnerable to drying conditions—most likely because it has smaller
pseudobulbs that have less storage capacity. This research provides
evidence that climate change will potentially influence the
population dynamics of epiphytic orchids.
Djordjevic et al. sampled orchids along an elevation gradient in
the Balkans, with a focus on the belowground features of the
different species and their pollination. Results showed that speciesfrontiersin.org
Kindlmann et al. 10.3389/fevo.2023.1226614diversity peaked at 900–1,000 m, with variations in distribution
patterns for different life history traits and habitat types. Deceptive
orchids were most abundant at lower and mid-elevations. By
contrast, rewarding orchids were more common at mid to high
elevations. This study demonstrates that data that link orchid
species to habitats are important for conservation efforts.Author contributions
All authors listed have made a substantial, direct, and intellectual
contribution to the work and approved it for publication.Acknowledgments
We thank all contributors for submitting their research to make
this Research Topic diverse and informative. We also acknowledge
all the peer reviewers for providing constructive guidance to theFrontiers in Ecology and Evolution 03
7
authors—their contributions were crucial in promoting the rigor
and diversity of this Research Topic.Conflict of interest
The authors declare that the research was conducted in the
absence of any commercial or financial relationships that could be
construed as a potential conflict of interest.Publisher’s note
All claims expressed in this article are solely those of the authors
and do not necessarily represent those of their affiliated
organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
claim that may be made by its manufacturer, is not guaranteed or
endorsed by the publisher.frontiersin.org
fevo-10-894616 April 12, 2022 Time: 16:15 # 1
ORIGINAL RESEARCH
published: 15 April 2022
doi: 10.3389/fevo.2022.894616
Range Size and Niche Breadth as
Predictors of Climate-Induced
Habitat Change in Epipactis
(Orchidaceae)
Alexandra Evans* and Hans Jacquemyn
Plant Conservation and Population Biology, Department of Biology, Katholieke Universiteit Leuven, Leuven, Belgium
While there is mounting evidence that ongoing changes in the climate system are shifting
species ranges poleward and to higher altitudes, responses to climate change vary
considerably between species. In general, it can be expected that species responses
to climate change largely depend on how broad their ecological niches are, but
evidence is still scant. In this study, we investigated the effects of predicted future
Edited by: climate change on the availability of suitable habitat for 14 Epipactis (Orchidaceae)
Pavel Kindlmann,
Charles University, Czechia species, and tested whether habitat specialists would experience greater changes
Reviewed by: in the extent of their habitats than habitat generalists. We used Maxent to model
Vladan Djordjević, the ecological niche of each species in terms of climate, soil, elevation and land-
University of Belgrade, Serbia use and projected it onto climate scenarios predicted for 2061–2080. To test the
Przemyslaw Baranow,
University of Gdańsk, Poland hypothesis that temperate terrestrial orchid species with small ranges or small niche
*Correspondence: breadths may be at greater risk under climate change than species with wide
Alexandra Evans ranges or large niche breadths, we related niche breadth in both geographic and
alexandra.evans@kuleuven.be
environmental space to changes in size and location of suitable habitat. The habitat
Specialty section: distributions of half of the species shifted northwards in future projections. The area
This article was submitted to of suitable habitat increased for eight species but decreased for the remaining six
Conservation and Restoration
Ecology, species. If expansion at the leading edge of the distribution was not possible, the
a section of the journal area of suitable habitat decreased for 12 species. Species with wide niche breadth
Frontiers in Ecology and Evolution in geographic space experienced greater northwards expansions and higher habitat
Received: 11 March 2022
Accepted: suitability scores than species with small niche breadth. Niche breadth in environmental25 March 2022
Published: 15 April 2022 space was not significantly related to change in habitat distribution. Overall, these
Citation: results indicate that terrestrial orchid species with a wide distribution will be more
Evans A and Jacquemyn H (2022) capable of shifting their distributions under climate change than species with a limited
Range Size and Niche Breadth as
Predictors of Climate-Induced Habitat distribution, but only if they are fully able to expand into habitats at the leading edge of
Change in Epipactis (Orchidaceae). their distributions.
Front. Ecol. Evol. 10:894616.
doi: 10.3389/fevo.2022.894616 Keywords: climate change, ecological niche, ENMTools, Epipactis, Maxent, range size
Frontiers in Ecology and Evolution | www.frontiersin.org 81 April 2022 | Volume 10 | Article 894616
fevo-10-894616 April 12, 2022 Time: 16:15 # 2
Evans and Jacquemyn Climate Change and Epipactis Distributions
INTRODUCTION Djordjević and Tsiftsis, 2022). Species traits related to growth
and reproduction in a habitat, such as root system and
Climate plays an important role in the distribution of plant and pollination, can affect spatial distribution. For example, wide
animal species and in light of the global climate crisis, the effects spatial distributions of orchids in the Czech Republic were
of changing climate on plant species distributions is a prominent associated with a rhizomatous root system (Štípková et al., 2021),
topic in ecology (Chen et al., 2011; Tayleur et al., 2015; Lehikoinen and the wide variety of pollinators utilised by the terrestrial
and Virkkala, 2016). In order to survive climate change, species orchid Epipactis helleborine is likely an important contributor
must either shift their range limits to environments that are able to its large range and ability to colonise various habitats
to support them or adapt to the new conditions in their current (Rewicz et al., 2017).
environments (Thuiller, 2007; Kelly and Goulden, 2008; Scheffers Recently, it has become clear that weather conditions can
et al., 2016; Ash et al., 2017). Predicting how a species’ suitable have a strong impact on orchid population dynamics, suggesting
habitat alters due to climate change is necessary when planning its that changing climatic conditions have the potential to affect
long-term conservation, but can be difficult because of the wide the geographic distribution of orchids. For example, climatic
variety of habitat needs and tolerances among species. changes during the last three decades have been shown to
Species differ in their responses to climate change based have a positive effect on the survival of the terrestrial orchid
on how broad their ecological niches are (Thuiller et al., Himantoglossum hircinum at the northern edge of its population
2005). Previous research has already shown that species within in the United Kingdom (van der Meer et al., 2016) and
a genus can vary considerably in habitat preferences and warmer winter weather conditions have also been shown to
distributions (Brown et al., 1996; Grossenbacher and Whittall, be beneficial to German populations of this species (Pfeifer
2011; Anacker and Strauss, 2014; Duffy and Jacquemyn, 2019). et al., 2006). Williams et al. (2015) demonstrated that the
Habitat generalists tend to have wider ranges of conditions population dynamics, vital rates and reproduction of the lady
where they can survive, grow and reproduce and are therefore orchid (Orchis purpurea) at the northern edge of its distribution
assumed to be more adaptable to environmental change (Marvier were affected by seasonal temperature and precipitation and,
et al., 2004; Thuiller et al., 2005). Specialist species, on the other specifically, that milder winters and wetter springs were beneficial
hand, tend to have more specific environmental requirements for its population growth. These results suggest that a warmer
and therefore can only occupy a narrow ecological niche. It climate will generally benefit orchids at the northern edges of
is expected that species which have narrow temperature or their distributions. A recent modelling study has indeed shown
precipitation tolerances are the most likely to be affected by that predicted changes in climatic conditions increased habitat
climate change (Slatyer et al., 2013). However, empirical evidence suitability available to threeOrchis species by 2050 at the northern
is still limited (Shay et al., 2021) and for many species we do not edge of their distribution (Evans et al., 2020). However, given
know the factors that limit their distributions, whether leading that these species showed very similar distribution areas and
edge expansions are sustainable, or how these species respond to often co-occur, such a generalisation may not be appropriate
climate change. Gaining a better understanding of the physical and it remains unclear how differences in range size or
factors underlying the distribution of organisms is crucial to environmental niche breadth predict vulnerability under global
predict how species will respond to climate change (Hagsater change (Shay et al., 2021).
et al., 1996; Tsiftsis et al., 2008). In this study, we tested the hypothesis that orchid species
Although orchids are generally considered rare and have with small ranges or small niche breadths may be at greater risk
small population sizes (Tremblay et al., 2005; Otero and under climate change than species with wide ranges or large
Flanagan, 2006; Shefferson et al., 2020), there is often large niche breadths. We used the orchid genus Epipactis as a study
variation in range size and environmental tolerance between system. Epipactis is a widespread genus occurring throughout the
species, both within and among orchid genera (McCormick and European and Asian continents with 37 species according to the
Jacquemyn, 2014; Evans and Jacquemyn, 2020). What drives The Euro+Med Plantbase Project (2022) although the results of
variation in orchid species range size is not well known, but phylogenetic research in recent years has brought into question
is likely a combination of factors including niche breadth, the status of many species (Sramkó et al., 2019; Bateman, 2020).
species age, niche availability and range position (Sheth et al., Previous research has shown that among fourteen European
2020). Previous research has shown that orchid species vary in Epipactis species, range size differed by more than three orders of
their dependence on specific abiotic environmental conditions, magnitude between species with the smallest and largest ranges
with some species being limited primarily by temperature and (Evans and Jacquemyn, 2020). The distribution of small-range
precipitation (McCormick et al., 2009; Djordjević et al., 2016; species was strongly associated with local habitat conditions and
Evans et al., 2020) and others being limited more by local landscape structure, while that of large-range species was more
growth conditions related to bedrock and soil (Bowles et al., associated with climatic conditions (Evans and Jacquemyn, 2020).
2005; Tsiftsis et al., 2008; Bunch et al., 2013). Consequently, However, whether the habitat distributions of generalist species
specialist orchid species are often associated with the habitat are more strongly affected by climate change than small-range,
types that arise from the specific combinations of these abiotic specialist species, is yet unknown. Specifically, we investigated
characteristics, from coastal dunes to temperate forests, and the how the habitat of the same fourteen Epipactis species would be
spatial extent of these habitats therefore can limit the range affected by changes in temperature and precipitation in Europe
of the species they support (McCormick and Jacquemyn, 2014; predicted for 2061–2080, and assessed whether species with
Frontiers in Ecology and Evolution | www.frontiersin.org 92 April 2022 | Volume 10 | Article 894616
fevo-10-894616 April 12, 2022 Time: 16:15 # 3
Evans and Jacquemyn Climate Change and Epipactis Distributions
small ranges or narrow ecological niches would suffer greater because they are the most representative of the mean climate
changes in size and latitudinal position of habitat than species of an area, and are therefore appropriate for a continent-wide
with large ranges. distribution study such as this. We also obtained the mean
annual temperature and annual precipitation rasters predicted
for the years 2061–2080 predicted by two Shared Socio-economic
MATERIALS AND METHODS Pathways (SSPs), SSP 2-4.5 and SSP 5-8.5 from WorldClim.3
SSP 2-4.5 models the climate in a scenario where greenhouse
Study Species and Occurrence Data gas emissions are at their highest (∼44 GT CO2) in 2040
The genus Epipactis contains a large number of terrestrial orchids and then decrease to 9.6 GT in 2100, while in SSP 5-8.5,
which vary greatly in distribution area and habitat type (Sramkó emissions increase steeply until the year 2080 (∼130 GT) before
et al., 2019; Evans and Jacquemyn, 2020). Some species (e.g., starting to stabilise and decrease (Riahi et al., 2017). Maps of
E. dunensis and E. albensis) have very localised distributions the distribution of temperature and precipitation values in
and are restricted to particular habitats such as coastal dunes Europe were created by calculating the mean temperature for
and beech forests, whereas others (e.g., E. helleborine and each cell of a 50 km2 cell grid of Europe and summarising
E. atrorubens) are widespread and can tolerate a relatively the values in QGIS.
wide range of habitat conditions. There are several ecotypes The other seven variables used were the same as those used to
of E. helleborine that can be found in specific habitats such as model Epipactis species in Evans and Jacquemyn (2020). These
coastal dunes and forests (Jacquemyn et al., 2018). Species are include the first two components of two PCAs run on two
autogamous, allogamous, or facultative allogamous (Claessens topsoil datasets (physical and biochemical measures) acquired
and Kleynen, 2011; Brys and Jacquemyn, 2016). The numerous through the European Soil Data Centre (ESDAC) (Hiederer,
seeds produced by Epipactis species are very small, dispersed by 2013; Ballabio et al., 2019), dominant bedrock from the ESDAC
wind, and rely on the presence of mycorrhizal fungi in the soil to database (Van Liedekerke et al., 2006), Corine Land Cover (CLC)
germinate and establish (Bidartondo and Read, 2008; Smith and from the Copernicus programme of the European Environmental
Read, 2010; McCormick and Jacquemyn, 2014; Jacquemyn et al., Program (Heymann, 1994) and elevation (Amatulli et al.,
2018; Xing et al., 2020). Differences in mycorrhizal communities 2018). All raster processing was performed in RStudio v4.0.2
between localities may contribute to reproductive isolation and (R Core Team, 2021).
spatial distribution of Epipactis species and populations (Ogura-
Tsujita and Yukawa, 2008; Jacquemyn et al., 2016, 2018; but see Ecological Niche Modelling
Těšitelová et al., 2012). Defining and quantitatively comparing plant niches can be
Records of each species’ occurrence from 2000 to 2020 on achieved using ecological niche models (ENMs). Ecological niche
the continent of Europe were obtained from the online database modelling has been applied successfully to numerous species
GBIF1 (Supplementary Material). We discarded records with to investigate ecological niches and to assess the impacts of
missing GIS coordinates, ambiguous species identification or climate change and land use on species ranges (Guisan and
with coordinates with a spatial resolution lower than 100 m. Thuiller, 2005). We used the programme Maxent v3.4.1. (Phillips
This resulted in between 31 (Epipactis lusitanica) and 45,354 et al., 2017) to model the effects of predicted climate change
(E. helleborine) occurrences per species. Records for each on species’ habitats. Maxent is a popular ENM tool that uses
species were aggregated into 10 km2 grid cells to reduce species occurrence data and environmental rasters to calculate a
the effects of spatial clustering resulting from sampling bias, Gibbs value for each pixel of the study area, or the probability
by extracting the centre coordinates of each grid cell in that the pixel has suitable habitat conditions for the species
which the species was recorded (Supplementary Table 1). (Phillips, 2005) and performs well in comparison to other
Processing of occurrence data was performed in QGIS v3.4.9 modelling methods (Elith et al., 2006; Phillips and Dudík, 2008;
(QGIS Development Team, 2019). Valavi et al., 2021). Maxent creates habitat suitability maps
over the study area from these data, as well as a table of
Ecogeographic Variables the contribution of each predictor variable to the distribution
Previous studies have shown that land cover, bedrock, of suitable habitat for each species. The choice of Maxent
precipitation, and temperature are important variables predicting settings was informed by Barbet-Massin et al. (2012) and
the distributions of some Epipactis species (Tsiftsis et al., 2008; Merow et al. (2013). Each model was run using a random
Djordjević et al., 2016; Evans and Jacquemyn, 2020). We seed and 100 bootstrap replicates with 75% of the data used
therefore used nine raster-format predictor variables with to train the model and 25% to test it. The rest of the
<0.5 correlation with one other. Two of the 19 bioclimatic settings were left as the default (convergence threshold of
variables available at the WorldClim v2 online database (Fick 0.00001, regularisation threshold of 1 and a maximum of 10,000
and Hijmans, 20172) were used in our model, mean annual background points) and allowed for linear, quadratic, product
temperature and annual precipitation, projected for the near- and hinge features to be chosen automatically, producing a
present climate (1970–2000). These two variables were chosen cloglog output. The models were run for the current climatic
features and projected onto the SSP climate data to produce
1www.GBIF.org
2https://www.worldclim.org/data/worldclim21.html 3https://www.worldclim.org/data/cmip6/cmip6climate.html
Frontiers in Ecology and Evolution | www.frontiersin.org 130 April 2022 | Volume 10 | Article 894616
fevo-10-894616 April 12, 2022 Time: 16:15 # 4
Evans and Jacquemyn Climate Change and Epipactis Distributions
separate environmental niche map outputs for the current and using Kruskal–Wallis tests. All analyses were performed in
future climate conditions. RStudio v4.0.2 (R Core Team, 2021).
Data Analysis
For each habitat suitability map, the mean Gibbs value with RESULTS
standard error was calculated for every latitudinal interval of
0.5 decimal degree of the study area (Europe) using the Zonal The mean temperature in continental Europe will increase from
Statistics tool in QGIS. We ran a Wilcoxon signed rank test a near-current mean of 9.21 0.10◦± C (standard error) to
on these mean Gibbs values multiplied by their corresponding 12.16± 0.09◦C for SSP 2-4.5 and to 13.42± 0.09◦C for SSP 5-8.5
latitudes to test whether the suitable habitat of each species will (Figure 1) predicted for the years 2061–2080. The mean annual
shift in latitude in future climate scenarios. Before running the precipitation will increase slightly in the future projections, with a
Wilcoxon tests, the set of Gibbs values for each climate scenario current mean of 742.74± 5.86–761.95± 5.86 mm3 for SSP 2-4.5
was centred by dividing by the mean to eliminate the influence of and 761.43± 5.93 mm3 for SSP 5-8.5.
different mean Gibbs values between climate scenarios and test The mean current habitat suitability or probability of
only for shifts in latitude. occurrence (Gibbs p-value) predicted by the Maxent model
The continuous probabilistic maps produced by Maxent were ranged from 0.0026 ± 0.0006 for E. lusitanica to 0.29 ± 0.022
converted into binary presence maps using the Maximum Test for E. helleborine. When the model was projected for the
Sensitivity plus Specificity (MTSS) value of each species as a climate of 2061–2080, there was no significant difference in
threshold. The pixels with a Gibbs value of greater than the mean habitat suitability between the current climatic conditions
MTSS were extracted and plotted as a new map for each species, and either of the two future climate scenarios (χ2 = 0.18,
with each pixel representing the species being present at that p-value = 0.91). Although the mean species’ habitat suitability
location. The total numbers of pixels occupied by these habitat did not change, when species were tested individually, the
distribution maps were compared between current and future habitat suitability of E. helleborine, lusitanica, phyllanthes, and
climate scenarios. The overlapping pixels between the current tremolsii significantly increased under both SSP scenarios, and
and future distributions (i.e., pixels for which occurrence equalled E. albensis increased significantly for SSP 5-8.5 (Table 1;
one for both maps) of each species were extracted and counted see Supplementary Table 2 for mean Gibbs values). Seven
to provide a measurement of the area suitable for a species if it species (E. albensis, fageticola, kleinii, leptochila, microphylla,
were unable to expand into any newly available areas created by muelleri, and tremolsii) demonstrated significant northwards
future climate change. shifts in their habitat distributions in both future climate
Mean species occurrence per climate scenario and per species, scenarios (Table 2 and Supplementary Material for illustration
for both the continuous Gibbs values and the number of pixels of individual range shifts).
occupied of the binary maps, were compared between the climate The area of suitable habitat available (pixels where the Gibbs
scenarios using Kruskal–Wallis and Dunn tests with a Holm p-value was above the species’ MTSS threshold) increased
correction or ANOVA and Tukey tests if the data were normally for eight species (E. albensis, dunensis, fageticola helleborine,
distributed. lusitanica, microphylla, phyllanthes, and tremolsii) in the
We calculated Levins’ B2 values of niche breadth in future scenarios, but decreased for the remaining six species
geographic (B2geo) and environmental space (B2env) for each (E. atrorubens, kleinii, leptochila, muelleri, palustris, and
species using the functions raster.breadth and env.breadth, purpurata; Table 3). For species that responded positively to
respectively, in the ENMTools 1.0.5 R package (Warren et al., the climatic changes, the increase in habitat ranged between
2021). B2 ranges from 0 to 1, with values closer to 0 5 and 1000% (E. dunensis and E. lusitanica), while for those
representing narrow (specialised) niche breadth and values that responded negatively, decrease in habitat area ranged
closer to 1 representing wide (generalised) niche breadth. between 5% (E. kleinii, muelleri, and palustris) and 88%
Finally, we investigated whether species niche breadth predicts (E. purpurata).
changes in distribution in response to climate change by Overlap in habitat distribution areas between current and
dividing the range change of a species from current to future climate scenarios was fairly high, ranging from 57 to
future scenarios, converting to the proportional change for 100% for SSP 2-4.5 and 33 to 100% for SSP 5-8.5, except for
each species, and comparing these values to each species’ E. purpurata which showed notably low overlap (16 and 4%
B2 value using ordinary least-squares regression. Ordinary for the two scenarios, respectively; see Supplementary Data for
least-squares regression was also used to compare the Levins’ range values). The change in habitat area experienced by species
B2 values between geographic and environmental space. The if they would not be able to track the climate in the future
difference in niche breadth between species with positive decreased by up to 95% (E. purpurata) for all except two species,
range changes and negative range changes in response to E. lusitanica and E. phyllanthes, which showed no decrease in
habitat change was investigated using Kruskal–Wallis tests. distribution area (100% overlap, only expansion).
The effect of mating system on response to climate change The niche breadth values of Levins’ B2geo (in geographic space)
was tested by comparing the mean changes in latitudinal ranged from 0.39 for E. fageticola to 0.85 for E. palustris, while
habitat distribution and proportional range size between B2env ranged from 0.16 for E. dunensis to 0.90 for E. lusitanica
autogamous, allogamous and facultative autogamous species, (see Supplementary Material). B2 values in geographic and
Frontiers in Ecology and Evolution | www.frontiersin.org 141 April 2022 | Volume 10 | Article 894616
fevo-10-894616 April 12, 2022 Time: 16:15 # 5
Evans and Jacquemyn Climate Change and Epipactis Distributions
FIGURE 1 | Distribution of mean annual temperature and annual precipitation in Europe predicted for the current climate, and projected to occur in the years
2061–2080 under two Shared Socio-economic Pathways (SSP2-4.5 and SSP5-8.5). Bars represent the number of 50 km grid squares of Europe with a
corresponding mean temperature or precipitation value, and dotted lines represent the mean values.
TABLE 1 | Change in mean Gibbs values (habitat suitability) for Epipactis species from current to future (2061–2080) climate scenarios (SSP 2-4.5 and SSP 5-8.5) and
results of Dunn tests comparing current and future mean Gibbs p-values (showing only results for significant differences, in bold, and marginally significant
differences, in italics).
Species Change mean Gibbs p Current – SSP 2-4.5 Current – SSP 5-8.5
SSP 2-4.5 SSP 5-8.5 Z p Z p
E. albensis 0.0025 0.0037 −2.1573 0.0620 −3.1508 0.0049
E. atrorubens −0.0196 −0.0378
E. dunensis 0.0007 0.0059 −0.3820 0.7024 −2.3288 0.0596
E. fageticola 0.0003 0.0002
E. helleborine 0.0645 0.0791 −2.2945 0.0435 −2.5807 0.0296
E. kleinii −0.0002 −0.0008
E. leptochila −0.0038 −0.0060
E. lusitanica 0.0064 0.0102 −3.4435 0.0011 −4.9054 <0.0001
E. microphylla 0.0024 0.0008
E. muelleri −0.0014 −0.0065
E. palustris 0.0193 0.0136
E. phyllanthes 0.0056 0.0084 −2.7392 0.0123 −3.8179 0.0004
E. purpurata −0.0113 −0.0182
E. tremolsii 0.0044 0.0065 −2.5263 0.0231 −3.7593 0.0005
Frontiers in Ecology and Evolution | www.frontiersin.org 152 April 2022 | Volume 10 | Article 894616
fevo-10-894616 April 12, 2022 Time: 16:15 # 6
Evans and Jacquemyn Climate Change and Epipactis Distributions
TABLE 2 | Latitudinal shifts in habitat distribution from current to future climate scenarios and results of Wilcoxon tests of differences in habitat distributions (showing only
results for significant differences).
Species Change in mean Gibbs p*latitude Current – SSP 2-4.5 Current – SSP 5-8.5
SSP 2-4.5 SSP 5-8.5 W p W p
E. albensis 86792.3 179258.2 1016 0.0091 918 0.0019
E. atrorubens 370209.9 558240.5
E. dunensis 7321.7 3531.9
E. fageticola 45463.3 112859.5 986 0.0058 985 0.0057
E. helleborine 319826.6 459016.5
E. kleinii 49895.4 165549.1 998 0.0069 904 0.0015
E. leptochila 155493.8 204799.6 1121 0.0369 1087 0.0241
E. lusitanica 105798.9 199849.8
E. microphylla 243820.9 385056.3 1076 0.0208 1110 0.0322
E. muelleri 246742.1 397285.9 1063 0.0175 1081 0.0222
E. palustris 332171.1 506155.0
E. phyllanthes 91731.5 164969.6
E. purpurata 433625.6 748751.2
E. tremolsii 253596.0 412612.4 802 0.0002 829 0.0004
environmental spaces were not correlated with one another DISCUSSION
(p-value = 0.74). The means Gibbs value of habitat suitability
was positively correlated with B2geo for both current (R2 = 0.33, In this study we investigated how the distribution of suitable
F1,12 = 7.52, p-value = 0.012) and future (SSP 2-4.5: R2 = 0.50, habitat of Epipactis species would be affected by predicted climate
F 21,12 = 13.81, p-value = 0.0029; SSP 5-8.5: R = 0.53, change and whether species with small ranges or narrow niche
F1,12 = 15.81, p-value= 0.0020) climate projections (Figure 2A). breadths are at greater risk from climate change than species with
Similarly, B2geo had a positive relationship with range size for wide ranges or large niche breadths. Our results showed that
current (R2 = 0.32, F1,12 = 7.11, p-value = 0.021) and future the habitat available increased on the leading (northern) edge
(SSP 2-4.5: R2 = 0.53, F1,12 = 15.41, p-value = 0.0020; SSP of the distribution for half of the species but decreased for the
5-8.5: R2 = 0.52, F1,12 = 13.19, p-value = 0.0034) climate
projections (Figure 2B). Species with higher B2geo values also
experienced greater changes in Gibbs values between current
and future climate scenarios (Figure 2C; SSP 2-4.5: R2 = 0.68, TABLE 3 | Changes in area of suitable habitat above the Maximum Test Sensitivity
F 17.93, p-value 0.0039; SSP 5-8.5: R2 0.43, and Specificity threshold of each species from the current climate conditions to1,12 = = = future climate scenarios (SSP 2-4.5 and SSP 5-8.5), as well as changes if species
F1,12 = 7.09, p-value = 0.032). There was a positive relationship are unable to expand their ranges into the climatic envelope of future scenarios
between B2geo and the change in mean Gibbs value between (climate tracking).
current climate and SSP 2-4.5 multiplied by latitude (Figure 2D;
R2 = 0.24, F1,12 = 4.99, p-value 0.045), indicating that Species Proportional Proportional range change=
species with higher B2 values would experience a greater range change without trackinggeo
northwards shift in suitable habitat than those with low B2geo SSP 2-4.5 SSP 5-8.5 SSP 2-4.5 SSP 5-8.5
values, if they were able to track the suitable climate. This
was also marginally significant for SSP 5-8.5 (R2 = 0.18, E. albensis 1.9094 3.4189 −0.0025 −0.0214
F1,12 = 3.77, p-value = 0.076). No comparisons involving B2env E. atrorubens −0.3250 −0.5481 −0.4259 −0.6711
were significant at α = 0.05, but marginally significant positive E. dunensis 0.0554 0.5265 −0.0019 −0.0136
relationships were detected between B2env and proportional E. fageticola 0.2203 0.5554 −0.0059 −0.2330
range change (proportional to the species’ current range) from E. helleborine 0.2001 0.1391 −0.1960 −0.2970
current to future climate scenarios (SSP 2-4.5: R2 = 0.31, E. kleinii −0.0476 −0.3597 −0.1228 −0.4138
F1 12 = 4.20, p-value = 0.086; SSP 5-8.5: R2 = 0.17, F1 12 = 3.64, E. leptochila −0.3257 −0.4843 −0.3862 −0.5520, ,
p-value = 0.081). Species with higher B2env also showed E. lusitanica 5.9117 10.8262 < 0.0001 < 0.0001
some evidence for experiencing a greater northwards shift in E. microphylla 0.3052 0.2582 −0.2469 −0.4614
suitable habitat from current to SSP 5-8.5 climate (R2 = 0.16, E. muelleri −0.0467 −0.2476 −0.3005 −0.5037
F1,12 = 3.77, p-value = 0.076). The mean proportional range E. palustris −0.0479 −0.1842 −0.2636 −0.4495
change (with and without tracking) and change in latitudinal E. phyllanthes 1.5237 2.3245 < 0.0001 < 0.0001
habitat distribution in response to climate change in either SSP E. purpurata −0.6798 −0.8866 −0.8357 −0.9555
scenario were not significantly different between mating systems E. tremolsii 0.6798 1.0058 −0.1054 −0.1888
(p-value > 0.05). Range changes are reported as proportional to the current range.
Frontiers in Ecology and Evolution | www.frontiersin.org 163 April 2022 | Volume 10 | Article 894616
fevo-10-894616 April 12, 2022 Time: 16:15 # 7
Evans and Jacquemyn Climate Change and Epipactis Distributions
FIGURE 2 | Relationships between Levins’ B2 measure for niche breadth in geographic space and (A) the mean Gibbs p-value (habitat suitability), (B) the spatial
range of suitable habitat, (C) change in mean Gibbs p from current to future climate scenarios, and (D) change in latitudinal habitat distribution of 14 Epipactis
species. Two climate scenarios were used, SSP 2-4.5 and SSP 5-8.5 to predict the climate for 2061–2080.
remaining species, and decreased for all but two species if climate values for the majority of the species individually was predicted
tracking was not possible. Levins’ B2 metric for niche breadth in to increase in 2061–2080. The area of suitable habitat increased
geographic space was highly correlated with the spatial extent into the north for some species and decreased in the south for
and mean Gibbs value (habitat suitability) of species habitat most species in the future, resulting in a mean northern shift in
distributions and species with a higher B2 value were predicted to habitat. When expansion into the north (climate tracking) was
experience a greater northwards expansion in response to climate restricted, the area of habitat decreased by up to 95% for all except
change. We did not detect significant effects of Levins’ B2 in two small-range species.
environmental space, although there was marginally significant Despite the expectation that species with narrow
patterns similar to those of B2 in geographic space. environmental tolerances are most threatened by climate
change, in the case of Epipactis, the habitats of most of the small-
range localised species that we investigated were predicted to
Impact of Climate Change on the increase with future climate change. Some northern hemisphere
Distribution of Epipactis in Europe herbaceous species benefit from increased temperatures at the
Although there was no change in the mean Gibbs value of the 14 northern edge of their distribution through increased population
species between current and future climate scenarios, the Gibbs growth, which in turn can lead to an increase in geographic
Frontiers in Ecology and Evolution | www.frontiersin.org 174 April 2022 | Volume 10 | Article 894616
fevo-10-894616 April 12, 2022 Time: 16:15 # 8
Evans and Jacquemyn Climate Change and Epipactis Distributions
range at this edge (Bremer and Jongejans, 2009). This includes The distributions of all investigated species, even generalists,
orchids such as H. hircinum where climatic changes in the tended to lag behind climate warming, without being able to
United Kingdom were shown to be partially responsible for fully track the upward shift in suitable climate resulting in a
the species’ expansion between 1991 and 2001, as well as for range contraction, in both our study and Geppert et al. (2020).
projected future scenarios (van der Meer et al., 2016). Similarly, Plant species inhabiting forests may be somewhat buffered
Ongaro et al. (2018) predicted that the habitat range for nine from the effects of climate warming (De Frenne et al., 2013;
orchid species will increase by 2070 on the island of Sardinia, Zellweger et al., 2020) and those in grasslands tend to have high
although the probability of presence in the newly colonised thermal ranges because of the lack of this buffering (Geppert
habitats was not predicted to increase. However, Vogt-Schilb et al., 2020). Similarly, Vogt-Schilb et al. (2015) found higher
et al. (2015) found that the distributions of many orchids in rates of disappearances in wetland orchid species in Western
Western Europe have declined in the last two decades due to Europe than those in grassland, and more appearances in forest.
land-use change, particularly in the northern parts of their There did not seem to be any clear pattern in response to
distributions. If land-use continues to change in more northern climate change and habitat-use in our species (other than with
latitudes, this could limit the areas into which Epipactis can move niche breadth), with woodland species such as E. muelleri
in response to climate change. The results of our study provide decreasing in suitable habitat and E. microphylla increasing.
further support for the potential for orchid ranges to increase at However, our study used a broad-scale specification of land cover,
their leading edges in response to climate change, but go further while more may be revealed at a finer resolution that captures
to demonstrate that this does not necessarily mean an increase microclimate gradients.
in available habitat, particularly if they cannot move into the An important caveat to consider when carrying out niche
northern habitats in time. breadth studies is that the metric used to describe niche breadth
can greatly affect the results. Levin’s B2 is the reciprocal of
Simpson’s diversity index (Levins, 1968) and has been a popular
Testing the Range Size Vulnerability metric of niche breadth for more than 50 years. However, it
Hypothesis has been noted that the traditional calculation of this metric is
A widely supported paradigm is that the maximum range limits in geographic space (see Peterson and Soberón, 2012) for more
of a species coincide with its ecological niche limits and that, on geographic and environmental space) and more accurately
given the opportunity to disperse, range limits will shift to match represents the “flatness” of the geographic distribution of suitable
the geographical extent of the niche under climate change (Reed habitat (Warren et al., 2019), which may be a useful measure of
et al., 2021; Shay et al., 2021). The pattern of species with spatial habitat-use but is not niche breadth in terms of specificity
wider niche breadths demonstrating greater latitudinal shifts in of resource-use. This is demonstrated clearly in our results, where
response to climate change has been documented in a number of B2 in geographic space was consistently correlated with measures
terrestrial plant taxa (Thuiller et al., 2005; Alarcón and Cavieres, of the size of the habitat distribution and the mean Gibbs value.
2018). This was also demonstrated in orchids by Geppert et al. B2 in environmental space as proposed by Warren et al. (2019)
(2020) where the distributions of generalist orchid species and and developed in Warren et al. (2021) filters the geographic
those inhabiting forests and semi-natural grasslands tended to habitat suitability distribution through the set of environmental
be less affected than the more specialised and rare species in variables that was used to create the Maxent model, resulting in
subalpine, natural grassland and wetland habitats, whose rear and a B2 value that is closer to the concept of niche breadth as being
leading edges shifted upward. This corresponds with our finding the specificity in environmental conditions of a species’ habitat.
that Epipactis species with wider niche breadths (generalists) It is important to note, however, that although closer to what
experience greater change in habitat area in response to changing we understand to be niche breadth, B2 in environmental space is
climate than specialists. If we were to assign species to the groups still dependent on the availability of habitats in geographic space
of specialist and generalist based on current spatial ranges and (Petraitis, 1979; Warren et al., 2019). Although the values of B2 in
values of B2 in geographic space, E. fageticola, albensis, kleinii, geographic and environmental space were not correlated, B2 in
and lusitanica would be considered the most specialist (relative environmental space showed some evidence for having the same
to the other species in this study), followed by E. tremolsii, relationship with habitat changes as B2 in geographic space. This
dunensis, and leptochila as moderately specialist (Supplementary indicates that B2 in environmental space has the potential to be
Table 2). E. muelleri, microphylla, and purpurata are moderately a useful representation of niche breadth for Epipactis in Europe,
generalist, while E. helleborine, palustris, and atrorubens could but further study is required to conclude this.
be considered generalists. However, E. purpurata had a low B2
value but a fairly large spatial distribution and E. phyllanthes
a high B2 value and small range, which is in contrast to this Other Factors Contributing to Range
pattern. E. purpurata was predicted to experience a significant Shifts
decrease in suitable habitat under climate change which may Although the abiotic characteristics discussed here are important
indicate that species with relatively large current ranges may for predicting orchid ranges, biotic interactions and species-
still have fairly narrow niches which are nonetheless currently specific characteristics are also essential contributors to
common in the environment, but are under threat from changing the realised niche, and including these interactions can
climate. improve the accuracy and performance of niche models
Frontiers in Ecology and Evolution | www.frontiersin.org 185 April 2022 | Volume 10 | Article 894616
fevo-10-894616 April 12, 2022 Time: 16:15 # 9
Evans and Jacquemyn Climate Change and Epipactis Distributions
(Flores-Tolentino et al., 2020; Phillips et al., 2020). Orchids rely in light of individual patterns in addition to drawing general
on insect pollinators and mycorrhizal fungi to reproduce and conclusions. This is demonstrated in the contrast between mean
germinate (Rasmussen, 2008; McCormick and Jacquemyn, 2014). change in Gibbs value (no change) and the change in Gibbs value
As with other pollinator-reliant plants, allogamous Epipactis will for individual species, where a number of species were predicted
only persist and track shifting climate if their pollinators are to increase in the future and, the increase in habitat area for
also able to disperse (Benning and Moeller, 2019; Shay et al., some species and the decrease for other. This disparity between
2021), such as has been predicted for a Neotropical orchid bee general vs. individual species patterns has also been demonstrated
which is predicted to persist and increase its habitat range under in Geppert et al. (2020), who showed high interspecific variation
future climate change (Silva et al., 2015). Mating system was not among orchids grouped by habitat preference. We also show how
significantly associated with changes in habitat distribution in some species with large areas of habitat such as E. purpurata
response to climate change, indicating that in the specific case should not be considered immune to the detrimental effects
of these species, autogamous and allogamous species did differ of future climate change as they may suffer considerable range
in response to predicted climate change. This is not surprising, reductions if they are not able to sufficiently disperse northwards.
considering that mating system was not significantly associated
with niche breadth or range size for Epipactis species in previous
studies (Evans and Jacquemyn, 2020) and niche breadth in DATA AVAILABILITY STATEMENT
geographic space is directly linked to range size. The presence
of soil microbes has also been linked to the ability of plants The original contributions presented in the study are included
to expand into newly available habitats (David et al., 2019; in the article/Supplementary Material, further inquiries can be
Bueno de Mesquita et al., 2020; Benning and Moeller, 2021; directed to the corresponding author.
Shay et al., 2021). The diversity of mycorrhizal fungi is linked
to latitudinal gradients for some orchid species (Duffy et al.,
2019), but it is unclear whether the northern shifts in orchid AUTHOR CONTRIBUTIONS
distributions will be supported by the lower diversity of fungi in
more northern latitudes. Our understanding and predictions of AE and HJ: conceptualisation, methodology, writing, review
orchid distribution changes in response to climate change would and editing, and funding acquisition. AE: formal analysis, data
be greatly improved with the addition of pollinator and fungal curation, and visualisation. HJ: supervision. Both authors have
symbiont distributional data. read and agreed to the published version of the manuscript.
Implications for Conservation FUNDING
Studies that model the ecological niches of species are useful
for conservation planning, particularly for identifying newly This work was supported by the Flemish Fund for Scientific
accessible areas available to plants (more so than predicting range Research (Grant: G093019N).
contractions) and assessing the risk faced by populations as a
consequence of their range size (Schwartz, 2012; Shay et al.,
2021). This study provides a useful estimate of new areas into SUPPLEMENTARY MATERIAL
which Epipactis can expand, which in conjunction with more
information on predicted land change in these areas, could be The Supplementary Material for this article can be found
used in conservation schemes to allow the genus to flourish under online at: https://www.frontiersin.org/articles/10.3389/fevo.2022.
climate change. However, it is important to assess the results 894616/full#supplementary-material
REFERENCES Ballabio, C., Lugato, E., Fernández-Ugalde, O., Orgiazzi, A., Jones, A., Borrelli,
P., et al. (2019). Mapping LUCAS topsoil chemical properties at European
Alarcón, D., and Cavieres, L. A. (2018). Relationships between ecological niche scale using Gaussian process regression. Geoderma 355:113912. doi: 10.1016/
and expected shifts in elevation and latitude due to climate change in South j.geoderma.2019.113912
American temperate forest plants. J. Biogeography 45, 2272–2287. doi: 10.1111/ Barbet-Massin, M., Jiguet, F., Albert, C. H., and Thuiller, W. (2012). Selecting
jbi.13377 pseudo-absences for species distribution models: how, where and how many?
Amatulli, G., Domisch, S., Tuanmu, M.-N., Parmentier, B., Ranipeta, A., Malczyk, Methods Ecol. Evol. 3, 327–338. doi: 10.1111/j.2041-210x.2011.00172.x
J., et al. (2018). A suite of global, cross-scale topographic variables for Bateman, R. M. (2020). Implications of next-generation sequencing for the
environmental and biodiversity modeling. Sci. Data 5:180040. doi: 10.1038/ systematics and evolution of the terrestrial orchid genus Epipactis, with
sdata.2018.40 particular reference to the British Isles. Kew Bull. 75:4.
Anacker, B. L., and Strauss, S. Y. (2014). The geography and ecology Benning, J. W., and Moeller, D. A. (2019). Maladaptation beyond a geographic
of plant speciation: range overlap and niche divergence in sister range limit driven by antagonistic and mutualistic biotic interactions across an
species. Proc. R. Soc. B: Biol. Sci. 281:20132980. doi: 10.1098/rspb.2013. abiotic gradient. Evolution 73, 2044–2059. doi: 10.1111/evo.13836
2980 Benning, J. W., and Moeller, D. A. (2021). Microbes, mutualism, and range
Ash, J. D., Givnish, T. J., and Waller, D. M. (2017). Tracking lags in historical plant margins: testing the fitness consequences of soil microbial communities across
species’ shifts in relation to regional climate change. Global Change Biol. 23, and beyond a native plant’s range. New Phytol. 229, 2886–2900. doi: 10.1111/
1305–1315. doi: 10.1111/gcb.13429 nph.17102
Frontiers in Ecology and Evolution | www.frontiersin.org 196 April 2022 | Volume 10 | Article 894616
fevo-10-894616 April 12, 2022 Time: 16:15 # 10
Evans and Jacquemyn Climate Change and Epipactis Distributions
Bidartondo, M. I., and Read, D. J. (2008). Fungal specificity bottlenecks during misestimated if biotic interactions are ignored: the case of the orchid Laelia
orchid germination and development. Mol. Ecol. 17, 3707–3716. doi: 10.1111/j. speciosa. Sci. Rep. 10:9542. doi: 10.1038/s41598-020-63638-9
1365-294X.2008.03848.x Geppert, C., Perazza, G., Wilson, R. J., Bertolli, A., Prosser, F., Melchiori, G., et al.
Bowles, M., Zettler, L., Bell, T., and Kelsey, P. (2005). Relationships between (2020). Consistent population declines but idiosyncratic range shifts in Alpine
soil characteristics, distribution and restoration potential of the federal orchids under global change. Nat. Commun. 11:5835. doi: 10.1038/s41467-020-
threatened eastern prairie fringed orchid, Platanthera leucophaea (Nutt.) Lindl. 19680-2
Am. Midland Natural. 154, 273–286. doi: 10.1674/0003-0031(2005)154[0273: Grossenbacher, D. L., and Whittall, J. B. (2011). Increased floral divergence in
rbscda]2.0.co;2 sympatric monkeyflowers. Evolution 65, 2712–2718. doi: 10.1111/j.1558-5646.
Bremer, P., and Jongejans, E. (2009). Frost and forest stand effects on the 2011.01306.x
population dynamics of Asplenium scolopendrium. Popul Ecol. 52:211. doi: 10. Guisan, A., and Thuiller, W. (2005). Predicting species distribution: offering more
1007/s10144-009-0143-7 than simple habitat models. Ecol. Lett. 8, 993–1009. doi: 10.1111/j.1461-0248.
Brown, J. H., Stevens, G. C., and Kaufman, D. M. (1996). The geographic range: 2005.00792.x
size, shape, boundaries, and internal structure. Annu. Rev. Ecol. Syst. 27, Hagsater, E., Dumont, V., and Pridgeon, A. M. (1996). Orchids: Status Survey and
597–623. doi: 10.1146/annurev.ecolsys.27.1.597 Conservation Action Plan. Gland: IUCN.
Brys, R., and Jacquemyn, H. (2016). Severe outbreeding and inbreeding depression Heymann, Y. (1994). CORINE Land Cover: Technical Guide. Luxembourg.: Office
maintain mating system differentiation in Epipactis (Orchidaceae). J. Evol. Biol. for Official Publ. of the Europ. Communities.
29, 352–359. doi: 10.1111/jeb.12787 Hiederer, R. (2013). Mapping Soil Properties for Europe—Spatial Representation of
Bueno de Mesquita, C. P., Sartwell, S. A., Schmidt, S. K., and Suding, K. N. Soil Database Attributes. Luxembourg: Scientific and Technical Research series.
(2020). Growing-season length and soil microbes influence the performance Publications Office of the European Union, EUR26082EN.
of a generalist bunchgrass beyond its current range. Ecology 101:e03095. doi: Jacquemyn, H., Kort, H. D., Broeck, A. V., and Brys, R. (2018). Immigrant and
10.1002/ecy.3095 extrinsic hybrid seed inviability contribute to reproductive isolation between
Bunch, W. D., Cowden, C. C., Wurzburger, N., and Shefferson, R. P. (2013). forest and dune ecotypes of Epipactis helleborine (Orchidaceae). Oikos 127,
Geography and soil chemistry drive the distribution of fungal associations in 73–84. doi: 10.1111/oik.04329
lady’s slipper orchid, Cypripedium acaule. Botany 91, 850–856. doi: 10.1139/cjb- Jacquemyn, H., Waud, M., Lievens, B., and Brys, R. (2016). Differences in
2013-0079 mycorrhizal communities between Epipactis palustris, E. helleborine and its
Chen, I.-C., Hill, J. K., Ohlemuller, R., Roy, D. B., and Thomas, C. D. (2011). Rapid presumed sister species E. neerlandica. Annals of Botany 118, 105–114. doi:
range shifts of species associated with high levels of climate warming. Science 10.1093/aob/mcw015
333, 1024–1026. doi: 10.1126/science.1206432 Kelly, A. E., and Goulden, M. L. (2008). Rapid shifts in plant distribution with
Claessens, J., and Kleynen, J. (2011). The Flower of the European Orchid: form and recent climate change. Proc. Natl. Acad. Sci. U S A. 105, 11823–11826. doi:
Function. Voerendaal: Jean Claessens & Jacques Kleynen. 10.1073/pnas.0802891105
David, A. S., Quintana-Ascencio, P. F., Menges, E. S., Thapa-Magar, K. B., Lehikoinen, A., and Virkkala, R. (2016). North by north-west: climate change
Afkhami, M. E., and Searcy, C. A. (2019). Soil microbiomes underlie population and directions of density shifts in birds. Global Change Biol. 22, 1121–1129.
persistence of an endangered plant species.Am.Nat. 194, 488–494. doi: 10.1086/ doi: 10.1111/gcb.13150
704684 Levins, R. (1968). Evolution in Changing Environments: Some Theoretical
De Frenne, P., Rodríguez-Sánchez, F., Coomes, D. A., Baeten, L., Verstraeten, Explorations. Princeton, NJ: Princeton University Press.
G., Vellend, M., et al. (2013). Microclimate moderates plant responses to Marvier, M., Kareiva, P., and Neubert, M. G. (2004). Habitat destruction,
macroclimate warming. Proc. Natl. Acad. Sci. U S A. 110, 18561–18565. doi: fragmentation, and disturbance promote invasion by habitat generalists in a
10.1073/pnas.1311190110 multispecies metapopulation. Risk Anal. 24, 869–878. doi: 10.1111/j.0272-4332.
Djordjević, V., and Tsiftsis, S. (2022). “The role of ecological factors in distribution 2004.00485.x
and abundance of terrestrial orchids,” in Orchids Phytochemistry, Biology and McCormick, M. K., and Jacquemyn, H. (2014). What constrains the distribution
Horticulture Reference Series in Phytochemistry, eds J.-M. Mérillon and H. of orchid populations? New Phytol. 202, 392–400. doi: 10.1111/nph.
Kodja (Cham: Springer International Publishing), 3–72. doi: 10.1007/978-3- 12639
030-38392-3_4 McCormick, M. K., Whigham, D. F., O’Neill, J. P., Becker, J. J., Werner,
Djordjević, V., Tsiftsis, S., Lakušić, D., Jovanović, S., and Stevanović, V. (2016). S., Rasmussen, H. N., et al. (2009). Abundance and distribution of
Factors affecting the distribution and abundance of orchids in grasslands Corallorhiza odontorhiza reflect variations in climate and ectomycorrhizae.
and herbaceous wetlands. Systematics Biodiversity 14, 355–370. doi: 10.1080/ Ecol. Monographs 79, 619–635. doi: 10.1890/08-0729.1
14772000.2016.1151468 Merow, C., Smith, M. J., and Silander, J. A. Jr. (2013). A practical guide to MaxEnt
Duffy, K. J., and Jacquemyn, H. (2019). Climate change increases ecogeographical for modeling species’ distributions: what it does, and why inputs and settings
isolation between closely related plants. J. Ecol. 107, 167–177. doi: 10.1111/ matter. Ecography 36, 1058–1069. doi: 10.1111/j.1600-0587.2013.07872.x
1365-2745.13032 Ogura-Tsujita, Y., and Yukawa, T. (2008). Epipactis helleborine shows strong
Duffy, K. J., Waud, M., Schatz, B., Petanidou, T., and Jacquemyn, H. (2019). mycorrhizal preference towards ectomycorrhizal fungi with contrasting
Latitudinal variation in mycorrhizal diversity associated with a European geographic distributions in Japan. Mycorrhiza 18, 331–338. doi: 10.1007/
orchid. J. Biogeography 46, 968–980. doi: 10.1111/jbi.13548 s00572-008-0187-0
Elith, J., Graham, C. H., Anderson, R. P., Dudík, M., Ferrier, S., Guisan, A., Ongaro, S., Martellos, S., Bacaro, G., Agostini, A. D., Cogoni, A., and Cortis,
et al. (2006). Novel methods improve prediction of species’ distributions from P. (2018). Distributional pattern of Sardinian orchids under a climate change
occurrence data. Ecography 29, 129–151. scenario. Community Ecol. 19, 223–232. doi: 10.1556/168.2018.19.3.3
Evans, A., and Jacquemyn, H. (2020). Impact of mating system on range size Otero, J. T., and Flanagan, N. S. (2006). Orchid diversity – beyond deception.
and niche breadth in Epipactis (Orchidaceae). Ann. Bot. 126, 1203–1214. doi: Trends Ecol. Evol. 21, 64–65. doi: 10.1016/j.tree.2005.11.016
10.1093/aob/mcaa142 Peterson, A. T., and Soberón, J. (2012). Species distribution modeling and
Evans, A., Janssens, S., and Jacquemyn, H. (2020). Impact of climate change on ecological niche modeling: getting the concepts right. NatCon 10, 102–107.
the distribution of four closely related orchis (Orchidaceae) species. Diversity doi: 10.2166/wst.2007.494
12:312. doi: 10.3390/d12080312 Petraitis, P. S. (1979). Likelihood measures of niche breadth and overlap. Ecology
Fick, S. E., and Hijmans, R. J. (2017). WorldClim 2: new 1-km spatial resolution 60, 703–710. doi: 10.2307/1936607
climate surfaces for global land areas. Int. J. Climatol. 37, 4302–4315. doi: Pfeifer, M., Wiegand, K., Heinrich, W., and Jetschke, G. (2006). Long-term
10.1002/joc.5086 demographic fluctuations in an orchid species driven by weather: implications
Flores-Tolentino, M., García-Valdés, R., Saénz-Romero, C., Ávila-Díaz, I., Paz, for conservation planning. J. Appl. Ecol. 43, 313–324. doi: 10.1111/j.1365-2664.
H., and Lopez-Toledo, L. (2020). Distribution and conservation of species is 2006.01148.x
Frontiers in Ecology and Evolution | www.frontiersin.org 1170 April 2022 | Volume 10 | Article 894616
fevo-10-894616 April 12, 2022 Time: 16:15 # 11
Evans and Jacquemyn Climate Change and Epipactis Distributions
Phillips, R. D., Peakall, R., Niet, T., van der, and Johnson, S. D. (2020). Niche The Euro+Med Plantbase Project (2022). The Euro+Med PlantBase - the
perspectives on plant–pollinator interactions. Trends Plant Sci. 25, 779–793. Information Resource for Euro-Mediterranean Plant Diversity. Available
doi: 10.1016/j.tplants.2020.03.009 online at: http://ww2.bgbm.org/EuroPlusMed/PTaxonDetail.asp?NameCache=
Phillips, S. J. (2005). A brief tutorial on maxent. ATT Research 190, 231–259. Epipactis&PTRefFk=8000000 (accessed March 22, 2022).
Phillips, S. J., and Dudík, M. (2008). Modeling of species distributions with Maxent: Thuiller, W. (2007). Climate change and the ecologist. Nature 448, 550–552. doi:
new extensions and a comprehensive evaluation. Ecography 31, 161–175. doi: 10.1038/448550a
10.1111/j.0906-7590.2008.5203.x Thuiller, W., Lavorel, S., and Araujo, M. B. (2005). Niche properties
Phillips, S. J., Anderson, R. P., Dudík, M., Schapire, R. E., and Blair, M. E. (2017). and geographical extent as predictors of species sensitivity to climate
Opening the black box: an open-source release of Maxent. Ecography 40, change. Global Ecol. Biogeography 14, 347–357. doi: 10.1111/j.1466-822x.2005.
887–893. doi: 10.1111/ecog.03049 00162.x
QGIS Development Team (2019). QGIS Geographic Information System. (QGIS Tremblay, R. L., Ackerman, J. D., Zimmerman, J. K., and Calvo, R. N. (2005).
Association). Variation in sexual reproduction in orchids and its evolutionary consequences:
R Core Team (2021). R: A Language and Environment for Statistical Computing. a spasmodic journey to diversification. Biol. J. Linnean Soc. 84, 1–54. doi:
Vienna: R Core Team. 10.1111/j.1095-8312.2004.00400.x
Rasmussen, H. (2008). Terrestrial Orchids: from Seed to Mycotrophic Plant. Tsiftsis, S., Tsiripidis, I., Karagiannakidou, V., and Alifragis, D. (2008). Niche
Terrestrial Orchids: from Seed to Mycotrophic Plant. Cambridge: Cambridge analysis and conservation of the orchids of east Macedonia (NE Greece). Acta
University Press. Oecol. 33, 27–35. doi: 10.1016/j.actao.2007.08.001
Reed, P. B., Peterson, M. L., Pfeifer-Meister, L. E., Morris, W. F., Doak, D. F., Roy, Valavi, R., Guillera-Arroita, G., Lahoz-Monfort, J. J., and Elith, J. (2021). Predictive
B. A., et al. (2021). Climate manipulations differentially affect plant population performance of presence-only species distribution models: a benchmark study
dynamics within versus beyond northern range limits. J. Ecol. 109, 664–675. with reproducible code. Ecol. Monogr. 92:e01486.
doi: 10.1111/1365-2745.13494 van der Meer, S., Jacquemyn, H., Carey, P. D., and Jongejans, E. (2016).
Rewicz, A., Jaskuła, R., Rewicz, T., and Tończyk, G. (2017). Pollinator diversity Recent range expansion of a terrestrial orchid corresponds with climate-driven
and reproductive success of Epipactis helleborine (L.) Crantz (Orchidaceae) in variation in its population dynamics. Oecologia 181, 435–448. doi: 10.1007/
anthropogenic and natural habitats. PeerJ 5:e3159. doi: 10.7717/peerj.3159 s00442-016-3592-7
Riahi, K., van Vuuren, D. P., Kriegler, E., Edmonds, J., O’Neill, B. C., Fujimori, Van Liedekerke, M., Jones, A., and Panagos, P. (2006). ESDBv2 Raster Library—
S., et al. (2017). The shared socioeconomic pathways and their energy, land A Set of Rasters Derived from the European Soil Database Distribution v2. 0.
use, and greenhouse gas emissions implications: an overview. Global Environ. European Commission and the European Soil Bureau Network, CDROM, EUR
Change 42, 153–168. 19945.
Scheffers, B. R., De Meester, L., Bridge, T. C., Hoffmann, A. A., Pandolfi, J. M., Vogt-Schilb, H., Munoz, F., Richard, F., and Schatz, B. (2015). Recent declines
Corlett, R. T., et al. (2016). The broad footprint of climate change from genes to and range changes of orchids in Western Europe (France, Belgium and
biomes to people. Science 354:aaf7671. doi: 10.1126/science.aaf7671 Luxembourg). Biol. Conservation 190, 133–141. doi: 10.1016/j.biocon.2015.
Schwartz, M. W. (2012). Using niche models with climate projections to inform 05.002
conservation management decisions. Biol. Conserv. 155, 149–156. doi: 10.1016/ Warren, D. L., Beaumont, L. J., Dinnage, R., and Baumgartner, J. B. (2019). New
j.biocon.2012.06.011 methods for measuring ENM breadth and overlap in environmental space.
Shay, J. E., Pennington, L. K., Montiel-Molina, J. A. M., Toews, D., Hendrickson, Ecography 42, 444–446. doi: 10.1111/ecog.03900
B., and Sexton, J. (2021). Rules of plant species ranges: applications for Warren, D. L., Matzke, N. J., Cardillo, M., Baumgartner, J. B., Beaumont, L. J.,
conservation strategies. Front. Ecol. Evol. 9:700962. doi: 10.3389/fevo.2021. Turelli, M., et al. (2021). ENMTools 1.0: an R package for comparative
700962 ecological biogeography. Ecography 44, 504–511. doi: 10.1111/ecog.
Shefferson, R. P., Jacquemyn, H., Kull, T., and Hutchings, M. J. (2020). The 05485
demography of terrestrial orchids: life history, population dynamics and Williams, J. L., Jacquemyn, H., Ochocki, B. M., Brys, R., and Miller, T. E. (2015).
conservation. Botan. J. Linnean Soc. 192, 315–332. doi: 10.1093/botlinnean/ Life history evolution under climate change and its influence on the population
boz084 dynamics of a long-lived plant. J. Ecol. 103, 798–808. doi: 10.1111/1365-2745.
Sheth, S. N., Morueta-Holme, N., and Angert, A. L. (2020). Determinants of 12369
geographic range size in plants. New Phytol. 226, 650–665. doi: 10.1111/nph. Xing, X., Gao, Y., Zhao, Z., Waud, M., Duffy, K. J., Selosse, M., et al. (2020).
16406 Similarity in mycorrhizal communities associating with two widespread
Silva, D. P., Macêdo, A. C. B. A., Ascher, J. S., and De Marco, P. (2015). Range terrestrial orchids decays with distance. J. Biogeography 47, 421–433. doi: 10.
increase of a neotropical orchid bee under future scenarios of climate change. 1111/jbi.13728
J. Insect. Conserv. 19, 901–910. doi: 10.1007/s10841-015-9807-0 Zellweger, F., De Frenne, P., Lenoir, J., Vangansbeke, P., Verheyen, K.,
Slatyer, R. A., Hirst, M., and Sexton, J. P. (2013). Niche breadth predicts Bernhardt-Römermann, M., et al. (2020). Forest microclimate dynamics drive
geographical range size: a general ecological pattern. Ecol. Lett. 16, 1104–1114. plant responses to warming. Science 368, 772–775. doi: 10.1126/science.
doi: 10.1111/ele.12140 aba6880
Smith, S. E., and Read, D. J. (2010). Mycorrhizal Symbiosis. Cambridge, MA:
Academic press. Conflict of Interest: The authors declare that the research was conducted in the
Sramkó, G., Paun, O., Brandrud, M. K., Laczkó, L., Molnár, A., and Bateman, R. M. absence of any commercial or financial relationships that could be construed as a
(2019). Iterative allogamy–autogamy transitions drive actual and incipient potential conflict of interest.
speciation during the ongoing evolutionary radiation within the orchid genus
Epipactis (Orchidaceae). Ann. Bot. 124, 481–497. doi: 10.1093/aob/mcz103 Publisher’s Note: All claims expressed in this article are solely those of the authors
Štípková, Z., Tsiftsis, S., and Kindlmann, P. (2021). Distribution of orchids with and do not necessarily represent those of their affiliated organizations, or those of
different rooting systems in the czech republic. Plants 10:632. doi: 10.3390/ the publisher, the editors and the reviewers. Any product that may be evaluated in
plants10040632 this article, or claim that may be made by its manufacturer, is not guaranteed or
Tayleur, C., Caplat, P., Massimino, D., Johnston, A., Jonzén, N., Smith, H. G., endorsed by the publisher.
et al. (2015). Swedish birds are tracking temperature but not rainfall: evidence
from a decade of abundance changes. Global Ecol. Biogeography 24, 859–872. Copyright © 2022 Evans and Jacquemyn. This is an open-access article distributed
doi: 10.1111/geb.12308 under the terms of the Creative Commons Attribution License (CC BY). The use,
Těšitelová, T., Těšitel, J., Jersáková, J., Říhová, G., and Selosse, M. (2012). Symbiotic distribution or reproduction in other forums is permitted, provided the original
germination capability of four Epipactis species (Orchidaceae) is broader than author(s) and the copyright owner(s) are credited and that the original publication
expected from adult ecology. Am. J. Botany 99, 1020–1032. doi: 10.3732/ajb. in this journal is cited, in accordance with accepted academic practice. No use,
1100503 distribution or reproduction is permitted which does not comply with these terms.
Frontiers in Ecology and Evolution | www.frontiersin.org 1181 April 2022 | Volume 10 | Article 894616
fevo-10-929266 August 3, 2022 Time: 15:21 # 1
TYPE Original Research
PUBLISHED 09 August 2022
DOI 10.3389/fevo.2022.929266
Orchid diversity along an
OPEN ACCESS altitudinal gradient in the central
EDITED BY
Isabel Marques, Balkans
University of Lisbon, Portugal
REVIEWED BY
Hans Jacquemyn, Vladan Djordjević1*, Spyros Tsiftsis2, Pavel Kindlmann3 and
KU Leuven, Belgium Vladimir Stevanović1,4
Zhifeng Ding,
Institute of Zoology, Guangdong 1Faculty of Biology, Institute of Botany and Botanical Garden, University of Belgrade, Belgrade,
Academy of Science (CAS), China Serbia, 2Department of Forest and Natural Environment Sciences, International Hellenic University,
*CORRESPONDENCE Drama, Greece, 3Academy of Science of the Czech Republic, Global Change Research Institute,
Vladan Djordjević Brno, Czechia, 4Serbian Academy of Sciences and Arts, Belgrade, Serbia
vdjordjevic@bio.bg.ac.rs
SPECIALTY SECTION
This article was submitted to Understanding patterns of species diversity along an altitudinal gradient is the
Conservation and Restoration Ecology, major topic of much biogeographical and ecological research. The aim of
a section of the journal
Frontiers in Ecology and Evolution this study was to explore how richness and density of orchid species and
subspecies in terms of different categories of underground organ systems and
RECEIVED 26 April 2022
ACCEPTED 04 July 2022 pollination systems vary along an altitudinal gradient in the central Balkans.
PUBLISHED 09 August 2022 The altitudinal gradient of the study area was divided into 21 100-m vertical
CITATION intervals. Data were analyzed using both non-linear and linear regressions
Djordjević V, Tsiftsis S, Kindlmann P
and Stevanović V (2022) Orchid with three data sets (total orchids, orchids of forest habitats, orchids of
diversity along an altitudinal gradient non-forest habitats) in the case of species richness and three data sets
in the central Balkans.
Front. Ecol. Evol. 10:929266. (total orchids—total area, forest orchids—forest area, and orchids of non-
doi: 10.3389/fevo.2022.929266 forest habitats—non-forest area) in the case of species density. The results
COPYRIGHT showed a hump-shaped pattern of orchid richness and density, peaking
© 2022 Djordjević, Tsiftsis, Kindlmann at 900–1,000 m. The richness and density of orchids of forest habitats
and Stevanović. This is an open-access
article distributed under the terms of are generally slightly greater than the richness and density of orchids of
the Creative Commons Attribution non-forest habitats in lowland areas, whereas the orchids of herbaceous
License (CC BY). The use, distribution
or reproduction in other forums is vegetation types dominating at high altitudes. Tuberous orchids dominate
permitted, provided the original in low and mid-altitude areas, orchids with palmately lobed and fusiform
author(s) and the copyright owner(s)
are credited and that the original tubers (“intermediate orchids”) dominate at high altitudes, while rhizomatous
publication in this journal is cited, in orchids are predominate in mid-altitude forest stands. Both deceptive and
accordance with accepted academic self-pollinated orchids show a unimodal trend with a peak at mid-altitude
practice. No use, distribution or
reproduction is permitted which does areas. This study underlines the importance of low and mid-altitude areas for
not comply with these terms. the survival of deceptive orchids and the importance of mid- and high-altitude
areas for the survival of rewarding orchids. In addition, forest habitats at mid-
altitudes have been shown to be crucial for the survival of self-pollinated
orchids. The results suggest that the altitudinal patterns of orchid richness
and density in the central Balkans are determined by mechanisms related
to land area size and habitat cover, partially confirming the species-area
relationship (SAR) hypothesis. This study contributes significantly to a better
understanding of the potential impacts of habitat changes on orchid diversity,
thereby facilitating more effective conservation planning.
KEYWORDS
Orchidaceae, ecology, altitudinal patterns, distribution, life history strategies, species
richness, species diversity, Balkan Peninsula
Frontiers in Ecology and Evolution 01 frontiersin.org
19
fevo-10-929266 August 3, 2022 Time: 15:21 # 2
Djordjević et al. 10.3389/fevo.2022.929266
Introduction There are some studies and books that provide detailed
information on the altitudinal range of individual orchid species
The family Orchidaceae is one of the most species-rich in Europe or specific countries, suggesting that the altitudinal
families in the plant kingdom, with an estimated 26,000–28,000 range of the same species can vary considerably within the
species from 749 genera (Christenhusz and Byng, 2016; Chase range of its distribution (Baumann et al., 2006; Delforge, 2006;
et al., 2017). Although orchids grow in almost all terrestrial Jersáková et al., 2015). To date, many studies provide important
ecosystems, they are most diverse in the tropics and subtropics, information on how altitudinal gradients affect orchid species
where species of different life forms can be found. In Europe, richness, but most of them have been conducted for Asian
orchids are exclusively terrestrial, inhabiting both forest habitats (Acharya et al., 2011; Zhang et al., 2015a,b; Timsina et al.,
and herbaceous plant communities (Djordjević and Tsiftsis, 2021), American (Cardelús et al., 2006; Ackerman et al., 2007;
2022). Because of their germination limitation, mycorrhizal Štípková et al., 2016), and African countries (Jacquemyn et al.,
specificity, and pollinator specialization, many orchid species 2005b, 2007), while there are just few studies on orchid richness
are particularly vulnerable to environmental change (Waterman along the altitudinal gradient in Europe (Tsiftsis et al., 2019;
and Bidartondo, 2008; Swarts and Dixon, 2009). Intensive Štípková et al., 2020, 2021). The species-area relationship (SAR)
anthropogenic impacts resulting in habitat alteration and has been investigated for some countries in Asia (Acharya et al.,
loss have led to the extinction or decline in abundance 2011; Zhang et al., 2015a), whereas in Europe there are only
and distribution of many orchids (Kull and Hutchings, two studies (Štípková et al., 2020, 2021) that consider this
2006). Understanding patterns of orchid species richness and relationship by analysis of density. However, it has not been
abundance along the geographical and environmental gradients studied in detail how the area of specific habitats affects the
is a central goal of much ecological and biogeographical research patterns of orchid diversity along the altitudinal gradient.
(Tsiftsis et al., 2008; Acharya et al., 2011; Zhang et al., 2015a; In recent years, the diversity patterns of species classified in
Djordjević et al., 2016, 2020). In addition, knowledge of diversity different functional groups have been used to understand the
patterns along the altitudinal gradient and factors influencing relationships between these traits and environmental variation
these patterns can contribute not only to a better understanding (Laughlin et al., 2012; Taylor et al., 2021). There are several
of orchid ecology and distribution, but also to planning studies on the distribution of certain orchid life forms,
strategies necessary for a successful species conservation plan. including commonly terrestrial, epiphytic, and saprophytic
In general, there are two main patterns of species richness— orchids (Cardelús et al., 2006; Acharya et al., 2011; Zhang
an altitude relationship: a monotonic decrease in number et al., 2015a), while knowledge on the distribution patterns
of species with altitude; and a hump-shaped pattern with of orchid life forms in Europe is limited (Tsiftsis et al.,
the highest species number at mid-altitudes (Rahbek, 1995; 2019; Štípková et al., 2021). Species diversity patterns related
McCain and Grytnes, 2010; Timsina et al., 2021). Nearly half to specific life forms along gradients (e.g., altitude, latitude)
of the studies showed that the hump-shaped patterns are the not only may be useful from a basic ecology perspective,
most common ones, whereas other studies suggested either a but they can also contribute to a better understanding of
monotonic decrease or an increase in the number of species orchid evolutionary history, prediction of their distribution,
with altitude (McCain and Grytnes, 2010; Timsina et al., 2021). and effective orchid species conservation. Some studies have
Although several hypotheses have been proposed to explain focused particularly on the distribution and species richness of
patterns of orchid diversity along the altitudinal gradient, most orchids possessing certain floral traits and breeding systems,
studies address the influence of climatic factors, then the as well as pollination systems (Arroyo et al., 1982; Jacquemyn
mid-domain effect (MDE), while less attention has been paid et al., 2005b; Pellissier et al., 2010; Štípková et al., 2020).
to the species-area relationship (SAR). The climatic gradient Although it was found that the relative occurrence of food-
hypothesis predicts that species richness peaks at a particular deceptive orchids decreases with increasing altitude in the
altitude where a combination of growing conditions is optimal territory of Switzerland (Pellissier et al., 2010), there is a lack of
for the species. According to Colwell and Lees (2000), most knowledge on how orchid diversity patterns vary when it comes
species live in mid-altitude areas due to the geometric limit to other orchid pollination systems, including rewarding, self-
of the species’ range. This pattern, known as the “mid-domain pollinated and other deceptive orchids. Furthermore, there is a
effect” (MDE), results from random overlap of the altitudinal lack of knowledge about the relationship between altitude and
range of species (Colwell and Hurtt, 1994; Colwell et al., 2004; richness/density of orchids, which are characterized by different
Dunn et al., 2007). The concept of a species-area relationship life forms and pollination systems in different habitats (e.g.,
suggests that species richness varies depending on size of the forests and herbaceous plant communities) and regions.
area of a certain altitude range, i.e., that maximum species Although the Balkan Peninsula is one of the parts of Europe
richness occurs in the altitudinal zones that cover the largest with the highest number of orchid taxa (Djordjević et al., 2020),
area (Acharya et al., 2011; Karger et al., 2011; Trigas et al., the patterns of species richness and density along the altitudinal
2013). gradient in the central Balkans have not yet been explored.
Frontiers in Ecology and Evolution 02 frontiersin.org
20
fevo-10-929266 August 3, 2022 Time: 15:21 # 3
Djordjević et al. 10.3389/fevo.2022.929266
Therefore, our study aims to explore patterns of diversity of region, which belongs to the Dinaric mountain system. The
orchids along the altitudinal gradient in the central Balkans, with altitude ranges from 65 m (Šabac) to 2,154 m (Mokra Gora—
the goal of analyzing both the total orchid flora and the orchid Pogled). The climate in Serbia can be described as temperate-
flora of individual life forms and pollination systems. Special continental. The average annual temperature varies from 6.7◦C
attention had to be paid to the analysis of the influence of the in the coldest parts to 11.6◦C in the warmest parts, while
size of the area of individual altitudinal intervals on the patterns the average annual temperature in the areas above 1,500 m
of diversity, focusing on the area of specific habitat types (forest above sea level is about 3.0◦C. Annual precipitation varies from
and non-forest). Consequently, the importance of this study lies 726.4 mm in the lower-lying regions to about 1,500 mm in the
in the contribution to the knowledge of orchid life histories, mountainous areas of south-western Serbia (climatic data from
ecology and distribution, but also in the creation of a good the Hydrometeorological Service of the Republic of Serbia).
basis for more effective orchid conservation. We hypothesized In general, vegetation in the study area is structured
that spatial patterns of forest and non-forest habitats along according to climatic differentiation. In the northernmost parts
the altitudinal gradient affect orchid species diversity patterns. of western Serbia, near the Sava and Kolubara rivers, there
Moreover, we expected that orchids of different traits have are floodplain (Fraxino-Quercion roboris) forests, while in the
different diversity patterns as well. Based on the evolutionary rest of the study area (especially at low to medium altitudes)
development of the underground organs of orchids (Averyanov, oak (Quercion confertae and Quercion petraeo-cerridis) forests
1990; Tsiftsis et al., 2019), we assume that orchids with spheroid predominate. Mesophilous deciduous beech and hornbeam
or ovoid tubers dominate at lower altitudes because they (Fagion sylvaticae and Carpinion betuli) forests are predominant
generally tolerate drought and warmer conditions best. On the in the zone of middle altitudes, while coniferous (Vaccinio-
other hand, orchids with palmately lobed and fusiform tubers Piceetea) forests are found in the high-mountain regions. The
are assumed to dominate at higher altitudes because their origin density of forest cover in the study area is shown in Figure 1B.
is related to the emergence of colder climates and they have Western Serbia is geologically diverse, with a large occurrence
the best adaptations that allow them to grow in habitats with of carbonate and ultramafic rocks and various types of silicate
low temperatures and high humidity characteristic of highland rocks (Djordjević and Tsiftsis, 2019).
areas. Given the different pollination systems of orchids and the
studies already published (Pellissier et al., 2010; Štípková et al.,
2020), we expect that the richness of rewarding orchids is greater Data collection
than that of deceptive ones in high-altitude areas.
The main objectives of this study were: (i) to determine The total database consists of data on 55 orchid species and
altitudinal range size of individual orchids and compare subspecies recorded at 3,580 sites (Supplementary Table 1).
altitudinal range size and mean altitude of occurrence of orchid Data on 53 orchid species and subspecies from 2,610 sites were
species and subspecies with different life traits (underground collected during field observations between 1995 and 2021. In
organ systems, pollination systems); (ii) to analyze orchid addition, the dataset included published data on 48 species
species richness and density along the altitudinal gradient; (iii) and subspecies from 683 sites and herbarium data on 44 taxa
to determine how the richness and density of orchid species from 287 sites collected in the Herbarium of the University of
with different life traits (underground organ systems, pollination Belgrade (BEOU) and the Herbarium of the Natural History
systems) vary with altitude. Patterns of species richness and Museum in Belgrade (BEO). The number of sampling sites for
density along the altitudinal gradient were explored for the total each altitudinal interval is shown in Figure 2. This number does
orchid flora, as well as for the orchid flora recorded in forest and not include the sites we visited and did not find any orchids
non-forest habitats. there. Orchid taxa were identified according to Delforge (2006),
while Djordjević et al. (2021) was used for nomenclature. During
field surveys, geographic coordinates (longitude, latitude) and
Materials and methods altitude were determined by a Garmin eTrex 30 hand-held
GPS device in the WGS 84, while reliable data from published
Study area sources and herbarium collections were georeferenced using Ozi
Explorer 3.95.4s software.
The study area covers the entire territory of western We studied each altitudinal interval with the same effort,
Serbia (19◦09′-20◦39′ E, 42◦50′-44◦58′ N) and encompasses number of days spent in the field, and mileage. The minimum
approximately 18,000 km2 (Figure 1A). It is located in the distance between sites was 250 m (i.e., two populations found
central Balkans and belongs to the eastern Dinaric Alps. Two closer than 250 m from each other were considered as one site),
basic units are distinguished in the study area: (a) the flatlands except in the case when large differences in altitude and habitats
of the southern part of the Pannonian Plain, which occupy of the studied places were observed. Due to the relatively small
the northern parts of western Serbia, and (b) the mountainous size of the area searched and the long duration of the study, we
Frontiers in Ecology and Evolution 03 frontiersin.org
21
fevo-10-929266 August 3, 2022 Time: 15:21 # 4
Djordjević et al. 10.3389/fevo.2022.929266
FIGURE 1
(A) Map of the study area (central Balkans: western Serbia) with sampling sites where orchids were found (the boundaries of the study area are
marked by the red line); (B) the density of forest cover in the study area.
FIGURE 2
The number of sampling sites for each altitudinal interval where orchids were found.
Frontiers in Ecology and Evolution 04 frontiersin.org
22
fevo-10-929266 August 3, 2022 Time: 15:21 # 5
Djordjević et al. 10.3389/fevo.2022.929266
FIGURE 3
The altitudinal range of orchids in western Serbia (the central Balkans).
are sure that the number of sites we missed during the search with palmate, fusiform, or stoloniferous tubers (intermediate
is negligible and therefore cannot have affected the outcome. in evolutionary history between rhizomatous orchids and
For the same reason, we can expect that the sampling effort was orchids with spheroid tubers); and (3) tuberous orchids, i.e.,
uniform throughout the region. For subsequent calculations, we orchids with spheroid tubers (considered the most specialized
only considered sites where orchids were found. and advanced orchids). Species of the genera Cephalanthera,
Corallorhiza, Epipactis, Epipogium, Goodyera, Limodorum,
Subdivision of orchid species by and Neottia were classified in the rhizomatous orchid group,
habitat type, life forms (underground while species of the genera Coeloglossum, Dactylorhiza,
organs), and pollination systems Gymnadenia, Nigritella, Platanthera, and Pseudorchis were
placed in the intermediate group. In addition, species of the
Orchid species were divided into two categories based on genera Anacamptis, Herminium, Himantoglossum, Neotinea,
habitat type: (1) orchids that were recorded in forest habitats Ophrys, Orchis, Spiranthes, and Traunsteinera were classified as
and (2) orchids that inhabit non-forest habitats (grasslands and tuberous orchids.
herbaceous wetlands) (Supplementary Table 1). Orchids that Based on their pollination system, orchids were divided
occurred in both forest and non-forest habitats were included into three categories: (1) rewarding orchids, i.e., those that
in both habitat categories (counted twice). In addition, orchids produce nectar and offer it as a reward to their pollinators,
were relegated to various categories based on their underground (2) deceptive, and (3) self-pollinated species (Supplementary
organs and pollination systems (Supplementary Table 1). We Table 1). Information on the pollination mechanism of orchids
classified orchids as belonging to one of three underground was obtained from Jacquemyn et al. (2005a), Jersáková et al.
organ systems, following the concept presented by Tsiftsis et al. (2006), Vereecken et al. (2010), and Inda et al. (2012), while
(2019) and Štípková et al. (2021): (1) rhizomatous orchids (the for the genus Epipactis the AHO-Bayern webpage (Aho-Bayern,
most primitive ones); (2) “intermediate orchids,” i.e., orchids 2021) was used. Orchids that have nectar and thus could
Frontiers in Ecology and Evolution 05 frontiersin.org
23
fevo-10-929266 August 3, 2022 Time: 15:21 # 6
Djordjević et al. 10.3389/fevo.2022.929266
TABLE 1 Overall statistics of the polynomial regressions used to Digital Altitude Model (Copernicus, EU-DEM version 1.1) was
determine the relationship between orchid species
richness and altitude. used by carrying out an aggregation process. The Tree Cover
Density layer (2015 was used as the reference year) at a 100-m
Orchid group R2 P-value resolution (available through the Copernicus Land Monitoring
Total orchids 0.9267 (c) 0.001 Service) was used to calculate the forest area at each 100-m<
Orchids of forest habitats 0.9295 (c) 0.001 vertical interval, and then the non-forest area was calculated<
Orchids of non-forest habitats 0.9061 (c) 0.001 by removing the forested area from the total area in each<
vertical interval.
(c): 3rd order polynomial regression. An orchid was considered as present in a 100-m interval
only if it was recorded at least once in this vertical interval.
be rewarding but can also be self-pollinated (e.g., Epipactis After constructing the total matrix for all the orchid taxa
spp.) were classified in both categories (rewarding and self- recorded in the study area, two series of orchid matrices
pollinated plants). were generated according to the traits studied (underground
organ system category, pollination system). Specifically, for each
orchid category the number of orchid taxa occurring in each
Data analysis vertical interval was calculated.
To explore whether (a) the altitudinal range and (b) the
The altitudinal gradient of the study area was divided into mean altitude of occurrence of the orchids with different
twenty-one 100 m vertical intervals (i.e., 0–100 m, 101–200 m, underground organ systems and pollination systems are
etc.). Species richness was calculated for each 100-m altitudinal statistically different, we used the Kruskal-Wallis test, followed
interval as the total number of orchid species and subspecies by Dunn’s post-hoc test with Bonferroni correction carried
in that interval. The area (in km2) of each 100-m interval was out on each pair of groups. The altitudinal range for
estimated by counting the number of 100-m grid cells of a each orchid was defined as the difference between the
Digital Altitude Model (DEM) having altitude values at a specific highest and the lowest site where each orchid has been
vertical interval. To achieve the 100-m map, the 25-m European recorded, whereas the mean altitude was calculated on the
FIGURE 4
Orchid species richness along an altitudinal gradient in the central Balkans (western Serbia).
Frontiers in Ecology and Evolution 06 frontiersin.org
24
fevo-10-929266 August 3, 2022 Time: 15:21 # 7
Djordjević et al. 10.3389/fevo.2022.929266
basis of all altitude values of the sites where each orchid
has been recorded.
Orchid species density (D) at each altitude interval was
calculated using the formula:
D = S/log(A),
where S is the number of orchid species recorded in each vertical
interval and A is the area of each vertical interval in km2.
Orchid density was calculated using (a) the total orchid
flora and total area of the vertical intervals, (b) orchids of
forest habitats and the forest area, and (c) orchids of non-forest
habitats and the non-forest area of the vertical intervals. The
three data sets were used to identify possible specific patterns
in the orchids of these broad habitat categories.
The associations between richness and density of orchid
species and subspecies and altitude were explored by analyzing
the data sets using regressions. As we did not have any
a priori hypothesis about the functions describing the shape
of the dependences studied, polynomial regressions were used.
We first used third-degree polynomials and always tested
significance of the cubic terms in order to determine whether a
second-degree or a linear regression would not be sufficient for
fitting the data. Linear regression was used in cases where both
cubic and quadratic terms were insignificant (Tsiftsis et al., 2019;
Štípková et al., 2020, 2021).
All analyses were performed in R version 4.0.5 (R Core
Team, 2021), whereas variable extraction was done using
ArcGIS 10.6 (ESRI, 2017). Kruskal-Wallis and Dunn’s post-hoc
tests were performed using the packages “stats” and “FSA” (Ogle
et al., 2022), respectively.
Results
Altitudinal range size
Altitudinal range profiles of orchid species and subspecies
of the central Balkans (western Serbia) showed that most
species occurred over wide altitudinal ranges (Figure 3 and FIGURE 5
Supplementary Table 1). Thus, 12 species and subspecies Richness of orchid species and subspecies with certain
underground organ systems along the altitudinal gradient: (A)
(21.82%) had altitudinal ranges less than 500 m, 13 species and total number of orchid taxa; (B) orchids of forest habitats; (C)
subspecies (23.64%) had altitudinal ranges from 500 to 1,000 m, orchids of non-forest habitats.
20 taxa (36.36%) had altitudinal ranges from 1,000 to 1,500 m,
while 10 taxa (18.18%) had altitudinal ranges of more than
1,500 m (Figure 3 and Supplementary Table 1).
There was no significant difference in altitudinal range altitude of occurrence than rhizomatous orchids (Z = 2.01,
between orchids with different types of underground organs p< 0.05) and tuberous orchids (Z = 4.589, p< 0.001). Moreover,
(Kruskal-Wallis χ2 = 0.314, p = 0.854) or pollination systems rhizomatous orchids had a significantly higher mean altitude
(χ2 = 1.635, p = 0.441). In contrast, the mean altitude of than tuberous orchids (Z = 2.707568, p < 0.01). Similarly, the
occurrence differed significantly between orchid species with mean altitude of occurrence differed between orchid species
different underground organs (χ2 = 21.617, p < 0.001). In with different pollination systems (χ2 = 6.6227, p < 0.05),
particular, intermediate orchids had a significantly higher mean with the mean altitude of occurrence of deceptive orchids
Frontiers in Ecology and Evolution 07 frontiersin.org
25
fevo-10-929266 August 3, 2022 Time: 15:21 # 8
Djordjević et al. 10.3389/fevo.2022.929266
being significantly lower that of rewarding orchids (Z = –2.573,
p< 0.05). No significant difference was found between the mean
altitude of occurrence of deceptive and self-pollinated orchids
(Z = –0.861, p = 0.973), and rewarding and self-pollinated
orchids (Z = 1.086, p = 0.832).
Altitudinal patterns of orchid species
richness
Regression analysis showed a strong influence of altitude on
orchid species richness in western Serbia (Table 1 and Figure 4).
The total orchid species richness showed a hump-shaped
relationship along the altitudinal gradient. Species richness
reached its maximum value in the mid-altitude zone between
901 and 1,000 m (41 orchid species and subspecies) and then
decreased to reach its minimum at high-altitude sites (Table 1
and Figure 4).
Regression analysis showed a significant effect of altitude
on orchid species richness in both forest and non-forest
habitats (Table 1 and Figure 4). Both orchids of forest habitats
and orchids of non-forest habitats showed a hump-shaped
relationship with the altitudinal gradient, peaking between 901
and 1,000 m (Figure 4). In the altitudinal zone from 0 to
1,100 m, the richness of orchids of forest habitats is generally
higher than the richness of orchids of non-forest habitats
(Figure 4). However, with increasing altitude (from 1,101 to
2,100 m), the richness of orchids of non-forest habitats is higher
than the richness of orchids of forest habitats (Figure 4).
Orchid richness in terms of the number of rhizomatous,
intermediate, and tuberous orchid taxa for the three data
sets (total orchids, orchids of forest habitats, orchids of non-
forest habitats) are shown in Figure 5. The regression lines of
orchids of each orchid life form have rather the same shape
(a hump-shaped pattern). Tuberous species dominate at low
and mid-altitude zone, the rhizomatous orchids present their
highest richness at c. 1,100–1,300 m, whereas intermediate
orchids dominate at high-altitude areas (Figure 5A). The
results concerning orchids of the forest habitats were of the
similar shape for all three orchid groups (Figure 5B). However, FIGURE 6
tuberous orchids dominate at low altitude areas, whereas the Richness of orchid species and subspecies of specific pollination
systems along the altitudinal gradient: (A) total number of orchid
rhizomatous orchids dominate from mid-altitude area to high- taxa; (B) orchids of forest habitats; (C) orchids of non-forest
altitude zone (Figure 5B). In the case of orchids of non-forest habitats.
habitats, tuberous orchids dominate in low and mid-altitudinal
zone (0–1,200 m), whereas the intermediate orchids dominate
between 1,200 and 2,100 m (Figure 5C). 1,700 m, whereas the richness of rewarding orchids is higher
The trends in orchid species richness along the altitudinal than the richness of deceptive and self-pollinated orchids at
gradient based on the three pollination mechanisms are altitudes between 1,700 and 2,100 m (Figure 6A). The altitudinal
shown in Figure 6. The orchids of each orchid pollination patterns of orchid species richness of specific pollination systems
system showed a hump-shaped relationship with the altitudinal were hump-shaped also in cases when orchids of forest and non-
gradient, peaking at mid-altitude zone (Figure 6A). In general, forest habitats were considered separately (Figures 6B,C). All
the richness of deceptive orchids is greater than the richness of the correlations between orchid species richness and altitude
rewarding and self-pollinated orchids at altitudes between 0 and using the three datasets were statistically significant (P < 0.001
Frontiers in Ecology and Evolution 08 frontiersin.org
26
fevo-10-929266 August 3, 2022 Time: 15:21 # 9
Djordjević et al. 10.3389/fevo.2022.929266
TABLE 2 Overall statistics of the polynomial regressions used to TABLE 4 Overall statistics of the polynomial regressions used to
determine the relationship between the richness of orchid species determine the relationship between orchid species
with specific underground organ systems and altitude. density and altitude.
Orchid group R2 P-value Orchid group R2 P-value
Total orchids Total orchids 0.8967 (c) <0.01
Tuberous 0.906 (c) <0.001 Orchids of forest habitats 0.938 (c) <0.001
Rhizomatous 0.842 (b) <0.001 Orchids of non-forest habitats 0.7711 (c) <0.001
Intermediate 0.829 (b) <0.001
(c): 3rd order polynomial regression.
Orchids of forest habitats
Tuberous 0.822 (c) <0.001
Rhizomatous 0.861 (b) <0.001 stabilized and slightly increased up to the highest altitudes. In
Intermediate 0.676 (b) <0.001 the in lowland areas, the density of orchids of forest habitats
Orchids of non-forest habitats is generally higher than the density of orchids of non-forest
Tuberous 0.923 (c) <0.001 habitats, whereas the density of orchids of non-forest habitats
Rhizomatous 0.420 (b) <0.01 is higher than the density of orchids of forest habitats at mid- to
Intermediate 0.885 (c) <0.001 high-altitude zone (Figure 7).
Orchid densities in terms of the number of rhizomatous,
(b): 2nd order polynomial regression; (c): 3rd order polynomial regression. intermediate, and tuberous orchid taxa for the three data sets
TABLE 3 Overall statistics of the polynomial regressions used to (total orchids—total area, forest orchids—forested area, non-
determine the relationship between the richness of orchid species forest orchids—non-forested area) are shown in Figure 8. When
with specific pollination systems and altitude.
the orchid density was calculated based on the total orchid
Orchid group R2 P-value flora and the total area at each vertical interval, the curves of
rhizomatous and tuberous orchids have rather the same shape
Total orchids
(a hump-shaped pattern), but the number of tuberous species
Rewarding 0.938 (c) <0.001
is slightly higher (Figure 8A). Contrary to these two species
Deceptive 0.889 (b) <0.001
groups, the intermediate orchids show an increasing trend,
Self-pollinated 0.924 (b) <0.001
and their density is gradually stabilized above 1,500 m. The
Orchids of forest habitats
results concerning orchids of the forest area were of the same
Rewarding 0.914 (b) <0.001
shape for all three orchid groups (Figure 8B). In the case of
Deceptive 0.882 (c) <0.001
orchid density calculated using non-forest orchids and the non-
Self-pollinated 0.821 (b) <0.001
forest area, the intermediate orchids showed a sharp increase
Orchids of non-forest habitats
along the altitudinal gradient, whereas the tuberous orchids
Rewarding 0.869 (c) <0.001
showed a hump-shaped pattern (Figure 8C). Rhizomatous
Deceptive 0.891 (b) <0.001
orchids have the lowest species density compared to the other
Self-pollinated 0.664 (c) <0.01
two orchid groups.
(b): 2nd order polynomial regression; (c): 3rd order polynomial regression. The trends in orchid species density along the altitudinal
gradient based on the three pollination mechanisms are shown
or P< 0.01) (Tables 2, 3), whereas the predictive power was very in Figure 9. In the case of orchid density calculated based on
high in almost all regressions. the total number of orchids and the total area, both deceptive
and self-pollinated orchids show a unimodal trend with a peak
at about 900–1,000 m, the deceptive orchids being the richest
Altitudinal patterns of orchid species in terms of species (Figure 9A). On the contrary, density of
density rewarding orchids increases sharply up to 900 m and then
slightly decreases. The graph of orchid density of the forest
Regression analysis showed a strong influence of altitude habitat types is presented in Figure 9B. Here, all orchid groups
on orchid species density in the central Balkans (Table 4 and show a unimodal trend. When analyzing non-forest orchids
Figure 7). The regression lines of total orchids and orchids using the non-forested area, we found that the deceptive orchids
of forest habitats have rather the same shape (a hump-shaped showed a unimodal trend, reaching a maximum density at c.
pattern). Total species density reached its maximum value in 1,000 m, much higher than density of the other orchid groups
the mid-altitude zone (at c. 1,000 m) and then decreased to (Figure 9C). The density of rewarding orchids increases with
reach its minimum in high-altitude areas (Figure 7). Species increase in altitude, peaking at about 500 m, then is stabilized
density of orchids of non-forest habitats increased with increase or slightly decreases up to c. 1,400 m before increasing again up
in altitude, peaking at about 800 m, then slightly decreased or to the highest altitudes. Self-pollinated species are only poorly
Frontiers in Ecology and Evolution 09 frontiersin.org
27
fevo-10-929266 August 3, 2022 Time: 15:21 # 10
Djordjević et al. 10.3389/fevo.2022.929266
FIGURE 7
Orchid species density along an altitudinal gradient in the central Balkans (western Serbia).
represented in non-forested areas and show a slight hump- organs and pollination systems differ significantly along the
shaped pattern. altitudinal gradient when the orchid flora of specific habitat
All the correlations between orchid species density and types was analyzed.
altitude using the three datasets were statistically significant
(P < 0.001 or P < 0.01) (Tables 5, 6). Moreover, the predictive
power was very high in almost all regressions. Specifically, the Altitudinal range size
predictive power in the three datasets in the analyses performed
using the underground organ system categories was 51.4–92.5%, The results of this study show that 10 orchid taxa have
whereas when analyzing orchids on the basis of their pollination the largest altitudinal ranges (above 1,500 m) in the central
mechanisms the predictive power of the regressions was 77.5– Balkans (Figure 3 and Supplementary Table 1), highlighting
93.1%. their ecological plasticity and adaptability, as well as a lower
degree of specialization.
Our results show that orchids belonging to the Central
Discussion European and boreal chorological groups (Coeloglossum viride,
Dactylorhiza fuchsii, Goodyera repens, Epipactis leptochila subsp.
In this study we investigated how the richness and density of neglecta, Epipactis muelleri, Epipactis purpurata, Epipogium
orchids vary along the altitudinal gradient in the central Balkans. aphyllum, and Neottia cordata) occur in the middle and high
Specifically, we explored whether the forest and non-forest altitudes of the central Balkans. In contrast, in Central and
areas along the altitudinal gradient affect patterns of orchids Northern Europe, these species have higher altitude ranges,
richness and density using different functional traits. Our results from lowlands to high-mountain areas (Baumann et al., 2006;
showed a hump-shaped pattern of orchid richness and density, Delforge, 2006). In addition, in the central Balkans, orchids
peaking in the mid-altitude area. In addition, the richness characteristic primarily of Central and Northern Europe have
and density of orchids of forest habitats are generally slightly a greater elevational range or occur at lower altitudes compared
higher than the richness and density of orchids of non-forest to northeastern Greece (Tsiftsis et al., 2008). Furthermore, some
habitats in lowland areas, while orchids of non-forest habitats Mediterranean-submediterranean orchids (e.g., Anacamptis
dominate in high-altitude areas. The results showed that the papilionacea, Neotinea tridentata, and Orchis simia) have a
diversity patterns of orchid species with different underground lower altitudinal range and occur mainly at lower altitudes in
Frontiers in Ecology and Evolution 10 frontiersin.org
28
fevo-10-929266 August 3, 2022 Time: 15:21 # 11
Djordjević et al. 10.3389/fevo.2022.929266
thermophilous habitats are mainly present at lower and
middle altitudes.
Altitudinal patterns of orchid species
richness and density
The results of this study show hump-shaped patterns of
orchid richness and density along the altitudinal gradient in
western Serbia, both in the case of total orchids and in the
cases of orchids of specific habitat types. In all cases, the highest
species richness and density were observed between 500 and
1,200 m. This is consistent with previous studies indicating that
orchid species richness is highest at mid-altitudes and decreases
with increasing altitude (Acharya et al., 2011; Chen et al., 2014;
Liu et al., 2015; Zhang et al., 2015a). It is assumed that patterns
of orchid species richness and density along the altitudinal
gradient in the central Balkans (western Serbia) are primarily
determined by climatic factors and breadth of the climatic niche
of species composing the orchid pool in western Serbia. The
highest orchid species richness and density at mid-altitudes in
the central Balkans can be explained by the fact that most species
tolerate the moderate environmental conditions in the middle
altitudes better than the extreme environmental conditions,
in terms of temperature, precipitation, relative air humidity,
ultraviolet radiation, atmospheric pressure, partial pressure of
all atmospheric gases, and anthropogenic influences in the low
and high-altitudes (Lomolino, 2001; van der Meulen et al., 2001;
Körner, 2007). The hump-shaped patterns of orchid richness
and density along the altitudinal gradient can also be explained
by size of the area. In western Serbia, area of the high-altitude
zones (from 1,500 to 2,100 m) is rather restricted compared
to areas at middle altitudes. On the contrary, although low-
altitude areas (e.g., <500 m) are extensive in the study area,
species richness and density in such areas are quite low because
a large part of these areas has been converted to cultivated
land and the landscape is not very heterogeneous in terms of
habitats and geological substrates. Similarly, previous research
has indicated that habitat heterogeneity overrides the species-
FIGURE 8 area relationship and is the most important predictor of species
Density of orchid species and subspecies with certain richness (Báldi, 2008; Tsiftsis, 2020). In addition, it is assumed
underground organ systems along the altitudinal gradient: (A)
total number of orchid taxa—total area; (B) forest that the lower richness and density of orchids of non-forest
orchids—forest area; (C) non-forest orchids—non-forest area. habitats at lower altitudes can be explained by the intense
anthropogenic influences. In general, the species richness of a
given altitudinal range is related to its extent. However, this
is correct for orchids of forest habitats, but not for orchids of
western Serbia than in northeastern Greece (Tsiftsis et al., 2008), non-forest habitats. Thus, our study partially confirms the SAR
which can be explained primarily by the climatic differences hypothesis (Karger et al., 2011). We could assume that the lower
between these two study areas. Indeed, north-eastern Greece richness of orchid species in the high-altitude areas of western
is under strong influence of the Mediterranean climate and Serbia is determined by the lower diversity of their pollinators
has a significant presence of thermophilous habitats along (Arroyo et al., 1982; Jacquemyn et al., 2005b), as well as by a
the altitudinal gradient. In contrast, in the central Balkans smaller pollen load of pollinators (Bingham and Orthner, 1998).
(western Serbia), due to the humid and continental climate, Furthermore, the greater richness and density of orchid species
Frontiers in Ecology and Evolution 11 frontiersin.org
29
fevo-10-929266 August 3, 2022 Time: 15:21 # 12
Djordjević et al. 10.3389/fevo.2022.929266
of forest habitat types compared to the species richness and
density of orchids of non-forest habitats in lowland areas can be
explained primarily by the high diversity of forest communities
in this altitude range. On the other hand, the greater richness
and density of orchid species of herbaceous vegetation in
relation to the richness and density of orchid species of forest
vegetation in the high-altitude regions of western Serbia can
be explained by the great diversity and high heterogeneity of
grasslands and herbaceous wetlands, as well as by the lower
diversity of forest vegetation.
Orchid species richness and density in terms of
different underground organ systems
The results of this study show that tuberous orchids
dominate in areas of lower and middle altitudes (Figures 5, 8).
This result was expected, bearing in mind that orchids of
this life form are best adapted to dry, semi-dry, and warm
habitat conditions, such as those found at the lower and
middle altitudes of the study area (Dafni, 1987; Averyanov,
1990; Tsiftsis et al., 2019; Štípková et al., 2021). Rhizomatous
orchids are predominant in mid-altitude areas, indicating that
moderate environmental conditions are appropriate for them.
However, the results showed that altitude strongly influences
rhizomatous orchid species richness and density in forest
habitats, whereas the influence of altitude is relatively weak
when it comes to the richness and density of rhizomatous
orchids in non-forest habitats. This can be explained by the fact
that representatives of the orchids of this life form, as the most
primitive representatives of European orchids, primarily grow
in forest habitats (Averyanov, 1990; Delforge, 2006).
This study shows that intermediate orchids dominate in
high-altitude areas, which is consistent with previous studies
suggesting that these orchids prefer lower temperatures and
higher humidity in their habitats and therefore occur in high-
altitude areas (Averyanov, 1990; Delforge, 2006; Pillon et al.,
2006; Tsiftsis et al., 2019). The results are understandable,
bearing in mind the evolutionary development of orchids.
Specifically, the evolution of the first intermediate orchids was
associated with the Alpine orogenesis, and the formation of
mountain habitats with lower temperatures (Averyanov, 1990).
FIGURE 9
Density of orchid species and subspecies of specific pollination
Orchid species richness and density in terms of systems along the altitudinal gradient: (A) total number of orchid
taxa—total area; (B) forest orchids—forest area; (C) non-forest
different pollination systems orchids—non-forest area.
The results of this study show that the richness and density
of deceptive orchids are higher through almost all the altitudinal
gradient studied, and that only in the highest regions of the as temperature, precipitation, or seasonality (Körner, 2007),
investigated area do rewarding orchids prevail (Figures 6, 9). as well as by factors that influence the decrease of pollinator
This result is consistent with those of Pellissier et al. (2010), who diversity and visitation rate at high altitudes (Arroyo et al., 1982;
found that the relative occurrence of food-deceptive orchids Jacquemyn et al., 2005b; Pellissier et al., 2010).
decreases with increasing altitude in the territory of Switzerland Štípková et al. (2020) used nectarless and nectariferous
and in the Vaud mountains, suggesting that deception may be orchids of the Czech Republic and found that both groups
less profitable at high compared to low altitudes. This may be showed a hump-shaped pattern of species density, with a
explained by climatic factors expressed through altitude, such maximum between 300 and 900 m, i.e., at lower altitudes
Frontiers in Ecology and Evolution 12 frontiersin.org
30
fevo-10-929266 August 3, 2022 Time: 15:21 # 13
Djordjević et al. 10.3389/fevo.2022.929266
TABLE 5 Overall statistics of the polynomial regressions used to for the survival of rhizomatous orchids. In addition, non-forest
determine the relationship between the density of orchid species with
specific underground organ systems and altitude. habitats at mid- and high-altitudes are most important for the
survival of intermediate orchids. Given the resulting diversity
Orchid group R2 P-value patterns and the fact that intermediate orchids inhabit colder
Total orchids and higher precipitation areas (Tsiftsis et al., 2019; Štípková
Tuberous 0.919 (c) <0.001 et al., 2021), our study suggests that these orchids may be
Rhizomatous 0.789 (b) <0.001 affected by the rise of temperature and lower precipitation at
Intermediate 0.829 (b) <0.001 lower altitudes due to climate change.
Orchids of forest habitats Our study suggests that forest and non-forest habitats at
Tuberous 0.912 (c) <0.001 low and mid-altitudes are most important for the survival
Rhizomatous 0.874 (b) <0.001 of deceptive orchids. On the other hand, mid- and high-
Intermediate 0.667 (b) <0.001 altitudinal areas are important for the survival of rewarding
Orchids of non-forest habitats orchids. Since rewarding orchids are rarer at lower altitudes,
Tuberous 0.925 (c) <0.001
they are at high risk of extinction in these areas. In view of
Rhizomatous 0.514 (c) <0.01
Intermediate 0.831 (a) <0.001 the fact that the rewarding orchids in western Serbia occur in
almost equal numbers in forest and non-forest habitats, it is
(a): 1st order polynomial regression; (b): 2nd order polynomial regression; (c): 3rd order
polynomial regression. necessary to carefully plan their conservation. Deceptive orchids
in the central Balkans occur in slightly higher numbers in non-
TABLE 6 Overall statistics of the polynomial regressions used to forest habitats (grasslands and meadows), a circumstance that
determine the relationship between the density of orchid species with
specific pollination systems and altitude. requires careful conservation of these habitats. Finally, our study
indicates that most orchid species grow in mid-altitude areas,
Orchid group R2 P-value which coincide with the strong presence of tourist sites and
Total orchids facilities in the study area. It is therefore necessary to work on
Rewarding 0.822 (c) <0.001 a carefully designed plan for protection of these areas, including
Deceptive 0.867 (b) <0.001 the application of ecologically sustainable tourism that does not
Self-pollinated 0.931 (c) <0.001 threaten orchids to extinction.
Orchids of forest habitats
Rewarding 0.893 (b) <0.001
Deceptive 0.920 (c) <0.001 Conclusion
Self-pollinated 0.857 (b) <0.001
Orchids of non-forest habitats
Rewarding 0.807 (c) <0.001 This study demonstrates a hump-shaped pattern of orchid
Deceptive 0.826 (b) <0.001 richness and density peaking at 900–1,000 m and the fact
Self-pollinated 0.775 (c) <0.001 that orchid species diversity patterns differ significantly along
(b): 2nd order polynomial regression; (c): 3rd order polynomial regression. the altitudinal gradient when comparing forest vs. non-forest
habitats. In general, our results confirm the SAR hypothesis,
i.e., that the richness and density of orchid species along the
compared to orchids in the central Balkans. Similarly to our altitudinal gradient are significantly affected by size of the area of
results, species density of both nectariferous and nectarless a given altitudinal interval. However, this does not hold true for
orchids along the altitudinal gradient in the Czech Republic was the orchids of non-forest habitats. Furthermore, it does not hold
found to depend on habitat cover, i.e., the spatial distribution true for the intermediate orchids in non-forest habitats or for
of forest and non-forest habitats. Earlier studies of orchids have the rewarding orchids in the same habitats because the species
shown that most self-pollinated orchids occur in high-altitude density of these groups increases with altitude.
areas (Catling, 1990; Jacquemyn et al., 2005b). Self-pollinated Our study suggests that the diversity patterns of orchid
orchids in the central Balkans mostly inhabit forest vegetation species with different underground organs and pollination
types, so the density of these orchids is highest in mid-altitude systems differ significantly along the altitudinal gradient when
areas, in which forest orchids dominate. considering the total flora in the whole area, but also
when analyzing the orchid flora of specific habitat types.
In general, tuberous orchids dominate in low and mid-
Implications for conservation altitude areas, intermediate orchids dominate at high altitudes,
while rhizomatous orchids are predominate in mid-altitude
This study shows that forest and non-forest habitats at low forest stands. This confirms the hypothesis of evolutionary
and mid- altitudes have high conservation value for tuberous development of orchids with different underground organs
orchids, while forest habitats at mid-altitudes are important and their specific ecological requirements (Averyanov, 1990;
Frontiers in Ecology and Evolution 13 frontiersin.org
31
fevo-10-929266 August 3, 2022 Time: 15:21 # 14
Djordjević et al. 10.3389/fevo.2022.929266
Tsiftsis et al., 2019; Štípková et al., 2021). Our study highlights Acknowledgments
the importance of low and mid-altitude areas for the survival of
deceptive orchids and the importance of mid- and high-altitude We are grateful to the Tara National Park, the Forest
areas for the survival of rewarding orchids. In addition, forest Estate Golija Ivanjica, and the Tourist Organization of Čačak
habitats at mid-altitudes have been shown to be crucial for the for logistical field support. We thank reviewers for their
survival of self-pollinated orchids. useful suggestions and comments on a previous version of the
In general, our study shows that the strategies required manuscript. We would also like to thank Raymond Dooley for
to protect orchids change along the altitudinal gradient and proofreading the manuscript.
depend on both functional traits of species and habitat cover. In
addition, our results suggest that changes in habitat cover may
be reflected in patterns of orchid diversity along the altitudinal
gradient. Future research should reveal which climatic and other
environmental factors are crucial for the changes in orchid Conflict of interest
species richness and density along the altitudinal gradient in the
central Balkans. The authors declare that the research was conducted in the
absence of any commercial or financial relationships that could
be construed as a potential conflict of interest.
Data availability statement
The original contributions presented in this study are
included in the article/Supplementary material, further Publisher’s note
inquiries can be directed to the corresponding author.
All claims expressed in this article are solely those of the
authors and do not necessarily represent those of their affiliated
Author contributions organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
VD: fieldwork. VD and ST: methodology and formal claim that may be made by its manufacturer, is not guaranteed
analysis. All authors conceptualization, writing – review or endorsed by the publisher.
and editing and read and agreed to the published version
of the manuscript.
Funding Supplementary material
This work was supported by the Ministry of Education, The Supplementary Material for this article can be
Science and Technological Development of the Republic of found online at: https://www.frontiersin.org/articles/10.3389/
Serbia (number 451-03-68/2022-14/200178). fevo.2022.929266/full#supplementary-material
References
Acharya, K. P., Vetaas, O. R., and Birks, H. J. B. (2011). Orchid species richness Baumann, H., Künkele, S., and Lorenz, R. (2006). Die Orchideen Europas. Mit
along Himalayan elevational gradients. J. Biogeogr. 38, 1821–1833. doi: 10.1111/j. angrenzenden Gebieten. Stuttgart: Eugen Ulmer KG.
1365-2699.2011.02511.x
Bingham, R. A., and Orthner, A. R. (1998). Efficient pollination of alpine plants.
Ackerman, J. D., Trejo-Torres, J. C., and Crespo-Chury, Y. (2007). Orchids of Nature 391, 238–239. doi: 10.1038/34564
the West Indies: predictability of diversity and endemism. J. Biogeogr. 34, 779–786.
doi: 10.1111/j.1365-2699.2006.01676.x Cardelús, L. C., Colwell, R. K., and Watkins, J. E. Jr. (2006). Vascular epiphyte
distribution patterns: explaining the mid-elevation richness peak. J. Ecol. 94,
Aho-Bayern (2021). ArbeitskreisHeimischeOrchideen Bayern e.V. Available 144–156. doi: 10.1111/j.1365-2745.2005.01052.x
online at: http://www.aho-bayern.de (accessed 26 June, 2021)
Catling, P. (1990). “Auto-pollination in the Orchidaceae,” in Orchid Biology:
Arroyo, M. T. K., Primack, R., and Armesto, J. (1982). Community studies in Reviews and Perspectives, ed. J. E. Arditti (Portland, OR: Timber Press),
pollination ecology in the high temperate Andes of Central Chile. I. pollination 121–158.
mechanisms and altitudinal variation. Am. J. Bot. 69, 82–97. doi: 10.1002/j.1537-
2197.1982.tb13237.x Chase, M., Christenhusz, M., and Mirenda, T. (2017). The Book of Orchids: A
Life-Size Guide to Six Hundred Species from Around the World. London: Ivy Press.
Averyanov, L. V. (1990). “A review of the genus Dactylorhiza,” in Orchid Biology:
Reviews and Perspectives V, ed. J. Arditti (Portland, OR: Timber Press), 159–206. Chen, Y. K., Yang, X. B., Yang, Q., Li, D. H., Long, W. X., and Luo, W. Q.
(2014). Factors affecting the distribution pattern of wild plants with extremely
Báldi, A. (2008). Habitat heterogeneity overrides the species–area relationship. small populations in Hainan Island, China. PLoS One 9:e97751. doi: 10.1371/
J. Biogeogr. 35, 675–681. doi: 10.1111/j.1365-2699.2007.01825.x journal.pone.0097751
Frontiers in Ecology and Evolution 14 frontiersin.org
32
fevo-10-929266 August 3, 2022 Time: 15:21 # 15
Djordjević et al. 10.3389/fevo.2022.929266
Christenhusz, M. J. M., and Byng, J. W. (2016). The number of known plants Laughlin, D. C., Joshi, C., van Bodegom, P. M., Bastow, Z. A., and Fulé, P. Z.
species in the world and its annual increase. Phytotaxa 261, 201–217. doi: 10. (2012). A predictive model of community assembly that incorporates intraspecific
11646/phytotaxa.261.3.1 trait variation. Ecol. Lett. 15, 1291–1299. doi: 10.1111/j.1461-0248.2012.01852.x
Colwell, R. K., and Hurtt, G. C. (1994). Nonbiological gradients in species Liu, Q., Chen, J., Corlett, R. T., Fan, X., Yu, D., Yang, H., et al. (2015). Orchid
diversity and a spurious Rapoport effect. Am. Nat. 144, 570–595. doi: 10.1086/ conservation in the biodiversity hotspot of southwestern China. Conserv. Biol. 29,
285695 1563–1572. doi: 10.1111/cobi.12584
Colwell, R. K., and Lees, D. C. (2000). The mid-domain effect: geometric Lomolino, M. V. (2001). Elevation gradients of species-density: historical and
constraints on the geography of species diversity. Trends Ecol. Evol. 15, 70–76. prospective views. Glob. Ecol. Biogeogr. 10, 3–13. doi: 10.1046/j.1466-822x.2001.
doi: 10.1016/S0169-5347(99)01767-X 00229.x
Colwell, R. K., Rahbek, C., and Gotelli, N. J. (2004). The mid-domain effect and McCain, C. M., and Grytnes, J. A. (2010). Elevational Gradients in species
species diversity patterns: what have we learned so far? Am. Nat. 163, E1–E23. richness. Encyclopedia of Life Sciences (ELS). Chichester: John Wiley & Sons, Ltd.
doi: 10.1086/382056 Ogle, D. H., Doll, J. C., Wheeler, P., and Dinno, A. (2022). FSA: Fisheries Stock
Dafni, A. (1987). “Pollination in Orchis and related genera: evolution from Analysis. R Package Version 0.9.3. Available online at: https://github.com/fishR-
reward to deception,” in Orchid biology, Reviews and Perspectives IV, ed. J. Adritti Core-Team/FSA (accessed June 9, 2022).
(Ithaca, NY: Cornell University Press), 79–104. Pellissier, L., Vittoz, P., Internicola, A. I., and Gigord, L. D. B. (2010).
Delforge, P. (2006). Orchids of Europe, North Africa and the Middle East. Generalized food-deceptive orchid species flower earlier and occur at lower
London: A. & C. Black. altitudes than rewarding ones. J. Plant Ecol. 3, 243–250. doi: 10.1093/jpe/rtq012
Djordjević, V., Niketić, M., and Stevanović, V. (2021). “Orchids of Serbia: Pillon, Y., Fay, M. F., Shipunov, A. B., and Chase, M. W. (2006). Species diversity
taxonomy, life forms, pollination systems, and phytogeographical analysis,” versus phylogenetic diversity: a practical study in the taxonomically difficult genus
in Orchidaceae: Characteristics, Distribution and Taxonomy, ed. V. Djordjević Dactylorhiza (Orchidaceae). Biol. Conserv. 129, 4–13. doi: 10.1016/j.biocon.2005.
(New York, NY: Nova Science Publishers), 57–163. 06.036
Djordjević, V., and Tsiftsis, S. (2019). Patterns of orchid species richness and R Core Team (2021). R: A Language And Environment Statistical Computing.
composition in relation to geological substrates. Wulfenia 26, 1–21. Vienna: R Foundation for Statistical Computing. Available online at: https://https:
Djordjević, V., and Tsiftsis, S. (2022). “The role of ecological factors in //www.r-project.org/
distribution and abundance of terrestrial orchids,” in Orchids Phytochemistry, Rahbek, C. (1995). The elevational gradient of species richness: a uniform
Biology and Horticulture, Reference Series in Phytochemistry. Fundamentals and pattern? Ecography 18, 200–205. doi: 10.1111/j.1600-0587.1995.tb00341.x
Applications, eds J.-M. Mérillon and H. Kodja (Cham: Springer International
Publishing), 3–72. Štípková, Z., Traxmandlová, I., and Kindlmann, P. (2016). Determinants oforchid species diversity in Latin America. Lankesteriana 16, 293–297. doi: 10.
Djordjević, V., Tsiftsis, S., Lakušić, D., Jovanović, S., Jakovljević, K., and 15517/lank.v16i2.26013
Stevanović, V. (2020). Patterns of distribution, abundance and composition of
forest terrestrial orchids. Biodivers. Conserv. 29, 4111–4134. doi: 10.1007/s10531- Štípková, Z., Tsiftsis, S., and Kindlmann, P. (2020). Pollination mechanisms are
020-02067-6 driving orchid distribution in space. Sci. Rep. 10:850. doi: 10.1038/s41598-020-57871-5
Djordjević, V., Tsiftsis, S., Lakušić, D., Jovanović, S., and Stevanović, V. (2016).
Factors affecting the distribution and abundance of orchids in grasslands and Štípková, Z., Tsiftsis, S., and Kindlmann, P. (2021). Distribution of orchids
herbaceous wetlands. Syst. Biodivers. 14, 355–370. doi: 10.1080/14772000.2016. with different rooting systems in the Czech Republic. Plants 10:632. doi: 10.3390/
1151468 plants10040632
Dunn, R. R., McCain, C. M., and Sanders, N. (2007). When does diversity fit Swarts, N. D., and Dixon, K. W. (2009). Terrestrial orchid conservation in the
null model predictions? Scale and range size mediate the mid-domain effect. Glob. age of extinction. Ann. Bot. 104, 543–556. doi: 10.1093/aob/mcp025
Ecol. Biogeogr. 16, 305–312. doi: 10.1111/j.1466-8238.2006.00284.x Taylor, A., Keppel, G., Weigelt, P., Zotz, G., and Kreft, H. (2021). Functional
ESRI (2017). ArcGIS – ArcMap: ArcInfo (version 10.6). Redlands, CA: traits are key to understanding orchid diversity on islands. Ecography 44, 703–714.
Environmental Science Research Institute. Timsina, B., Kindlmann, P., Subedi, S., Khatri, S., and Rokaya, M. B. (2021).
Inda, L. A., Pimentel, M., and Chase, M. W. (2012). Phylogenetics of tribe Epiphytic orchid diversity along an altitudinal gradient in central Nepal. Plants
Orchideae (Orchidaceae: Orchidoideae) based on combined DNA matrices: 10:1381. doi: 10.3390/plants10071381
inferences regarding timing of diversification and evolution of pollination Trigas, P., Panitsa, M., and Tsiftsis, S. (2013). Elevational gradient of vascular
syndromes. Ann. Bot. 110, 71–90. doi: 10.1093/aob/mcs083 plant species richness and endemism in Crete – the effect of post-isolation
Jacquemyn, H., Brys, R., Hermy, M., and Willems, J. H. (2005a). Does nectar mountain uplift on a continental island system. PLoS One 8:e59425. doi: 10.1371/
reward affect rarity and extinction probabilities of orchid species? an assessment journal.pone.0059425
using historical records from Belgium and the Netherlands. Biol. Conserv. 121, Tsiftsis, S. (2020). The complex effect of heterogeneity and isolation in
257–263. doi: 10.1016/j.biocon.2004.05.002 determining alpha and beta orchid diversity on islands in the Aegean archipelago.
Jacquemyn, H., Micheneau, C., Roberts, D. L., and Pailler, T. (2005b). Syst. Biodivers. 18, 281–294. doi: 10.1080/14772000.2020.1738584
Elevational gradients of species diversity, breeding system and floral traits of Tsiftsis, S., Štípková, Z., and Kindlmann, P. (2019). Role of way of life, latitude,
orchid species on Réunion Island. J. Biogeogr. 32, 1751–1761. doi: 10.1111/j.1365- elevation and climate on the richness and distribution of orchid species. Biodivers.
2699.2005.01307.x Conserv. 28, 75–96. doi: 10.1007/s10531-018-1637-4
Jacquemyn, H., Honnay, O., and Pailler, T. (2007). Range size variation, Tsiftsis, S., Tsiripidis, I., Karagiannakidou, V., and Alifragis, D. (2008). Niche
nestedness and species turnover of orchid species along an altitudinal gradient analysis and conservation of the orchids of east Macedonia (NE Greece). Acta
on Réunion Island: Implications for conservation. Biol. Conserv. 136, 388–397. Oecol. 33, 27–35.
doi: 10.1016/j.biocon.2006.12.008 van der Meulen, M. A., Hudson, A. J., and Scheiner, S. M. (2001). Three
Jersáková, J., Johnson, S. D., and Kindlmann, P. (2006). Mechanisms and evolutionary hypotheses for the hump-shaped productivity–diversity curve. Evol.
evolution of deceptive pollination in orchids. Biol. Rev. 81, 219–235. doi: 10.1017/ Ecol. Res. 3, 379–392.
S1464793105006986 Vereecken, N. J., Dafni, A., and Cozzolino, S. (2010). Pollination syndromes in
Jersáková, J., Traxmandlová, I., Ipser, Z., Matthias, K., Pellegrino, G., Schatz, B., mediterranean orchids – implications for speciation, taxonomy and conservation.
et al. (2015). Biological flora of Central Europe: Dactylorhiza sambucina (L.) Soó. Bot. Rev. 76, 220–240. doi: 10.1007/s12229-010-9049-5
Perspect. Plant Ecol. Evol. Syst. 17, 318–329. doi: 10.1016/j.ppees.2015.04.002 Waterman, R. J., and Bidartondo, M. I. (2008). Deception above, deception
Karger, D. N., Kluge, J., Krömer, T., Hemp, A., Lehnert, M., and Kessler, M. below: linking pollination and mycorrhizal biology of orchids. J. Exp. Bot. 59,
(2011). The effect of area on local and regional elevational patterns of species 1085–1096. doi: 10.1093/jxb/erm366
richness. J. Biogeogr. 38, 1177–1185. doi: 10.1111/j.1365-2699.2010.02468.x Zhang, S.-B., Chen, W.-Y., Huang, J.-L., Bi, Y.-F., and Yang, X.-F. (2015a).
Körner, C. (2007). The use of ‘altitude’ in ecological research. Trends Ecol. Evol. Orchid species richness along elevational and environmental gradients in yunnan,
22, 569–574. doi: 10.1016/j.tree.2007.09.006 China. PLoS One 10:e0142621. doi: 10.1371/journal.pone.0142621
Kull, T., and Hutchings, M. J. (2006). A comparative analysis in decline in the Zhang, Z., Yan, Y., Tianb, Y., Lib, J., Hea, J.-S., and Tanga, Z. (2015b).
distribution ranges of orchid species in Estonia and the United Kingdom. Biol. Distribution and conservation of orchid species richness in China. Biol. Conserv.
Conserv. 129, 31–39. doi: 10.1016/j.biocon.2005.09.046 181, 64–72. doi: 10.1016/j.biocon.2014.10.026
Frontiers in Ecology and Evolution 15 frontiersin.org
33
fevo-10-994641 October 6, 2022 Time: 11:0 # 1
TYPE Original Research
PUBLISHED 06 October 2022
DOI 10.3389/fevo.2022.994641
Dominant Dendrobium
OPEN ACCESS officinale mycorrhizal partners
EDITED BY
Dennis Whigham, vary among habitats and
Smithsonian Institution, United States
REVIEWED BY strongly induce seed
Peter Zale,
Longwood Gardens, United States
Stefania Cevallos, germination in vitro
Universidad Técnica Particular de Loja,
Ecuador
*CORRESPONDENCE Liyue Zhang1, Kento Rammitsu2,3, Kenshi Tetsuka4,
Yuki Ogura-Tsujita 5
ytsujita@cc.saga-u.ac.jp Tomohisa Yukawa and Yuki Ogura-Tsujita1,2*
1
SPECIALTY SECTION Faculty of Agriculture, Saga University, Saga, Japan, 2United Graduate School of Agricultural
This article was submitted to Sciences, Kagoshima University, Kagoshima, Japan, 3Department of Health Chemistry, Showa
Conservation and Restoration Ecology, Pharmaceutical University, Tokyo, Japan, 4Yakushima Evergreen Broad-Leaved Forest Network,
a section of the journal Kagoshima, Japan, 5National Museum of Nature and Science, Ibaraki, Japan
Frontiers in Ecology and Evolution
RECEIVED 15 July 2022
ACCEPTED 21 September 2022
PUBLISHED 06 October 2022 Dendrobium officinale (Orchidaceae) is an endangered epiphytic orchid that
CITATION has been well studied as a medicinal plant. Although previous studies have
Zhang L, Rammitsu K, Tetsuka K,
Yukawa T and Ogura-Tsujita Y (2022) shown that various fungal isolates promote D. officinale seed germination
Dominant Dendrobium officinale and seedling development in vitro, mycorrhizal associations among its
mycorrhizal partners vary among
habitats and strongly induce seed wild populations remain poorly understood. In this study, we identified
germination in vitro. mycorrhizal fungi associated with D. officinale (36 individuals from six
Front. Ecol. Evol. 10:994641. sites) using Sanger sequencing and compared fungal communities among
doi: 10.3389/fevo.2022.994641
sites and habitats (lithophytic vs. epiphytic individuals). Among the obtained
COPYRIGHT
© 2022 Zhang, Rammitsu, Tetsuka, sequences, 76 belonged to orchid mycorrhizal fungi (OMF), among which
Yukawa and Ogura-Tsujita. This is an Tulasnellaceae accounted for 45.8% and Serendipitaceae for 28.1%. The
open-access article distributed under
the terms of the Creative Commons Serendipitaceae operational taxonomic unit (OTU) SE1 was the most dominant
Attribution License (CC BY). The use, partner, accounting for 27.1% of all detected fungal sequences, followed
distribution or reproduction in other
forums is permitted, provided the by a Tulasnellaceae OTU, TU27, which accounted for 15.6%. The relative
original author(s) and the copyright frequencies of Serendipitaceae and Tulasnellaceae differed greatly between
owner(s) are credited and that the lithophytic and epiphytic individuals. Serendipitaceae accounted for 47.3%
original publication in this journal is
cited, in accordance with accepted of the OMF sequences among lithophytes, and Tulasnellaceae for 95.2%
academic practice. No use, distribution among epiphytes. Mycorrhizal community composition also varied among
or reproduction is permitted which
does not comply with these terms. sites. We further conducted in vitro symbiotic culture from seeds with
six fungal isolates. Two Serendipitaceae and two Tulasnellaceae isolates,
including SE1 and TU27, significantly promoted seed germination and seedling
development. These results indicate that D. officinale is mainly associated with
Tulasnellaceae and Serendipitaceae as its main fungal partners, which strongly
induced seed germination and seedling development in vitro, suggesting their
association with D. officinale through its life cycle.
KEYWORDS
lithophytes, orchid, Serendipitaceae, Tulasnellaceae, wild populations, epiphytes
Frontiers in Ecology and Evolution 01 frontiersin.org
34
fevo-10-994641 October 6, 2022 Time: 11:0 # 2
Zhang et al. 10.3389/fevo.2022.994641
Introduction with humus and moss (Zhu et al., 2009; Hou et al., 2012).
Although various fungal isolates from Dendrobium species
Orchidaceae is among the largest angiosperm plant families, promote seed germination and seedling development in D.
comprising more than 28,000 species (Christenhusz and officinale (Guo and Xu, 1991; Wu et al., 2012; Shao et al., 2019;
Byng, 2016), 69% of which are epiphytic (Zotz, 2013). Wang et al., 2021), the mycorrhizal associations of its wild
Orchids form symbiotic associations with mycorrhizal fungi, populations remain unclear. Mycorrhizal fungi can vary among
in which fungal hyphae penetrate living plant cells to form sites and substrates (lithophytic or epiphytic individuals).
intracellular pelotons (Smith and Read, 2008). Orchid seeds D. officinale is an endangered species due to over collection;
are highly dependent on mycorrhizal fungi for carbon, therefore, understanding the mycorrhizal associations of its wild
nitrogen, and other nutrients during seed germination; such populations is important for its conservation.
associations generally persist in mature plants (Rasmussen In this study, we also examined the effects of major and
and Rasmussen, 2009). Most orchid mycorrhizal fungi (OMF) minor mycorrhizal fungal associations on seed germination,
belong to a rhizoctonia aggregate, a polyphyletic group protocorm formation, and seedling development. We examined
of fungi belonging to a combination of Tulasnellaceae, 36 wild D. officinale individuals (27 lithophytic and 9 epiphytic)
Serendipitaceae, and Ceratobasidiaceae (Rasmussen, 2002; sampled from six sites and conducted in vitro symbiotic seed
Dearnaley et al., 2012). Orchid mycorrhizal associations germination testing using D. officinale seeds and six fungal
do not always remain stable throughout the plant life isolates obtained from roots.
cycle, with some orchids continuing their association with
the same fungi and others switching partners from the
seed germination to adult stages (Ventre Lespiaucq et al., Materials and methods
2021). Habitat type, which can be terrestrial, epiphytic,
or lithophytic, also often affects mycorrhizal communities Sample collection
(Xing et al., 2019; Qin et al., 2020). OMF may have
a significant impact on the distribution, abundance, and In all, 36 D. officinale individuals were collected from six
population dynamics of orchid species (Jacquemyn et al., 2012; sites in Kochi and Kagoshima Prefectures in Japan (Table 1). To
McCormick et al., 2018). However, despite the rich diversity examine differences in mycorrhizal fungal associations between
of epiphytic orchids, far fewer studies have explored OMF epiphytic and lithophytic individuals, we collected epiphytic
associations among epiphytic orchids than among terrestrial root samples from seven tree species and lithophytic root
orchids. samples from rocks, cement bridges, and a roof. Root samples
The genus Dendrobium Swartz is among the largest (3–5 cm per plant) were washed with tap water, and hand-sliced
genera in Orchidaceae, including approximately 1,450 species sections were observed under a microscope to assess fungal
distributed in tropical and subtropical regions from India to colonization. Mycorrhizal root segments were cut into 1–2 cm
Southeast Asia, China, Japan, and Oceania (Schuiteman, 2014). fragments and stored in Tris-EDTA (TE) buffer at –20◦C for
Dendrobium species have long been studied for their economic, fungal molecular identification. Sections with living hyphal coils
medicinal, and ornamental value (Teixeira da Silva et al., 2015; were used for fungal isolation.
Teoh, 2016). Mycorrhizal associations with Dendrobium species
have also been investigated for the propagation of medicinal
species and conservation of endangered species (Chen et al., Fungal isolation
2021). However, most such studies have focused on symbiotic
culture with fungal isolates from roots, seeds, or seedlings Root sections with living hyphal coils were washed with
(Nontachaiyapoom et al., 2011; Mala et al., 2017; Maharjan et al., sterile distilled water (SDW) to remove bark debris from the root
2020), whereas mycorrhizal associations among wild orchid surface and crushed with forceps to disperse the viable hyphae
populations remain poorly understood, although a few studies coils into 100 mL SDW. Hyphal coils (pelotons) were collected
have revealed in situ associations with several wild populations using a micropipette and rinsed four times in sterile water.
(Xing et al., 2013; Rammitsu et al., 2021). For culture, these pelotons with 20–40 µL SDW were dropped
Dendrobium officinale Kimura and Migo (syn. Dendrobium onto 1.5% agar medium containing 50 ppm streptomycin
stricklandianum Rchb.f and Dendrobium tosaense Makino; Jin and tetracycline. Plates were incubated at 25 ± 1◦C for 1
and Huang, 2015) is a component of many traditional Chinese week. Fungal colonies that formed from single pelotons were
medicines and its symbionts have been well studied (Ding transferred to fresh potato dextrose agar (PDA) plates for
et al., 2008; Jin et al., 2017; Zuo et al., 2021). This species is subculture. The fungal isolates obtained in this study were
distributed from southern China to southern Japan, where it deposited in the Biological Resource Center of the National
grows on cliffs (lithophyte) or tree trunks (epiphyte) covered Institute of Technology and Evaluation (NBRC) (Table 2).
Frontiers in Ecology and Evolution 02 frontiersin.org
35
fevo-10-994641 October 6, 2022 Time: 11:0 # 3
Zhang et al. 10.3389/fevo.2022.994641
Molecular identification of mycorrhizal from the suspension as described previously (Izumitsu et al.,
fungi 2012). Polymerase chain reaction (PCR) amplification of the
internal transcribed spacer (ITS) sequences was performed
DNA was extracted from root samples as described using the fungal universal primer pairs ITS1F/ITS4 (White et al.,
previously (Rammitsu et al., 2021). Samples were crushed with 1990; Gardes and Bruns, 1993) and ITS1F/ITS4B (Gardes and
forceps to disperse hyphal coils into TE buffer. We collected Bruns, 1993). These primer pairs failed to amplify sequences
100–200 coils per fragment and homogenized these with 20 of Tulasnellaceae, which is a dominant mycorrhizal fungal
µL TE buffer using a BioMasher II homogenizer (Nippi Inc., family associated with orchids. Therefore, we also used the
Tokyo, Japan). For fungal isolate DNA, hyphae growing on Tulasnellaceae-specific primer pairs ITS5/ITS4-Tul2 (White
the culture medium were collected using a sterilized toothpick et al., 1990; Oja et al., 2015) and 5.8S-Tulngs/ITS4-Tul2 (Oja
and suspended in 50 µL TE buffer. DNA was extracted et al., 2015; Rammitsu et al., 2021). PCR amplification was
TABLE 1 Details of Dendrobium officinale and mycorrhizal samples used in this study.
Locality Site Habitata Substrate Total no. of Roots Isolates Total
no. individuals OMF
No. of No. of OMF No. of No. of OMF
individuals samples individuals isolates
Kami-shi, Kochi S1 L Cement block 3 3 8 8 2 4 4 12
Prefecture, Japan wall
E Aesculus 1 1 3 1 0 0 0 1
turbinata
Yakushima-cho, S2 L Cement bridge 5 5 9 9 0 0 0 9
Kagoshima
Prefecture, Japan E Distylium 1 1 1 0 0 0 0 0
racemosum
E Castanopsis 1 1 6 2 0 0 0 2
cuspidata
E Glochidion 1 1 2 2 0 0 0 2
obovatum
L Cement bridge 9 1 1 1 9 15 12 13
S3 L Rock wall 5 3 3 3 2 8 7 10
S4 L Cement bridge 4 4 11 8 3 3 2 10
E Athruphyllum 1 1 2 2 0 0 0 2
neriifolium
L Cement roof 1 1 3 1 0 0 0 1
E Quercus 2 1 5 5 1 2 2 7
salicina
S5 E Unknown 1 1 6 6 0 0 0 6
fallen tree
S6 E Ficus superba 1 0 0 0 1 2 1 1
aL, Lithophyte; E, Epiphyte.
TABLE 2 Fungal isolates from Dendrobium officinale used for symbiotic culture.
Family Fungal OTU Isolate ID Site no. Substrate DDBJ accession no. NBRC accession no.
Tulasnellaceae TU10 F205 S1 Cement block wall LC597350 NBRC 114085
TU22 F868 S2 Cement bridge LC683200 NBRC 115276
TU27 F763 S4 Cement bridge LC683202 NBRC 115262
Serendipitaceae SE1 F809 S4 Cement bridge LC683203 NBRC 115270
SE5 F859 S6 Ficus superba LC683204 NBRC 115275
Ceratobasidiaceae CE18 F356 S3 Rock wall LC597346 NBRC 114326
Frontiers in Ecology and Evolution 03 frontiersin.org
36
fevo-10-994641 October 6, 2022 Time: 11:0 # 4
Zhang et al. 10.3389/fevo.2022.994641
performed using MightyAmp DNA polymerase Ver.3 (TaKaRa, et al., 2000). The collected capsules were sterilized using 75%
Shiga, Japan) in a total volume of 10 µL, containing 1 µL sample ethanol and dried for 1 week using silica gel desiccant until they
DNA, 5 µL 2 × MightyAmp buffer, 5 pmol each primer, 0.2 µL had nearly ruptured. Seeds were collected from the capsules and
MightyAmp DNA Polymerase Ver.3, and 1 µL 10 × Additive stored at 5◦C until use. Seeds were sterilized with 1% sodium
for High Specificity (TaKaRa). hypochlorite solution for 3 min, sown on oatmeal agar medium
PCR amplification was performed with the following cycling (OMA; 2.5 g/L oatmeal and 15 g/L agar) and maintained at 25◦C
parameters: initial denaturation at 98◦C for 2 min, followed for 1 week for contamination checking. After 1 week without
by denaturation at 98◦C for 10 s, annealing at 58◦C for contamination, 1 cm × 1 cm discs were cut (5–10 seeds per disc)
15 s, extension at 68◦C for 40 s, for a total of 35 cycles. and transplanted to new OMA media. A total of 20 seeds on
The resulting amplicons were purified using the Fast Gene two to four discs were placed on each new medium plate. Each
Gel/PCR Extraction Kit (Nippon Genetics, Tokyo, Japan) and treatment consisted of 5–15 replicates, for a total of 100–300
sequenced using the BigDye Terminator v3.1 Cycle Sequencing seeds. A 6-mm plug of fungal culture was inoculated onto the
Kit (Thermo Fisher Scientific Baltics, Vilnius, Lithuania) and OMA medium, and the cultures were placed under a 12 h/12 h
3,130 Genetic Analyzer (Applied Biosystems, Tokyo, Japan) light/dark photoperiod at 25 1◦± C. Petri dishes without fungal
according to the manufacturer’s instructions. All ITS sequences inoculum were prepared as a control. After 90 days of culture,
were assigned to operational taxonomic units (OTUs) defined the seeds were counted under a stereomicroscope. Germination
by 97% sequence similarity. All ITS sequences were analyzed and seedling growth and development were scored on a scale of
using BLAST searches (Altschul, 1997) against the GenBank 0–5 as described previously (Stewart et al., 2003; Table 3). The
sequence database to find the closest matching sequence. The data were analyzed by one-way ANOVA and Turkey-Kramer
full-length ITS sequences of each OTU were edited using the test using IBM SPSS (ver. 27 IBM Corp., NY, USA).
ATGC v7 sequence assembly software (Genetyx, Tokyo, Japan) To confirm fungal colonization, the protocorms were
and deposited in the DNA Data Bank of Japan under accession cleared using 10% KOH solution, washed in 2% HCl, and
numbers LC597346, LC597350, and LC683198–LC683206. stained with 0.05% trypan blue in lactoglycerol, as described
previously (Phillips and Hayman, 1970), with modifications.
Stained protocorms were de-stained in lactoglycerol prior to
Phylogenetic analysis microscopic observation (Nikon Eclipse 50i, Nikon, Tokyo,
Japan).
OTUs belonging to Tulasnellaceae, Serendipitaceae and
Ceratobasidiaceae, which are known OMF, were considered
putative mycorrhizal associates and subjected to phylogenetic Results
analysis using ITS sequences. Sequences obtained from
Dendrobium species in previous studies were included in the Molecular identification of mycorrhizal
analysis (Wang et al., 2011; Shao et al., 2019; Zhang et al., 2020). fungi
The phylogenetic analysis was performed using the MEGA 11
software (Nei and Kumar, 2000; Stecher et al., 2020; Tamura
In total, 60 root samples and 34 isolates collected from 36
et al., 2021). Maximum likelihood (ML) trees were obtained
individuals from six sites were analyzed (Table 1). In total, 96
using the GTR + G + I model. Bootstrap (BS) analysis of the
fungal sequences were obtained from these samples and 79.2%
ML trees was performed using 1,000 replicates (Felsenstein,
of the sequences were OMF, including 45.8% Tulasnellaceae,
1985). All positions with < 90% site coverage was eliminated,
28.1% Serendipitaceae, 3.1% Ceratobasidiaceae, and 2.1%
i.e., < 10% of alignment gaps, missing dates, and ambiguous
Fusarium (Figure 1). Two or three different sequences were
bases were allowed at any position.
obtained from each of the six samples using different primer
Symbiotic culture TABLE 3 Seed germination and protocorm development in
Dendrobium officinale.
Six fungal isolates from D. officinale were used (Table 2). Stage description
A fungal colony of each isolate was transferred onto PDA as pre-
Stage 0 No germination, viable embryo
culture and cultured in the dark at 25 1◦± C for 7 days. Seeds
Stage 1 Enlarged embryo
were obtained from nine mature capsules from five individuals.
Stage 2 Continued embryo enlargement, rupture of testa
Seeds from four to five capsules of two or three individuals
Stage 3 Appearance of protomeristem
were mixed and used for symbiotic culture. Prior to each use,
Stage 4 Emergence of first leaf
seeds were tested using the TTC (2,3,5-triphenyl tetrazolium
Stage 5 Growing two leaves or a root
chloride) method to ensure high viability (> 90%) (Vujanovic
Frontiers in Ecology and Evolution 04 frontiersin.org
37
fevo-10-994641 October 6, 2022 Time: 11:0 # 5
Zhang et al. 10.3389/fevo.2022.994641
sets. The independent data sets of fungal sequences for root Mycorrhizal fungi were compared among sites and
samples and isolates showed that both data sets consisted substrates (Figure 2). We collected D. officinale samples from
of Tulasnellaceae, Serendipitaceae and Ceratobasidiaceae six different sites and 11 substrates. The dominant mycorrhizal
(Supplementary Figure 1). The OMF sequences were assigned fungi varied among both sites and substrates, even within the
to 10 OTUs including six Tulasnellaceae, two Serendipitaceae, same site. SE1 was the most frequently detected OTU, occurring
one Ceratobasidiaceae, and one Fusarium (Figure 2). Two in 26 samples from four sites and accounting for 27.1% of all
Fusarium sequences obtained in this study showed high detected fungal OTUs (Figure 1). The second most frequently
sequence similarity (99–100%) with Fusarium oxysporum detected OTU was TU27, which was found in 15 samples from
according to BLAST analysis. This species formed fungal two sites, accounting for 15.6%. TU22 was detected in 9 samples
coils in D. candidum root cells (Jiang et al., 2019) and has from two sites (9.4%), and TU10 in 7 samples from three sites
been sampled from D. officinale seedlings (Chen et al., 2021). (7.3%).
Therefore, we added these Fusarium sequences to the OMF The relative frequencies of Serendipitaceae and
OTUs as FU1. Tulasnellaceae differed greatly between lithophytic and
FIGURE 1
Frequency distribution of fungal sequences identified from Dendrobium officinale using 96 sequences. Identical sequences obtained from a
single sample using different primer pairs were discarded.
FIGURE 2
Binary matrix showing the relationship between the sampling sites, substrates, and detected fungal operational taxonomic units (OTUs). The
abundance of detected OTUs is indicated as a gradient from white to black. L indicates lithophytic and E indicates epiphytic habitats.
Frontiers in Ecology and Evolution 05 frontiersin.org
38
fevo-10-994641 October 6, 2022 Time: 11:0 # 6
Zhang et al. 10.3389/fevo.2022.994641
epiphytic individuals (Figure 3). Serendipitaceae accounted for (MN173026) and Sebacinales sp. SSCDO-6 from D. officinale
47.3% of the total in lithophytes (Figure 3A) and only 4.8% in (MH348615), sharing 96.9 and 97.3% ITS sequence similarity,
epiphytes (Figure 3B). By contrast, Tulasnellaceae accounted respectively.
for 43.6% in lithophytes and 95.2% in epiphytes. Among the The ITS sequences of 6 Tulasnellaceae OTUs obtained in this
10 detected OTUs, four (TU10, TU22, TU23, and TU27) were study and 44 obtained from the GenBank database were used to
present in both substrates, whereas four (TU12, SE1, CE18, and generate the phylogenetic tree (Figure 5). The second dominant
FU1) and two (TU21 and SE5) OTUs were unique to lithophytes mycorrhizal fungus, TU27, formed a monophyletic clade with
and epiphytes, respectively (Figure 3C). Serendipitaceae found four Tulasnellaceae sequences from D. officinale (MH348611,
in lithophytes consisted of only a single OTU, SE1, which was MH348612, MH348613, and MH348616), sharing 97.8–98.0%
unique to lithophytes and accounted for approximately half ITS sequence similarity, with BS = 98%. The TU22 sequence was
of the total frequency (Figure 3A). TU27 was dominant in closely related to the three mycorrhizal fungal sequences from
epiphytes, accounting for 52.4% of the total frequency, whereas D. officinale (MN545849, MN545657, and MN545858), sharing
it accounted for only 7.3% in lithophytes (Figures 3A,B). 96.7–98.2% similarity. TU12 formed a monophyletic clade with
two Tulasnella sequences from D. officinale (EF393629 and
MN544859) with BS = 99% and shared 97.0–98.8% similarity.
Phylogenetic analysis TU10 was clustered with mycorrhizal fungi from epiphytic
orchid, Ascocentrum himalaicum (JQ713573), with BS = 97%,
Phylogenetic analysis of Serendipitaceae was conducted and closely related to TU27 (BS = 82%). TU23 was clustered
using two Serendipitaceae OTUs obtained in this study with mycorrhizal fungi isolated from other epiphytic species
and 33 sequences obtained from the GenBank database (LC597355, LC568587, OL374168) with BS = 98%. TU21
(Figure 4). The most dominant mycorrhizal fungus, SE1, was closely related to epiphytic species, Liparis viridiflora
formed a monophyletic clade of Thanatephorus sp. SSCDO- (KP053821), BS = 98%, and distantly related to the other
8 (MH348617: 97.2% sequence similarity) from D. officinale Tulasnellaceae OTUs.
(as syn. D. catenatum in Zhu et al., 2009), with BS = 99%. Phylogenetic analysis of Ceratobasidiaceae was
SE5 was closely related to Sebacinales sp. from D. officinale conducted using one Ceratobasidiaceae OTU obtained in
FIGURE 3
Comparison of orchid mycorrhizal fungi (OMF) associating with lithophytic and epiphytic Dendrobium officinale individuals. Frequency
distribution of OMF sequences detected from (A) lithophytic and (B) epiphytic individuals. (C) Venn diagrams showing the numbers of OMF
OTUs.
Frontiers in Ecology and Evolution 06 frontiersin.org
39
fevo-10-994641 October 6, 2022 Time: 11:0 # 7
Zhang et al. 10.3389/fevo.2022.994641
FIGURE 4
Maximum likelihood tree for Serendipitaceae internal transcribed spacer (ITS) sequences, including those of two OTUs from this study. Symbols
indicate the origin of each sequence. Only bootstrap values ≥ 70% are shown. The tree is drawn to scale, with branch lengths reflecting the
number of substitutions per site. Auricularia auricula-judae was used as an outgroup taxon. The 35 assembled sequences were aligned, and the
final dataset included 382 bp sequences.
this study and 34 sequences obtained from the GenBank (LC278371), terrestrial orchid of Dactylorhiza (EF536969)
database (Supplementary Figure 2). The CE18 formed and three sequences from plant pathogens, Rhizoctonia
a monophyletic group with other Genbank sequences sp. AG-G (JF519837, KC825348), Ceratobasidium sp.
divided from mycobionts of epiphytic orchids containing AG-G (DQ102402), sharing 99.5–100% ITS sequence
D. officinale (JX545227), Aranda (AJ318429), Liparis similarity, with BS = 96%.
Frontiers in Ecology and Evolution 07 frontiersin.org
40
fevo-10-994641 October 6, 2022 Time: 11:0 # 8
Zhang et al. 10.3389/fevo.2022.994641
FIGURE 5
Maximum likelihood tree for Tulasnellaceae ITS sequences, including six OTUs from this study. Symbols indicate the origin of each sequence.
Only bootstrap values ≥ 70% are shown. The tree is drawn to scale, with branch lengths reflecting the number of substitutions per site.
Tulasnella alibida and Tulasnella hadrolaeliae were used as outgroup taxa. The 50 assembled sequences were aligned, and the final dataset
included 442 bp sequences.
Frontiers in Ecology and Evolution 08 frontiersin.org
41
fevo-10-994641 October 6, 2022 Time: 11:0 # 9
Zhang et al. 10.3389/fevo.2022.994641
Symbiotic culture Tulasnellaceae OTU (Rammitsu et al., 2021). Such high
specificity is also found in Dendrobium fimbriatum, which
Seeds from D. officinale were cultured symbiotically was associated with only two OTUs in 15 root samples
with six OTU isolates including three Tulasnellaceae, two from two sites (Xing et al., 2013). Mycorrhizal specificity
Serendipitaceae, and one Ceratobasidiaceae (Table 2). After may vary among Dendrobium species (Xing et al., 2017),
3 months of culture, all isolates except for CE18 promoted seed and D. officinale appears to have lower specificity than its
germination to different degrees (Table 4). Seeds inoculated with congeners.
TU22, TU27, SE1, and SE5 developed at stage 5, and TU22 and Orchid mycorrhizal communities of D. officinale varied
SE1 showed higher development rates than the other isolates. among sites in this study (Figure 2). Xing et al. (2013) also
TU10 also promoted seed germination, with seeds developing at found that D. officinale from two sites had distinct OMF
stage 4. Seeds cultured with CE18 became swollen and did not communities in Guangxi Province, China. Such community
develop past stage 2. All six isolates formed intracellular hyphal differences among sites have also been recorded in terrestrial
coils in protocorm cells (Supplementary Figure 3). orchids (Jacquemyn et al., 2012; Kohout et al., 2013; Oja et al.,
2015). There is some evidence that soil chemical characteristics
such as phosphorus, zinc, and organic matter (Kaur et al.,
Discussion 2021) and nitrogen, phosphorus, and water content (Han
et al., 2016), impact OMF communities in orchid roots and
In this study, 10 OTUs were detected as OMF in D. soils. These differences in substrate chemical and physical
officinale samples collected from six sites; eight were included characteristics may vary among sites, resulting in corresponding
in Tulasnellaceae and Serendipitaceae, accounting for 73.9% OMF community differences.
of all detected fungal sequences (Figures 1, 2). This implies Mycorrhizal community composition differed between
that these fungal families are the most dominant fungal lithophytic and epiphytic individuals in this study (Figure 3).
partners of D. officinale. Most previous studies of D. officinale The dominant mycorrhizal fungus among lithophytes was
sampled from southern China have also found Tulasnellaceae a Serendipitaceae OTU, SE1, whereas that of epiphytes was
and/or Serendipitaceae in roots or protocorms germinated a Tulasnellaceae OTU, TU27. Distinct OMF communities
in situ (Wang et al., 2011; Wu et al., 2012; Shao et al., between lithophytic and epiphytic individuals were also
2019). These results imply that independent of its distribution recorded for the orchid Coelogyne viscosa (Xing et al., 2015).
range, D. officinale has mycorrhizal associations mainly Among lithophytic and epiphytic individuals of Coelogyne
with Tulasnellaceae and Serendipitaceae fungi. Phylogenetic corymbosa, Serendipitaceae fungi contributed a relatively large
analysis showed that three of the six Tulasnellaceae OTUs portion of the OTU communities specific to lithophytic orchids
and two Serendipitaceae OTUs showed greater than 97% (Qin et al., 2020). Yokoya et al. (2021) surveyed 11 growing
sequence similarity to mycorrhizal fungi associated with D. Cynorkis orchid species within lithophytic and terrestrial
officinale from China (Figures 4, 5). These OTUs include habitats and found that Serendipitaceae OTUs were frequently
the most frequent OTUs detected in this study, SE1 and found in species inhabiting granite/rock, whereas Tulasnellaceae
TU27 (Figure 1). Although D. officinale is associated with a OTUs were found in both habitat types; they also reported
wide range of basidiomycetous mycorrhizal partners, its main that most Serendipitaceae OTUs were found in the habitat
fungal partners may be widely shared among D. officinale with higher phosphorus and nitrogen content, which may
populations. In Dendrobium okinawense, 11 mature plants indicate that Serendipitaceae prefers soil conditions with high
from four sites were predominantly associated with a single phosphorus and nitrogen levels. These differences in nutrient
TABLE 4 Effects of fungal isolates on Dendrobium officinale seed germination and protocorm development after 3 months of culture.
Treatment Ratio of seed germination and protocorm development (%)a
Stage 0 Stage 1 Stage 2 Stage 3 Stage 4 Stage 5
Controlb 13.72 ± 1.67a 10.40 ± 1.42b 75.71 ± 2.40d 0.17 ± 0.17a 0.00 ± 0.00a 0.00 ± 0.00a
TU10 10.19 ± 3.35a 1.54 ± 1.54a 13.11 ± 1.45b 54.76 ± 5.25c 20.40 ± 4.83b 0.00 ± 0.00a
TU22 13.30 ± 1.90a 0.00 ± 0.00a 0.69 ± 0.46a 0.00 ± 0.00a 8.77 ± 8.77ab 77.25 ± 7.93c
TU27 13.14 ± 1.71a 3.47 ± 1.67a 9.92 ± 3.86ab 9.77 ± 2.88ab 16.83 ± 3.66ab 46.87 ± 9.22b
SE1 12.45 ± 2.20a 0.21 ± 0.21a 1.74 ± 0.74ab 3.48 ± 1.10ab 10.18 ± 1.65ab 71.94 ± 3.46bc
SE5 13.20 ± 1.92a 0.99 ± 0.74a 5.94 ± 2.09ab 11.28 ± 3.41b 22.84 ± 3.32b 45.75 ± 6.79b
CE18 10.60 ± 2.21a 27.36 ± 2.52c 62.04 ± 2.63c 0.00 ± 0.00a 0.00 ± 0.00a 0.00 ± 0.00a
aGermination percentage (mean ± SE, n = 5–15) within columns marked by different letters are significantly different at P < 0.05 (Tukey Kramer).
bSeeds without fungal inoculation.
Frontiers in Ecology and Evolution 09 frontiersin.org
42
fevo-10-994641 October 6, 2022 Time: 11:0 # 10
Zhang et al. 10.3389/fevo.2022.994641
conditions may contribute to OMF community differences as its main fungal partners, such as SE1 and TU27. These
between lithophytic and epiphytic individuals in D. officinale. fungal partners induced D. officinale seed germination and
Germination of D. officinale seeds was promoted by five seedling development in vitro, suggesting that they are
of the six OMF used in this study (Table 4). Although all its main fungal partners throughout its life cycle. The
six OMF formed coiled fungal hyphae within the protocorm in situ seed baiting technique, which was proposed as an
cells according to histological observation (Supplementary effective and simple technique for obtaining seed germination-
Figure 3), the ability to promote seed germination varied enhancing fungi in situ (Rasmussen and Whigham, 1993),
greatly among OMF (Table 4). All OMF, except CE18, exhibited will contribute to a more comprehensive understanding of
germination-promoting effects, and seedlings with TU22, TU27, the mycorrhizal associations of D. officinale throughout its life
SE1, and SE5 were able to reach stage 5. Phylogenetic analysis cycle. Our results show that the OMF community differed
showed that the sequences of these four OTUs shared ≥ 97% between lithophytic and epiphytic individuals, suggesting that
sequence similarity with fungal isolates obtained in previous mycorrhizal specificity may vary by habitat type. Our findings
studies of D. officinale (Figures 4, 5). Tulasnellaceae sp. SSCDO- contribute to understanding of mycorrhizal associations among
7, which is closely related to TU27 (Figure 5), strongly wild Dendrobium species, the conservation of endangered
promotes seed germination in D. officinale (Shao et al., 2019). Dendrobium species, and the industrial production of medicinal
Tulasnellaceae sp. TPYD1, TPYD2, and TPYD3, which share Dendrobium species.
97–98% sequence similarity with TU22, promote the growth
of D. officinale seedlings produced in vitro (Chen et al., 2021).
Serendipitaceae isolates SSCDO-8 and SSCDO-6, which are
closely related to SE1 and SE5, respectively, also induce D. Data availability statement
officinale seed germination and seedling growth (Shao et al.,
2019). Our molecular analysis showed that SE1, TU27, and Sequence data have been deposited in DNA Data Bank of
TU22 were the most frequent fungal OTUs in adult individuals Japan (DDBJ) under accession numbers LC597346, LC597350,
(Figure 1), and these fungi promoted seed germination and and LC683198–LC683206.
protocorm development (Table 4). These results suggest that
the main fungal partners at the adult stage can induce seed
germination and support seedling development in D. officinale.
Seedlings with TU10 developed at stage 4 after 3 months Author contributions
of culture (Table 4) and continued growth, reaching stage
5 after 6 months (data not shown). This fungus induced YO-T and TY involved in the study conception and design.
seed germination, but with slower seedling growth than other KT contributed to the field survey. LZ and KR performed
effective fungal isolates. By contrast, seedlings with CE18 the sampling, experiments, data collection, and analysis. LZ
reached stage 2 after 2 months and showed no further growth, and YO-T wrote the manuscript. All authors commented on
despite our detection of coiled fungal hyphae in protocorm previous versions of the manuscript, read, and approved the
cells (Supplementary Figure 3). Hence, this fungal strain final manuscript.
appears not to contribute to seed germination in D. officinale.
Ceratobasidiaceae fungi are considered important partners
of other orchid genera such as Goodyera (Shefferson et al., Funding
2010), Tolumnia (Otero et al., 2004), and Pterostylis (Bougoure
et al., 2005; Bonnardeaux et al., 2007). Phylogenetic analysis This research was funded by JSPS KAKENHI (grant
showed that CE18 was closely related to OMF from epiphytic nos. 21K06306 to YO-T, and 18H02500 to TY) and
and terrestrial orchids (Supplementary Figure 2). However, Research Grant from Yakushima Environmental and Cultural
it has rarely been sampled from D. officinale roots. Because Foundation to KR.
all root samples bearing the Ceratobasidiaceae sequence were
accompanied by Tulasnellaceae or Serendipitaceae sequences in
this study, Ceratobasidiaceae may not a main fungal partner for
D. officinale. Acknowledgments
We are very grateful to T. Hashimoto, A. Maeda, T.
Conclusion Saito, and T. Tetsuka for sampling and K. Watanabe for
DNA analysis. The DNA sequencing analyses were made
In conclusion, our results demonstrate that D. officinale using a Genetic Analyzer at Analytical Research Center for
mainly forms OMF with Tulasnellaceae and Serendipitaceae Experimental Sciences, Saga University.
Frontiers in Ecology and Evolution 10 frontiersin.org
43
fevo-10-994641 October 6, 2022 Time: 11:0 # 11
Zhang et al. 10.3389/fevo.2022.994641
Conflict of interest organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
The authors declare that the research was conducted in the claim that may be made by its manufacturer, is not guaranteed
absence of any commercial or financial relationships that could or endorsed by the publisher.
be construed as a potential conflict of interest.
Supplementary material
Publisher’s note
The Supplementary Material for this article can be
All claims expressed in this article are solely those of the found online at: https://www.frontiersin.org/articles/10.3389/
authors and do not necessarily represent those of their affiliated fevo.2022.994641/full#supplementary-material
References
Altschul, S. (1997). Gapped BLAST and PSI-BLAST: A new generation of Jin, X., and Huang, L. (2015). Proposal to conserve the name Dendrobium
protein database search programs. Nucleic Acids Res. 25, 3389–3402. doi: 10.1093/ officinale against D. stricklandianum, D. tosaense, and D. pere-fauriei
nar/25.17.3389 (Orchidaceae). Taxon 64, 385–386. doi: 10.12705/642.19
Bonnardeaux, Y., Brundrett, M., Batty, A., Dixon, K., Koch, J., and Jin, Z., Li, D., Liu, T., Zhang, Z., Su, C., Wang, Y., et al. (2017). Cultural
Sivasithamparam, K. (2007). Diversity of mycorrhizal fungi of terrestrial orchids: endophytic fungi associated with Dendrobium officinale: Identification, diversity
Compatibility webs, brief encounters, lasting relationships and alien invasions. estimation and their antimicrobial potential. Curr. Sci. 16, 90–97.
Mycol. Res. 111, 51–61. doi: 10.1016/j.mycres.2006.11.006 Kaur, J., Phillips, C., and Sharma, J. (2021). Host population size is linked
Bougoure, J. J., Bougoure, D. S., Cairney, J. W. G., and Dearnaley, J. D. W. to orchid mycorrhizal fungal communities in roots and soil, which are
(2005). ITS-RFLP and sequence analysis of endophytes from Acianthus, Caladenia shaped by microenvironment. Mycorrhiza 31, 17–30. doi: 10.1007/s00572-020-00
and Pterostylis (Orchidaceae) in southeastern Queensland. Mycol. Res. 109, 452– 993-5
460. doi: 10.1017/S095375620500225X Kohout, P., Tìšitelová, T., Roy, M., Vohník, M., and Jersáková, J. (2013). A
Chen, D.-Y., Wang, X.-J., Li, T.-Q., Li, N.-Q., and Gao, J.-Y. (2021). In situ diverse fungal community associated with Pseudorchis albida (Orchidaceae) roots.
seedling baiting to isolate plant growth-promoting fungi from Dendrobium Fungal Ecol. 6, 50–64. doi: 10.1016/j.funeco.2012.08.005
officinale, an over-collected medicinal orchid in China. Glob. Ecol. Conserv.
28:e01659. doi: 10.1016/j.gecco.2021.e01659 Maharjan, S., Thakuri, L. S., Thapa, B. B., Pradhan, S., Pant, K. K., Joshi, G. P.,et al. (2020). In vitro propagation of the endangered orchid Dendrobium chryseum
Christenhusz, M. J. M., and Byng, J. W. (2016). The number of known plants Rolfe from protocorms culture. Nepal J. Sci. Technol. 19, 39–47. doi: 10.3126/njst.
species in the world and its annual increase. Phytotaxa 261, 201–217. doi: 10. v19i1.29737
11646/phytotaxa.261.3.1 Mala, B., Kuegkong, K., Sa-ngiaemsri, N., and Nontachaiyapoom, S. (2017).
Dearnaley, J. D. W., Martos, F., and Selosse, M. A. (2012). Orchid mycorrhizas: Effect of germination media on in vitro symbiotic seed germination of three
Molecular ecology, physiology, evolution and conservation aspects. Fungal Assoc. Dendrobium orchids. S. Afr. J. Bot. 112, 521–526. doi: 10.1016/j.sajb.2017.0
9, 207–230. doi: 10.1007/978-3-642-30826-0_12 5.008
Ding, G., Zhang, D., Ding, X., Zhou, Q., Zhang, W., and Li, X. (2008). Genetic McCormick, M. K., Whigham, D. F., and Canchani-Viruet, A. (2018).
variation and conservation of the endangered Chinese endemic herb Dendrobium Mycorrhizal fungi affect orchid distribution and population dynamics. New Phytol.
officinale based on SRAP analysis. Plant Syst. Evol. 276, 149–156. doi: 10.1007/ 219, 1207–1215. doi: 10.1111/nph.15223
s00606-008-0068-1 Nei, M., and Kumar, S. (2000). Molecular evolution and phylogenetics. New York,
Felsenstein, J. (1985). Confidence limits on phylogenies: An approach using the NY: Oxford University Press.
bootstrap. Evolution 39, 783–791. doi: 10.1111/j.1558-5646.1985.tb00420.x Nontachaiyapoom, S., Sasirat, S., and Manoch, L. (2011). Symbiotic seed
Gardes, M., and Bruns, T. D. (1993). ITS primers with enhanced specificity for germination of Grammatophyllum speciosum Blume and Dendrobium draconis
basidiomycetes - application to the identification of mycorrhizae and rusts. Mol. Rchb. f., native orchids of Thailand. Sci. Hortic. 130, 303–308. doi: 10.1016/j.
Ecol. 2, 113–118. doi: 10.1111/j.1365-294X.1993.tb00 scienta.2011.06.040
Guo, S. X., and Xu, J. T. (1991). Studies on the effects of fungi on the course of Oja, J., Kohout, P., Tedersoo, L., Kull, T., and Kõljalg, U. (2015). Temporal
seed germination of Dendrobium lohohens and Dendrobium candidum. Acta Acad. patterns of orchid mycorrhizal fungi in meadows and forests as revealed by 454
Med. Sin. 13, 46–49. pyrosequencing. New Phytol. 205, 1608–1618. doi: 10.1111/nph.13223
Han, J. Y., Xiao, H., and Gao, J. (2016). Seasonal dynamics of mycorrhizal fungi Otero, J. T., Ackerman, J. D., and Bayman, P. (2004). Differences in mycorrhizal
in Paphiopedilum spicerianum (Rchb. f) Pfitzer — A critically endangered orchid preferences between two tropical orchids. Mol. Ecol. 13, 2393–2404. doi: 10.1111/
from China. Glob. Ecol. Conserv. 6, 327–338. doi: 10.1016/j.gecco.2016.03.011 j.1365-294X.2004.02223.x
Hou, B., Tian, M., Luo, J., Ji, Y., Xue, Q., and Ding, X. (2012). Genetic diversity Phillips, J. M., and Hayman, D. S. (1970). Improved procedures for clearing
assessment and ex situ conservation strategy of the endangered Dendrobium roots and staining parasitic and vesicular-arbuscular mycorrhizal fungi for rapid
officinale (Orchidaceae) using new trinucleotide microsatellite markers. Plant Syst. assessment of infection. Trans. Br. Mycol. Soc. 55, 18–29. doi: 10.1016/S0007-
Evol. 298, 1483–1491. doi: 10.1007/s00606-012-0651-3 1536(70)80110-3
Izumitsu, K., Hatoh, K., Sumita, T., Kitade, Y., Morita, A., Tanaka, C., et al. Qin, J., Zhang, W., Zhang, S. B., and Wang, J. H. (2020). Similar mycorrhizal
(2012). Rapid and simple preparation of mushroom DNA directly from colonies fungal communities associated with epiphytic and lithophytic orchids of
and fruiting bodies for PCR. Mycoscience 53, 396–401. Coelogyne corymbosa. Plant Diversity 42, 362–369. doi: 10.1016/j.pld.2020.
Jacquemyn, H., Brys, R., Lievens, B., and Wiegand, T. (2012). Spatial variation in 07.005
below-ground seed germination and divergent mycorrhizal associations correlate Rammitsu, K., Abe, S., Abe, T., Kotaka, N., Kudaka, M., Kudaka, N., et al.
with spatial segregation of three co-occurring orchid species. J. Ecol. 100, 1328– (2021). The endangered epiphytic orchid Dendrobium okinawense has a highly
1337. doi: 10.1111/j.1365-2745.2012.01998.x specific mycorrhizal association with a single Tulasnellaceae fungus. J. For. Res.
Jiang, J., Zhang, K., Cheng, S., Nie, Q., Zhou, S. X., Chen, Q., et al. (2019). 26, 215–221. doi: 10.1080/13416979.2021.1876587
Fusarium oxysporum KB-3 from Bletilla striata: An orchid mycorrhizal fungus. Rasmussen, H. N. (2002). Recent developments in the study of orchid
Mycorrhiza 29, 531–540. doi: 10.1007/s00572-019-00904-3 mycorrhiza. Plant Soil 244, 149–163. doi: 10.1023/A:1020246715436
Frontiers in Ecology and Evolution 11 frontiersin.org
44
fevo-10-994641 October 6, 2022 Time: 11:0 # 12
Zhang et al. 10.3389/fevo.2022.994641
Rasmussen, H. N., and Rasmussen, F. N. (2009). Orchid mycorrhiza: Wang, H., Fang, H., Wang, Y., Duan, L., and Guo, S. (2011). In situ seed baiting
Implications of a mycophagous life style. Oikos 118, 334–345. doi: 10.1111/j.1600- techniques in Dendrobium officinale Kimura et Migo and Dendrobium nobile
0706.2008.17116.x Lindl.: The endangered Chinese endemic Dendrobium (Orchidaceae). World J.
Rasmussen, H. N., and Whigham, D. F. (1993). Seed ecology of dust seeds in situ: Microbiol. Biotechnol. 27, 2051–2059. doi: 10.1007/s11274-011-0667-9
A new study technique and its application in terrestrial orchids. Am. J. Bot. 80, Wang, X. J., Wu, Y. H., Ming, X. J., Wang, G., and Gao, J. Y. (2021). Isolating
1374–1378. doi: 10.1002/j.1537-2197.1993.tb15381.x ecological-specific fungi and creating fungus-seed bags for epiphytic orchid
Schuiteman, A. (2014). Genera orchidacearum: Orchidoideae, Vol. 6, eds A. M. conservation. Glob. Ecol. Conserv. 28:e01714. doi: 10.1016/j.gecco.2021.e01714
Pridgeon, P. J. Cribb, and M. W. Chase (Oxford: Oxford University Press), 51–73. White, T. J., Bruns, T., Lee, S. J. W. T., and Taylor, J. (1990). Amplification and
Shao, S. C., Xi, H. P., and Mohandass, D. (2019). Symbiotic mycorrhizal fungi direct sequencing of fungal ribosomal RNA genes for phylogenetics. PCR Protoc.
isolated via ex situ seed baiting induce seed germination of Dendrobium catenatum 18, 315–322.
lindl. (Orchidaceae). Appl. Ecol. Environ. Res. 17, 9753–9771. doi: 10.15666/aeer/ Wu, H. F., Song, X. Q., and Liu, H. X. (2012). Ex-situ symbiotic seed germination
1704_97539771 of Dendrobium catenatum. Acta Ecol. Sin. 32, 2491–2497.
Shefferson, R. P., Cowden, C. C., McCormick, M. K., Yukawa, T., Ogura-Tsujita, Xing, X., Gai, X., Liu, Q., Hart, M. M., and Guo, S. (2015). Mycorrhizal fungal
Y., and Hashimoto, T. (2010). Evolution of host breadth in broad interactions: diversity and community composition in a lithophytic and epiphytic orchid.
Mycorrhizal specificity in East Asian and North American rattlesnake plantains Mycorrhiza 25, 289–296. doi: 10.1007/s00572-014-0612-5
(Goodyera spp.) and their fungal hosts. Mol. Ecol. 19, 3008–3017. doi: 10.1111/j.
1365-294X.2010.04693.x Xing, X., Jacquemyn, H., Gai, X., Gao, Y., Liu, Q., Zhao, Z., et al. (2019). Theimpact of life form on the architecture of orchid mycorrhizal networks in tropical
Smith, S. E., and Read, D. J. (2008). Mycorrhizal symbiosis, 3rd Edn. San Diego, forest. Oikos 128, 1254–1264. doi: 10.1111/oik.06363
CA: Academic Press. Xing, X., Ma, X., Deng, Z., Chen, J., Wu, F., and Guo, S. (2013). Specificity and
Stecher, G., Tamura, K., and Kumar, S. (2020). Molecular evolutionary genetics preference of mycorrhizal associations in two species of the genus Dendrobium
analysis (MEGA) for macOS. Mol. Biol. Evol. 37, 1237–1239. doi: 10.1093/molbev/ (Orchidaceae). Mycorrhiza 23, 317–324. doi: 10.1007/s00572-012-0473-8
msz312 Xing, X., Ma, X., Men, J., Chen, Y., and Guo, S. (2017). Phylogenetic constrains
Stewart, S. L., Zettler, L. W., Minso, J., and Brown, P. M. (2003). Symbiotic on mycorrhizal specificity in eight Dendrobium (Orchidaceae) species. Sci. China
germination and reintroduction of Spiranthes brevilabris Lindley, an endangered Life Sci. 60, 536–544. doi: 10.1007/s11427-017-9020-1
orchid native to Florida. Selbyana 24, 64–70. Yokoya, K., Jacob, A. S., Zettler, L. W., Kendon, J. P., Menon, M., Bell, J., et al.
Tamura, K., Stecher, G., and Kumar, S. (2021). MEGA11: Molecular (2021). Fungal diversity of selected habitat specific Cynorkis species (Orchidaceae)
evolutionary genetics analysis version 11. Mol. Biol. Evol. 38, 3022–3027. doi: in the Central Highlands of Madagascar. Microorganisms 9:792. doi: 10.3390/
10.1093/molbev/msab120 microorganisms
Teixeira da Silva, J. A., Tsavkelova, E. A., Ng, T. B., Parthibhan, S., Zhang, Y., Li, Y. Y., Chen, X. M., Guo, S. X., and Lee, Y. I. (2020). Effect
Dobránszki, J., Cardoso, J. C., et al. (2015). Asymbiotic in vitro seed propagation of different mycobionts on symbiotic germination and seedling growth of
of Dendrobium. Plant Cell Rep. 34, 1685–1706. doi: 10.1007/s00299-015- Dendrobium officinale, an important medicinal orchid. Bot. Stud. 61:2. doi: 10.
1829-2 1186/s40529-019-0278-6
Teoh, E. S. (2016). Medicinal orchids of Asia. Cham: Springer. Zhu, G. H., Ji, Z. H., Wood, J. J., and Wood, H. P. (2009). Dendrobium. Flora of
Ventre Lespiaucq, A., Jacquemyn, H., Rasmussen, H. N., and Méndez, M. China 25, 367–397.
(2021). Temporal turnover in mycorrhizal interactions: A proof of concept with Zotz, G. (2013). The systematic distribution of vascular epiphytes - a critical
orchids. New Phytol. 230, 1690–1699. doi: 10.1111/nph.17291 update. Bot. J. Linn. Soc. 171, 453–481. doi: 10.1111/boj.12010
Vujanovic, V., St-Arnaud, M., Barabé, D., and Thibeault, G. (2000). Viability Zuo, J., Zu, M., Liu, L., Song, X., and Yuan, Y. (2021). Composition and
testing of orchid seed and the promotion of colouration and germination. Ann. diversity of bacterial communities in the rhizosphere of the Chinese medicinal
Bot. 86, 79–86. doi: 10.1006/anbo.2000.1162 herb Dendrobium. BMC Plant Biol. 21:127. doi: 10.1186/s12870-021-02893-y
Frontiers in Ecology and Evolution 12 frontiersin.org
45
fevo-10-1058550 November 21, 2022 Time: 10:9 # 1
TYPE Original Research
PUBLISHED 23 November 2022
DOI 10.3389/fevo.2022.1058550
Delimiting species in the
OPEN ACCESS taxonomically challenging
EDITED BY
Pavel Kindlmann, orchid section Pseudophrys:
Charles University, Czechia
REVIEWED BY Bayesian analyses of genetic
Przemyslaw Baranow,
University of Gdańsk, Poland
Tiiu Kull, and phenotypic data
Estonian University of Life Sciences,
Estonia
*CORRESPONDENCE Nina Joffard1, Bruno Buatois2, Véronique Arnal2, Errol Véla3,
Bertrand Schatz
bertrand.schatz@cefe.cnrs.fr Claudine Montgelard2† and Bertrand Schatz2*†
†These authors have contributed 1Univ. Lille, CNRS, UMR 8198–Evo-Eco-Paleo, Lille, France, 2Centre d’Ecologie Fonctionnelle et
equally to this work and share last Evolutive (CEFE), UMR 5175, CNRS, Université de Montpellier, Université Paul Valéry Montpellier,
authorship EPHE, Montpellier, France, 3AMAP, Université de Montpellier, CIRAD, CNRS, INRAE, IRD, Montpellier,
France
SPECIALTY SECTION
This article was submitted to
Conservation and Restoration Ecology,
a section of the journal Accurate species delimitation is critical for biodiversity conservation.
Frontiers in Ecology and Evolution
Integrative taxonomy has been advocated for a long time, yet tools allowing
RECEIVED 30 September 2022
24 October 2022 true integration of genetic and phenotypic data have been developed quiteACCEPTED
PUBLISHED 23 November 2022 recently and applied to few models, especially in plants. In this study,
CITATION we investigated species boundaries within a group of twelve Pseudophrys
Joffard N, Buatois B, Arnal V, Véla E, taxa from France by analyzing genetic, morphometric and chemical (i.e.,
Montgelard C and Schatz B (2022)
Delimiting species floral scents) data in a Bayesian framework using the program integrated
in the taxonomically challenging Bayesian Phylogenetics and Phylogeography (iBPP). We found that these
orchid section Pseudophrys: Bayesian
analyses of genetic and phenotypic twelve taxa were merged into four species when only genetic data were used,
data. while most formally described species were recognized as such when only
Front. Ecol. Evol. 10:1058550.
doi: 10.3389/fevo.2022.1058550 phenotypic (either morphometric or chemical) data were used. The result of
the iBPP analysis performed on both genetic and phenotypic data supports
COPYRIGHT
© 2022 Joffard, Buatois, Arnal, Véla, the proposal to merge Ophrys bilunulata and O. marmorata on the one
Montgelard and Schatz. This is an hand, and O. funerea and O. zonata on the other hand. Our results show
open-access article distributed under
the terms of the Creative Commons that phenotypic data are particularly informative in the section Pseudophrys
Attribution License (CC BY). The use, and that their integration in a model-based method significantly improves
distribution or reproduction in other
forums is permitted, provided the the accuracy of species delimitation. We are convinced that the integrative
original author(s) and the copyright taxonomic approach proposed in this study holds great promise to conduct
owner(s) are credited and that the
original publication in this journal is taxonomic revisions in other orchid groups.
cited, in accordance with accepted
academic practice. No use, distribution
KEYWORDS
or reproduction is permitted which
does not comply with these terms. integrative taxonomy, species delimitation, iBPP, floral scents, orchids
Introduction
Accurately delimiting species is of critical importance for many fields of
research in biology, including conservation biology. Species are commonly defined
as independently evolving linages that can be delimited using various criteria (Hey,
2006; De Queiroz, 2007). As any single line of evidence may fail at detecting species
Frontiers in Ecology and Evolution 01 frontiersin.org
46
fevo-10-1058550 November 21, 2022 Time: 10:9 # 2
Joffard et al. 10.3389/fevo.2022.1058550
boundaries (Knowles and Carstens, 2007), many authors have floral scents may cause pollinator shifts, which may in turn
advocated the use of an integrative approach combining mediate reproductive isolation between conspecific populations
several lines of evidence, both genetic and phenotypic and drive speciation (Sedeek et al., 2014). Distinct views on
(Dayrat, 2005; Will et al., 2005; Padial et al., 2010; Pires which criteria should be used to delimit species has led to the
and Marinoni, 2010). However, until recently, genetic and recognition of dozens (Devey et al., 2008) versus hundreds
phenotypic data were almost always integrated in a purely (Paulus, 2006) of Ophrys species. In addition, even authors who
qualitative way, as no quantitative methods were available for favor the same criteria sometimes disagree on where along the
processing simultaneously both data types (Schlick-Steiner et al., speciation continuum independently evolving lineages should
2010; Yeates et al., 2011). Fortunately, model-based species be recognized as species, i.e., “splitters” (e.g., Devillers and
delimitation methods, which were originally developed for DNA Devillers-Terschuren, 1994; Delforge, 2016) versus “lumpers”
sequences (Fujita et al., 2012; Naciri and Linder, 2015), were (e.g., Pedersen and Faurholdt, 2007; Kühn et al., 2020). In this
later extended to integrate quantitative traits (Guillot et al., 2012; context, model-based species delimitation methods integrating
Solís-Lemus et al., 2015), thereby improving objectivity and genetic and phenotypic data could be particularly helpful.
repeatability of integrative species delimitation. Such methods In this study, we aim at delimiting species through the
have been applied to various animal (Huang and Knowles, integration of molecular markers, morphometric characters
2016; Pyron et al., 2016; Olave et al., 2017; Núñez et al., and floral scents in a group of twelve Pseudophrys taxa. We
2022) and plant (Yang et al., 2019; Zhang et al., 2020) clades compare species boundaries based on genetic and phenotypic
and have proven useful in several cases. Because model-based data alone or in combination and we discuss the potential of
species delimitation methods may cause oversplitting when integrative taxonomy in solving long-standing debates about
solely based on genetic data (Sukumaran and Knowles, 2017; Ophrys taxonomy.
Mason et al., 2020), combining the latter with phenotypic data
may provide more conservative estimates of species numbers
(e.g., Pyron et al., 2016). Conversely, in recently radiating Materials and methods
clades, in which species often lack clear genetic differentiation,
integrating morphological or ecological data may increase the Studied species and populations
power to detect species boundaries (e.g., Edwards and Knowles,
2014; Solís-Lemus et al., 2015). The monophyletic section Pseudophrys Godfery comprises
Hyperdiverse clades deserve particular conservation twelve groups, each of them including one to twelve taxa
attention but may be taxonomically challenging. This is, for (Delforge, 2016). Here, we focused on the twelve Pseudophrys
example, the case of the Orchidaceae family, which comprises taxa that are described in France (Table 1 and Figure 1;
more than 30,000 named species [Plants of the World Online Bournérias and Prat, 2005). Among them, eight belong to
[POWO], 2022], including some of the most threatened species the O. fusca group (namely O. bilunulata, O. delforgei subsp.
in the world (Fay, 2018), but in which species boundaries are “O. forestieri” sensu neotypus 1999, O. funerea, O. lupercalis,
sometimes blurred (Barrett and Freudenstein, 2011; Pessoa O. marmorata, O. peraiolae, O. sulcata, and O. zonata), one
et al., 2012). Within this family, the Mediterranean genus Ophrys to the O. iricolor group (O. eleonorae), two to the O. lutea
L. is of particular interest, due to its high level of ecological group (namely O. corsica and O. lutea) and one to the
specialization and endemism rate, but it is also considered as O. omegaifera group (O. vasconica). These twelve taxa differ
a textbook example of taxonomic confusion (Bertrand et al., in their geographical distribution, some of them being widely
2021; Cuypers et al., 2022), which may affect its conservation distributed (e.g., O. bilunulata, O. lupercalis, and O. lutea), while
(Agapow et al., 2004; Pillon and Chase, 2007; Vereecken et al., others have restricted distribution areas, e.g., in South-eastern
2010; Schatz et al., 2014). Some of this confusion arises from France (O. deforgei), South-western France and Northern Spain
conflicting views on which operational criteria should be used (O. vasconica) or Corsica and Sardinia (O. corsica, O. eleonorae,
to delimit species in this genus. Specifically, some authors O. funerea, O. marmorata, O. peraiolae, and O. zonata). By
support that taxa should have achieved reciprocal monophyly contrast, these twelve taxa do not strongly differ in their
(Devey et al., 2008; Bateman et al., 2011) to be considered as flowering phenology or habitats: except for O. sulcata and
“good” species, while others argue that interactions between O. vasconica, they all flower in early spring and grow in open,
Ophrys and pollinators are more informative than neutral dry habitats typical of the Mediterranean region (Bournérias
markers due to their key role in speciation (Schiestl and and Prat, 2005). Among them, O. lupercalis, O. lutea, O.
Ayasse, 2002; Ayasse et al., 2011; Vereecken et al., 2011; sulcata, and O. vasconica are regionally protected in France
Baguette et al., 2020). Indeed, Ophrys species attract one or (Bournérias and Prat, 2005) and several of them are currently
a few pollinator species (Joffard et al., 2019; Schatz et al., considered as threatened at the national or regional level, such
2020) using sex pheromones-mimicking floral scents (Schiestl as O. eleonorae (considered as endangered at the national
et al., 1999; Ayasse et al., 2003). In these species, changes in level and as critically endangered in the Corsican region) and
Frontiers in Ecology and Evolution 02 frontiersin.org
47
fevo-10-1058550 November 21, 2022 Time: 10:9 # 3
Joffard et al. 10.3389/fevo.2022.1058550
TABLE 1 Number of populations and individuals sampled for molecular, morphometric, and chemical data.
Taxon Population Molecular Morphometric Chemical
data data data
Ophrys bilunulata Risso (1844) Gruissan, 11, France 3 (1) 15 10
Clapier, 34, France 3 (1) 10 5
La Gaude, 06, France 2 (1) − −
Ophrys corsica Soleirol ex Foelsche and Foelsche (2002) Bonifacio, 2A, France* 1 25 15
Ophrys delforgei Devillers-Terschuren and Devillers (2006) Martigues, 13, France* 2 25 20
Ophrys eleonorae Paulus and Gack (2004) Antisanti, 2B, France 1 − 2
Ophrys funerea Viviani (1824) Palasca, 2B, France 3 (2) 20 9
Corte, 2B, France 3 (2) 15 9
Laconi, Sardinia, Italy 2 − −
Ophrys lupercalis Devillers and Devillers-Terschuren (1994) Armissan, 11, France* 2 0 0
Saint Bauzille de Montmel, 34, France 2 (1) 20 15
Saint-Florent, 2B, France − 5 5
Ophrys lutea Cavanilles (1793) Montferrier sur Lez, 34, France 3 (1) 10 10
Montarnaud, 34, France 3 (1) 20 10
Cassis, 13, France 1 − −
Maala, Kabylia, Algeria 1 − −
Benicolet, Valencian community, Spain 1 − −
Sempere, Valencian community, Spain 1 − −
Ophrys marmorata Foelsche and Foelsche (1998) Bonifacio, 2A, France* 5 (3) 20 8
Ophrys peraiolae Foelsche et al. (2000) Palasca, 2B, France* 3 15 8
Ophrys sulcata Devillers and Devillers-Terschuren (1994) Lapanouse, 12, France 3 (2) 25 10
Oléron, 17, France* 2 − −
Vence, 06, France 1 − −
Ophrys vasconica Delforge (1991) Belpech, 11, France 1 20 15
Ophrys zonata Devillers and Devillers-Terschuren (1994) Saint-Florent, 2B, France 3 (1) 25 15
*: Populations located at the locus classicus. (): Number of newly-published sequences.
O. marmorata (considered as vulnerable in the Corsican region) few days. DNA extraction was performed with a Plant Minikit
(IUCN et al., 2010; Delage and Hugot, 2015). (©R Quiagen). Three genes were amplified and sequenced in 52
Four hundred ninety individuals belonging to one to six individuals: the internal transcribed spacers (ITS) 1 and 2, the
populations per taxon were selected and sampled for molecular, first intron of the beta-galactosidase-like (BGP) gene and the
morphometric, or chemical analysis between 2013 and 2016 first intron of the LEAFY/FLORICULA (LFY) gene. For 36
(Table 1). Within populations, molecular, morphometric, and individuals, sequences have been published in Joffard et al.
chemical data were not collected on the same individuals as (2020), while for 16 individuals, sequences are published for the
the iBPP program (see below) requires independence of genetic first time in this study (Supplementary Table 1).
and phenotypic data. Molecular data were collected in one Polymerase chain reaction (PCR) and sequencing were
to six populations per taxon, distributed over most of their carried out as described in Joffard et al. (2020). Sequences
geographic range, in up to five individuals per population. were edited using CodonCode Aligner v.4.2.7 (CodonCode
Morphometric and chemical data were collected in one or two Corporation). Uncertainties and alleles from heterozygous
of these populations only, but on up to 25 individuals per individuals were merged into consensus sequences using
population. One population was sampled at the locus classicus International Union for Pure and Applied Chemistry (IUPAC)
(i.e., site where the species was described for the first time) for coding. Consensus sequences were aligned using the Muscle
six of these twelve taxa. algorithm (Edgar, 2004) as implemented in SeaView v.4.4.2
(Gouy et al., 2010) prior to concatenation.
A phylogenetic analysis was performed on the concatenated
Genetic data collection and analysis alignment using MrBayes v.3.1.2 (Ronquist and Huelsenbeck,
2003). Ophrys cinerophila from Samos (Greece) was used as
One leaf of one to five individual(s) per population were outgroup based on Joffard et al. (2020). The best partitioning
collected between 2014 and 2017 and dried in silica gel for a scheme and the best model for each partition was chosen
Frontiers in Ecology and Evolution 03 frontiersin.org
48
fevo-10-1058550 November 21, 2022 Time: 10:9 # 4
Joffard et al. 10.3389/fevo.2022.1058550
FIGURE 1
Photographs of the 12 French Pseudophrys taxa sampled for molecular, morphometric, and chemical data. From left to right and top to bottom:
Ophrys bilunulata, O. delforgei, O. funerea, O. lupercalis, O. marmorata, O. peraiolae, O. sulcata and O. zonata (O. fusca group), O. eleonorae
(O. iricolor group), O. corsica, O. lutea (O. lutea group), and O. vasconica (O. omegaifera group). © N. Joffard and B. Schatz.
using the Bayesian Information Criterion (BIC) as estimated by recursive manner until no further splits are possible, while
PartitionFinder v.1.1.1 (Lanfear et al., 2012). Bayesian analysis integrating priors on maximum and minimum intraspecific
was conducted with two separate runs of four Markov chain differentiation and barcode gap width. In this study, pairwise
Monte Carlo (MCMC) chains for 10 million generations with distances were computed as K2P-corrected distances. We left
tree sampling every 1,000 generations. 25% of the sampled trees the default values of 10 steps from Pmin = 0.001 to Pmax = 0.1
were discarded as burn-in, and the 75% best scoring trees were for number of steps and intraspecific differentiation, and the
used to calculate the consensus tree. default value of 1.5 for barcode gap width.
A DNA barcoding analysis was performed on the
concatenated alignment using the Automatic Barcode Gap
Detection (ABGD) website (Puillandre et al., 2012). ABGD is a Phenotypic data collection and
tool designed to infer species hypotheses based on automatized analysis
identification of barcode gaps between intra- and interspecific
pairwise distances. It aims at revealing a significant barcoding For morphometric data, fifteen to thirty-five individuals
gap in the distribution of pairwise genetic distances, reflecting a per taxon were sampled in 2015 and 2016 (Table 1). Ophrys
discontinuity between intra- and interspecific distances among eleonorae was not sampled for morphometric data as no
individuals. ABGD partitions individuals into groups in a flowering individuals were found in 2015 nor in 2016. However,
Frontiers in Ecology and Evolution 04 frontiersin.org
49
fevo-10-1058550 November 21, 2022 Time: 10:9 # 5
Joffard et al. 10.3389/fevo.2022.1058550
this species is known to be morphologically distinct from the have evolved according to the assumptions of a BM process, but
eleven other taxa in that it has a particularly long (∼15 to the results of the program iBPP have been shown to be robust
25 mm) and wide (∼10 to 20 mm) labellum (Bournérias and to such a violation (Solís-Lemus et al., 2015). The program
Prat, 2005). In each individual, twelve morphometric characters begins with a strictly bifurcating guide tree, which in our case
were measured to the nearest 0.01 mm in the field using a was constructed with the software MrBayes (see above), and
digital caliper. Four of these characters concerned the labellum, collapses internal nodes sequentially. We used a prior gamma
whose size and shape are important because they must match distribution G (2, 2,000) for τ (branch lengths) and θ (product
those of the pollinator, but the length and/or width of the of Ne the population size and µ the mutation rate) for genetic
stigmatic cavity, lateral petals, and sepals were also measured data and left the default values of 0 for σ2 (variance) and
(Supplementary Figure 1). λ (within/between species ratio) for phenotypic data (non-
For chemical data, eight to twenty individuals per taxon informative priors). A reversible MCMC analysis was ran over
were sampled for floral scents in 2014 and 2015 (Table 1) 1,00,000 generations, sampled every ten generations, with 1,000
using solid phase microextraction (SPME) (except in the case generations (10%) discarded as burn-in. Seven analyses were
of O. eleonorae in which only two individuals were sampled) performed: (i) with genetic data only, (ii) with morphometric
as described in Joffard et al. (2016). Floral scents were then data only, (iii) with chemical data only, (iv) with both genetic
analyzed by GC-MS using a Shimadzu QP2010 Plus gas and morphometric data, (v) with both genetic and chemical
©R
chromatograph-mass spectrometer with an OPTIMA 5-MS data, (vi) with both morphometric and chemical data, and
capillary column (30 m × 0.25 mm × 0.25 µm, Macherey-Nagel, (vii) with the entire dataset. Because phylogenetic relationships
Düren, Germany) and helium as carrier gas with the method between the taxa O. bilunulata, O. delforgei, and O. marmorata,
described in Joffard et al. (2016). Retention times of a series as well as between the taxa O. funerea, O. sulcata, and
of n-alkanes (Qualitative retention time mix, ASTM, Sigma O. zonata could not be resolved, several alternative topologies
©R
Aldrich ) were used to convert retention times into retention were tested for the guide tree and the topology that gave the
index. Compounds were identified based on retention index most conservative species delimitation model for these two
and mass spectra which were compared to those recorded in triplets was retained. The robustness of the results was tested by
databases (NIST, 2007, Wiley Registry 9th) and in the literature analyzing the data with both the fine tune settings of zero and
(Adams, 2007) and, for some of them, to retention index and one (Yang and Rannala, 2010), and by repeating each analysis
mass spectra of analytical standards. Peak areas were measured five times.
©R
with the software GCMSsolution (4.11) (Shimazu ).
Two partial least square discriminant analyses (PLS-
DA) were performed, one for morphometric characters and Results
one for floral scents. PLS-DA was chosen over Principal
Component Analysis (PCA) because it is suitable for data Genetic data
that are non-independent (due to allometry in the case of
morphometric characters and shared biosynthetic pathways ITS, BGP, and LFY sequences were obtained for 52, 49,
in the case of floral scents). Because variances were non- and 52 individuals, respectively (153 sequences, including
homogenous among compounds, floral scents data were centred 45 that are newly published). Sequences were obtained for
log-ratio-transformed prior to analysis. Statistical analyses were at least three individuals per taxon, except for O. corsica,
performed in R version 3.1.2 (R Development Core Team, 2008). O. eleonorae, and O. vasconica (sequences obtained for one
individual only). ITS, BGP, and LFY sequences contained 73,
562, and 603 parsimony-informative sites on 809, 948, and
IBPP species delimitation 2,210 sites, respectively. The phylogenetic tree (Figure 2) was
congruent with the one described in Joffard et al. (2020), with
A joint Bayesian inference based on genetic and phenotypic two well-supported clades, one comprising the taxa O. lutea,
data was used to delimit species using the program iBPP v.2.1.3 O. corsica, O. lupercalis, O. peraiolae, O. delforgei, O. bilunulata,
(Solís-Lemus et al., 2015). This program is an extension of the and O. marmorata and one comprising the taxa O. eleonorae,
multispecies coalescent model-based program BPP (Rannala O. vasconica, O. sulcata, O. funerea, and O. zonata. On the nine
and Yang, 2003; Yang, 2015) which includes models of evolution taxa for which several individuals were sampled for molecular
for phenotypic data under a Brownian Motion (BM) process. analyses, only two - namely O. lutea and O. peraiolae - were
Because the program assumes independence of phenotypic data, found to be monophyletic with a posterior probability of 0.98
scores on the first two (for morphometric characters) and five and 1, respectively. The AGBD method detected three species
(for floral scents) components resulting from two preliminary only: it recognized O. eleonorae as a species but merged O. lutea,
PCA were included in the analysis. Note that given the role O. corsica, O. lupercalis, O. peraiolae, O. delforgei, O. bilunulata,
of floral scents in pollinator attraction, these scents may not and O. marmorata on the one hand, and O. vasconica, O. sulcata,
Frontiers in Ecology and Evolution 05 frontiersin.org
50
fevo-10-1058550 November 21, 2022 Time: 10:9 # 6
Joffard et al. 10.3389/fevo.2022.1058550
FIGURE 2
Phylogenetic relationships between the twelve French Pseudophrys taxa represented by the 50-majority rule consensus tree from the MrBayes
analysis. Posterior probabilities are indicated at each node.
O. funerea, and O. zonata on the other hand. Mean K2P- labella compared to other species, while O. sulcata, O. funerea,
corrected distances between individuals were of 2.38 10−3× and O. zonata were characterized by long petals and sepals
substitutions per site within and 7.09 × 10−3 substitutions but a relatively small labellum, with a high length/width ratio.
per site between these species. The barcoding gap was located By contrast, O. bilunulata, O. marmorata, and O. peraiolae
between 3.00 10−3× and 4.00 × 10−3 substitutions per site were characterized by larger labella with a lower length/width
(Supplementary Figure 2). ratio. Finally, the yellow-flowered O. corsica and O. lutea were
characterized by short sepals and petals and a short but wide
labellum (Figure 3 and Supplementary Table 2).
Morphometric data
Labellum length ranged from 6.11 to 14.35 mm, with a Chemical data
mean of 9.11 (±1.45) mm, and labellum width from 5.88 to
14.23 mm, with a mean of 8.98 (±1.51) mm. Ophrys lupercalis Over one hundred VOCs were detected in the blends of the
and O. vasconica were characterized by large sepals, petals, and twelve studied taxa, mostly alkanes (19), alkenes and alkadienes
Frontiers in Ecology and Evolution 06 frontiersin.org
51
fevo-10-1058550 November 21, 2022 Time: 10:9 # 7
Joffard et al. 10.3389/fevo.2022.1058550
FIGURE 3
Partial least squares-discriminant analysis (PLS-DA) of morphometric characters measured on the twelve French Pseudophrys taxa.
(29), aldehydes (23), acids (13) and fatty acid esters (24) IBPP species delimitation
(Supplementary Table 3). Blends were dominated by alkenes
and alkadienes (58.6%) as well as alkanes (28.2%), but aldehydes Results of iBPP analyses (best species delimitation
and fatty acid esters both accounted for more than 5% of the models, with their respective species numbers and posterior
blends. The blends of the O. fusca, O. iricolor, O. omegaifera, probabilities) are summarized Figure 5 and detailed
and O. lutea groups were well differentiated, both qualitatively Supplementary Table 4.
and quantitatively (Figure 4). More precisely, taxa from the Whatever the type of data included in the analysis, the
O. fusca group generally did not produce any fatty acid esters, posterior probability of the best model never exceeded 70%,
while taxa from the O. iricolor, O. omegaifera, and O. lutea showing relatively weak support for this model compared to
groups produced significant amounts of nonyl, decyl, and octyl the next best ones. When only genetic data were considered,
esters, respectively. Within the O. fusca group, some species also the three best models (i.e., those for which the sum of posterior
had well-differentiated blends, although this differentiation was probabilities exceeded 80%) all recognized O. lutea, O. corsica,
often quantitative rather than qualitative. By contrast, several O. lupercalis, O. peraiolae, O. deforgei, O. bilunulata, and
taxa, such as the O. funerea/O. zonata pair, produced very O. marmorata as a single species. They also all considered
similar floral scents. O. eleonorae as a genuine species and merged O. funerea
Frontiers in Ecology and Evolution 07 frontiersin.org
52
fevo-10-1058550 November 21, 2022 Time: 10:9 # 8
Joffard et al. 10.3389/fevo.2022.1058550
FIGURE 4
Partial least squares-discriminant analysis (PLS-DA) of floral scents detected in the blends of the 12 French Pseudophrys taxa. Individuals were
represented along axes 1 and 3 to better visualize variation within the O. fusca group.
and O. zonata. The best model (PP = 61.86%) was a four- only, with comparable posterior probabilities (PP = 47.63
species model recognizing O. vasconica as a species but merging and 44.84%, respectively). Including molecular data in the
O. sulcata with the O. funerea/O. zonata pair. By contrast, analysis increased support for the first model (PP = 53.59%)
when only morphometric data were considered, the two best compared to the second one (PP = 39.50%). When only
models recognized most taxa as genuine species: the first phenotypic (morphometric + chemical) data were considered,
one (PP = 67.10%) merged O. bilunulata and O. marmorata, the three best models either suggested to merge both the
while the second one (PP = 21.98%) delimited twelve species. O. bilunulata/O. marmorata and O. funerea/O. zonata pairs
The same results were obtained when molecular data were (PP = 44.83%) or one only (PP = 29.04% for the model
included in the analysis, but the posterior probabilities of merging O. marmorata with O. bilunulata and 16.02% for
these two best models decreased, while that of a ten-species the one merging O. zonata with O. funerea). Finally, the
model merging both O. marmorata with O. bilunulata and same results were obtained when genetic and phenotypic data
O. zonata with O. funerea increased (Supplementary Table 4). were combined, with only slight differences in the posterior
Likewise, when only chemical data were considered, the two probability attributed to each of these three best models
best models were the ten-species model merging both pairs, compared to the previous analysis (PP = 49.88, 27.63, and
and an eleven-species model merging O. zonata with O. funerea 14.19%, respectively).
Frontiers in Ecology and Evolution 08 frontiersin.org
53
fevo-10-1058550 November 21, 2022 Time: 10:9 # 9
Joffard et al. 10.3389/fevo.2022.1058550
FIGURE 5
Results from the iBPP analysis based on molecular (Mol.), morphometric (Mor.), and chemical (Che.) data analyzed alone or in combination.
These results were robust to the algorithm that was used hybridization (Soliva et al., 2001; Soliva and Widmer, 2003),
to collapse internal nodes, and repetitions of each analysis that are both likely given the recent diversification of the
gave similar results. By contrast, these results varied depending section Pseudophrys (Breitkopf et al., 2014; Baguette et al., 2020)
on the topology of the guide tree. More specifically, when and the weakness of post-zygotic barriers between sympatric
O. bilunulata and O. marmorata on the one hand, and O. funerea species (Cortis et al., 2009). Although the markers used in
and O. zonata on the other hand were not considered as sister- this study were selected because of their high resolution at the
species in the guide tree, the best model was the twelve-species scale of the genus, they may not be informative enough to
model, because these two taxa were distinct from O. delforgei discriminate between such closely related taxa. Just like other
and O. sulcata, respectively. non-model organisms, the Ophrys genus will likely benefit from
the democratization of high-throughput sequencing technics
allowing to develop more resolutive markers (e.g., Bateman
Discussion et al., 2018).
By contrast, phenotypic differentiation between the
Our study aimed at comparing species boundaries drawn twelve studied taxa was often significant, perhaps because
from molecular, morphometric, and chemical data alone or in morphometric characters and floral scents are selected by
combination in a group of twelve Pseudophrys taxa. Our results pollinators and may thus evolve faster and be less affected by
showed that including phenotypic data in the analysis helped hybridization than neutral markers (Sedeek et al., 2014). Indeed,
being more accurate when delimiting species in this group. the size and shape of the labellum are likely to be selected to
Based on this integrative taxonomic approach, eight formally match those of the pollinator’s body (Triponez et al., 2013),
described species were recognized as such, while the best model and floral scents to match sex pheromones of female insects
suggested merging two pairs of taxa into one species each. (Schiestl et al., 1999; Ayasse et al., 2003). Interestingly, our
results show that morphometric characters are as informative
as floral scents to discriminate between Pseudophrys species.
Integration of genetic and phenotypic Both are classically used as criteria to delimit Ophrys species
data in the section Pseudophrys (Bernardos et al., 2005; Mant et al., 2005), but in the past
decades, much more emphasis has been put on chemical signals
Our results showed that genetic differentiation between (Schiestl et al., 1999; Ayasse et al., 2003; Stökl et al., 2005;
the twelve studied taxa was often limited and that species Véla et al., 2007). However, our results suggest that using
delimitations drawn from genetic data only (using either morphometric characters for taxonomic purposes is relevant
ABGD or iBPP) were thus very conservative, with only a few in the section Pseudophrys and emphasize the potential role of
taxa recognized as genuine species. Such a limited genetic orchid enthusiasts in providing valuable data for taxonomic
differentiation could result from incomplete lineage sorting or research (Véla et al., 2015). As in the case of molecular markers,
Frontiers in Ecology and Evolution 09 frontiersin.org
54
fevo-10-1058550 November 21, 2022 Time: 10:9 # 10
Joffard et al. 10.3389/fevo.2022.1058550
more informative morphometric or chemical markers could questioned when confronted to molecular, morphometric or
be developed using more sophisticated techniques, such chemical evidence (e.g., O. arnoldii: Bernardos et al., 2005;
as geometric morphometrics (Rakosy et al., 2017; Gibert O. vallesiana: Gögler et al., 2016). Our analysis supports
et al., 2022). More importantly, it would be interesting to most Pseudophrys species that are described in France, with
distinguish between selected (i.e., functionally significant) and two remarkable exceptions: on the one hand, the first and
neutral phenotypic traits through electrophysiological and/or second best models suggested to merge O. marmorata with
behavioral studies (Schiestl et al., 1999; Rakosy et al., 2017). The the previously described species O. bilunulata, and on the
distinction between biologically active and non-active floral other hand, the first and third best models suggested to merge
scents, in particular, would likely provide further insights into O. zonata with the previously described species O. funerea. The
the taxonomy of the section Pseudophrys (Stökl et al., 2005, similarity within these two pairs of taxa has been emphasized
2009), as shown in the section Euophrys (e.g., Mant et al., 2005). before (Bournérias and Prat, 2005; Tison and de Foucault,
Because of these heterogeneous levels of resolution between 2014), O. marmorata being sometimes called “O. bilunulata
molecular, morphometric and chemical data, species boundaries from Corsica” (Bournérias and Prat, 2005). Tison and de
drawn from genetic versus phenotypic data were not congruent. Foucault (2014) also suggested merging O. delforgei with the
Such an incongruence mirrors disagreements between authors O. bilunulata/O. marmorata pair, and O. sulcata with the
favoring phylogenetic distinctness versus reproductive isolation O. funerea/O. zonata pair, but our analysis does not support
through attraction of distinct pollinator species as a criterion these proposals, since we found that both O. delforgei and
to delimit Ophrys species (Paulus, 2006; Devey et al., 2008; O. sulcata were morphologically and chemically distinct from
Bateman et al., 2010). Methods based on genetic data are often their closest relatives. The continental O. bilunulata and the
judged more reliable than methods based on phenotypic data Corsican O. marmorata were found to be genetically and
because they are not subject to investigator bias, nor affected phenotypically similar, whereas the Cyrno-Sardinian O. funerea
by environmental or maternal conditions (Fujita et al., 2012). and O. zonata were found to be slightly distinct morphologically
However, speciation sometimes leaves no signature at the level but similar both genetically and chemically. Interestingly, recent
of neutral markers, especially when it is recent and when only records suggest that both O. bilunulata and O. marmorata
a few loci mediate reproductive isolation, as it is assumed to are pollinated by Andrena flavipes (Schatz et al., 2021), which
be the case in the genus Ophrys (Xu and Schlüter, 2015). In supports our proposal to merge these two taxa. Likewise,
this case, neutral markers-based methods may fail at detecting both O. funerea and O. zonata are pollinated by this species
species boundaries by putting aside the data that are the most (Foelsche et al., 2000; Schatz et al., 2021), suggesting that
informative. Our results also show that integrating several these two taxa are not reproductively isolated and should be
phenotypic traits (in our case, morphometric and chemical) in considered as conspecifics. The proposal to merge O. marmorata
the analysis may be helpful. For example, in our study, two with O. bilunulata does not imply that this taxon should not
taxa were slightly distinct morphologically, but strictly similar be considered as vulnerable in Corsica anymore; however, it
from a chemical point of view. When only morphometric implies that it should not be considered as threatened at the
characters were analysed, these two taxa were recognized as national level. Another important conclusion of our study is
species, whereas when both morphometric characters and floral the fact that O. peraiolae – which is sometimes described as
scents were analysed, they were merged. This shows that a morph of O. marmorata (Delforge, 2005) and sometimes
integrating new data types – either new molecular markers merged with O. funerea and O. zonata (Delage and Hugot,
or new phenotypic traits – may challenge previous taxonomic 2015) – likely corresponds to a genuine species, although it
inferences, species boundaries being hypotheses which should may be of hybrid origin (Tison and de Foucault, 2014). More
be tested using many data types to increase their robustness generally, our analysis supports many species that are not
(Padial et al., 2010). recognized in the latest version of the European Red List (e.g.,
O. lupercalis, O. bilunulata, etc.), in which they are all referred
to as “O. fusca sensu lato” (Rankou, 2011). We hope that this
Taxonomy and conservation of the study will prompt the reassessment of their UICN status and the
section Pseudophrys implementation of appropriate conservation actions, especially
for species with extremely restricted distribution areas and
The section Pseudophrys is known to be taxonomically declining population sizes such as O. peraiolae. We encourage
challenging, due to the lack of resolution of classic molecular the use of the integrative taxonomic approach proposed in this
markers in this section (Schlüter et al., 2007; Devey et al., study to other orchid groups in which species boundaries are
2008) and to the striking morphological similarity between its blurred, as it provides a framework to interpret patterns of
members (Bernardos et al., 2005). Ninety-seven Pseudophrys genetic and phenotypic divergence among taxa and would speed
species are described across the Mediterranean region (Delforge, up taxonomic revisions that are urgently needed for defining
2016), but the taxonomic rank of many of them has been conservation priorities.
Frontiers in Ecology and Evolution 10 frontiersin.org
55
fevo-10-1058550 November 21, 2022 Time: 10:9 # 11
Joffard et al. 10.3389/fevo.2022.1058550
Data availability statement Méditerranéen de l’Environnement et de la Biodiversité for
technical support; M. Busi, P. Cortis, A. Delage, P. Escudié, P.
The data presented in the study are deposited in the NCBI Fouquet, P. Geniez, T. Guillausson, L. Hugot, and J. Oltra for
GenBank repository, and accession numbers are provided in the their help with fieldwork and C. Ané, M. Hervé, and C. Solís-
Supplementary material. Lemus for statistical advice.
Author contributions Conflict of interest
NJ, EV, CM, and BS conceived and designed the analysis. NJ, The authors declare that the research was conducted in the
EV, and BS collected the data. NJ, BB, VA, and CM performed absence of any commercial or financial relationships that could
the analysis. All authors discussed the results and contributed to be construed as a potential conflict of interest.
the final manuscript.
Publisher’s note
Funding
All claims expressed in this article are solely those of the
This study was funded by heSam Université and its Paris authors and do not necessarily represent those of their affiliated
Nouveaux Mondes program, the Conservatoire Botanique organizations, or those of the publisher, the editors and the
National de Corse (Contract N◦15/006, OEC-CBNC), and the reviewers. Any product that may be evaluated in this article, or
Observatoire de REcherche Méditerranéen de l’Environnement claim that may be made by its manufacturer, is not guaranteed
(SO Ocove, OSU OREME). or endorsed by the publisher.
Acknowledgments Supplementary material
We thank the Service des Marqueurs Génétiques en Ecologie The Supplementary Material for this article can be
(SMGE - CEFE) and the Plateforme d’Analyses Chimiques en found online at: https://www.frontiersin.org/articles/10.3389/
Ecologie (PACE - CEFE) of the LABoratoire d’EXcellence Centre fevo.2022.1058550/full#supplementary-material
References
Adams, R. (2007). Identification of essential oil components by gas Bateman, R., Devey, D., Malmgren, S., Bradshaw, E., and Rudall, P. (2010).
chromatography/mass spectrometry, 4th Edn. Carol Stream, IL: Allured Publishing. Conflicting species concepts underlie perennial taxonomic controversies in
Agapow, P., Bininda-Emonds, O., Crandall, K., Gittleman, J., Mace, G., Ophrys. Cah. Société Française Orchidophilie 7, 87–104.
Marshall, J., et al. (2004). The impact of species concept on biodiversity studies. Bateman, R., Murphy, A., Hollingsworth, P., Hart, M., Denholm, I., and Rudall,
Q. Rev. Biol. 79, 161–179. doi: 10.1086/383542 P. (2018). Molecular and morphological phylogenetics of the digitate-tubered
Ayasse, M., Schiestl, F., Paulus, H., Ibarra, F., and Francke, W. (2003). clade within subtribe Orchidinae s.s. (Orchidaceae: Orchideae). Kew Bull. 73, 1–30.
Pollinator attraction in a sexually deceptive orchid by means of unconventional doi: 10.1007/s12225-018-9782-1
chemicals. Proc. R. Soc. Lond. B 270, 517–522. doi: 10.1098/rspb.2002. Bernardos, S., Crespí, A., Del Rey, F., and Amich, F. (2005). The section
2271 pseudophrys (Ophrys, Orchidaceae) in the Iberian Peninsula: A morphometric
Ayasse, M., Stökl, J., and Francke, W. (2011). Chemical ecology and pollinator- and molecular analysis. Bot. J. Linn. Soc. 148, 359–375. doi: 10.1111/j.1095-8339.
driven speciation in sexually deceptive orchids. Phytochemistry 72, 1667–1677. 2005.00403.x
doi: 10.1016/j.phytochem.2011.03.023 Bertrand, J., Baguette, M., Joffard, N., and Schatz, B. (2021). “Challenges
Baguette, M., Bertrand, J., Stevens, V., and Schatz, B. (2020). Why inherent in the systematics and taxonomy of genera that have recently experienced
are there so many bee-orchid species? Adaptive radiation by intraspecific explosive radiation: The case of orchids of the genus Ophrys,” in Systematics and
competition for mnesic pollinators. Biol. Rev. 95, 1630–1663. doi: 10.1111/brv.1 the exploration of life, eds P. Grandcolas and M. C. Maurel (London: ISTE CNRS),
2633 doi: 10.1002/9781119476870.c
Barrett, C., and Freudenstein, J. (2011). An integrative approach Bournérias, M., and Prat, D. (2005). Les orchidées de France, Belgique et
to delimiting species in a rare but widespread mycoheterotrophic Luxembourg, deuxième édition. Biotope editions, ed. Société Française and
orchid. Mol. Ecol. 20, 2771–2786. doi: 10.1111/j.1365-294X.2011. d’Orchidophilie (Mèze: Biotope Editions), 504.
05124.x Breitkopf, H., Onstein, R., Cafasso, D., Schlüter, P. M., and Cozzolino, S. (2014).
Bateman, R., Bradshaw, E., Devey, D., Glover, B., Malmgren, S., Sramkó, G., et al. Multiple shifts to different pollinators fuelled rapid diversification in sexually
(2011). Species arguments: Clarifying competing concepts of species delimitation deceptive Ophrys orchids. N. Phytol. 207, 377–389. doi: 10.1111/nph.13219
in the pseudo-copulatory orchid genus Ophrys. Bot. J. Linn. Soc. 165, 336–347. Cortis, P., Vereecken, N., Schiestl, F., Barone Lumaga, M., Scrugli, A., and
doi: 10.1111/j.1095-8339.2011.01121.x Cozzolino, S. (2009). Pollinator convergence and the nature of species’ boundaries
Frontiers in Ecology and Evolution 11 frontiersin.org
56
fevo-10-1058550 November 21, 2022 Time: 10:9 # 12
Joffard et al. 10.3389/fevo.2022.1058550
in sympatric Sardinian Ophrys (Orchidaceae). Ann. Bot. 104, 497–506. doi: 10. Kühn, R., Perdersen, H., and Cribb, P. (2020). Orchidées d’europe et de
1093/aob/mcn219 méditerranée. Mèze: Biotope editions, 430.
Cuypers, V., Reydon, T., and Artois, T. (2022). Deceiving insects, deceiving Lanfear, R., Calcott, B., Ho, S., and Guindon, S. (2012). PartitionFinder:
taxonomists? Making theoretical sense of taxonomic disagreement in the Combined selection of partitioning schemes and substitution models for
European orchid genus Ophrys. Perspect. Plant Ecol. Evol. Syst 56:125686. doi: phylogenetic analyses. Mol. Biol. Evol. 29, 1695–1701. doi: 10.1093/molbev/mss020
10.1016/j.ppees.2022.125686 Mant, J., Peakall, R., and Schiestl, F. P. (2005). Does selection on floral odor
Dayrat, B. (2005). Towards integrative taxonomy. Biol. J. Linn. Soc. 85, 407–415. promote differentiation among populations and species of the sexually deceptive
doi: 10.1111/j.1095-8312.2005.00503.x orchid genus Ophrys? Evolution 59, 1449–1463. doi: 10.1111/j.0014-3820.2005.
De Queiroz, K. (2007). Species concepts and species delimitation. Syst. Biol. 56, tb01795.x
879–886. doi: 10.1080/10635150701701083 Mason, N., Fletcher, N., Gill, B., Funk, W., and Zamudio, K. (2020). Coalescent-
Delage, A., and Hugot, L. (2015). Liste rouge régionale de la flore vasculaire based species delimitation is sensitive to geographic sampling and isolation by
de Corse. Corte: Conservatoire Botanique National de Corse, Office de distance. Syst. Biodivers. 18, 269–280. doi: 10.1080/14772000.2020.1730475
l’Environnement de la Corse, 72. Naciri, Y., and Linder, H. P. (2015). Species delimitation and relationships: The
Delforge, P. (2005). Guide des orchidées d’Europe, d’afrique du nord et du dance of the seven veils. Taxon 64, 3–16. doi: 10.12705/641.24
proche-orient. Lonay: Delachaux et Niestlé, 640. Núñez, R., Willmott, K., Álvarez, Y., Genaro, J., Pérez-Asso, A., Quejereta, M.,
Delforge, P. (2016). Orchidées d’Europe, d’Afrique du Nord et du Proche-Orient. et al. (2022). Integrative taxonomy clarifies species limits in the hitherto monotypic
Delachaux et Niestlé, Neuchâtel-Paris, Switzerland, p. 544. passion-vine butterfly genera Agraulis and Dryas (Lepidoptera, Nymphalidae,Heliconiinae). Syst. Entomol. 47, 152–178. doi: 10.1111/syen.12523
Devey, D., Bateman, R., Fay, M., and Hawkins, J. (2008). Friends or relatives?
Phylogenetics and species delimitation in the controversial European orchid genus Olave, M., Avila, L., Sites, J., and Morando, M. (2017). Hidden diversity within
Ophrys. Ann. Bot. 101, 385–402. doi: 10.1093/aob/mcm299 the lizard genus Liolaemus: Genetic versus morphological divergence in the L.rothi complex (Squamata: Liolaeminae). Mol. Phylogenet. Evol. 107, 56–63. doi:
Devillers, P., and Devillers-Terschuren, J. (1994). Essai d’analyse systématique 10.1016/j.ympev.2016.09.009
du genre Ophrys. Les Naturalistes Belg. 75, 273–400. Padial, J., Miralles, A., De la Riva, I., and Vences, M. (2010). The integrative
Edgar, R. (2004). MUSCLE: Multiple sequence alignment with high accuracy future of taxonomy. Front. Zool 7:16. doi: 10.1186/1742-9994-7-16
and high throughput. Nucleic Acids Res. 32, 1792–1797. doi: 10.1093/nar/gkh340 Paulus, H. (2006). Deceived males–pollination biology of the mediterranean
Edwards, D., and Knowles, L. (2014). Species detection and individual orchid genus Ophrys (Orchidaceae). J. Eur. Orchideen 38, 303–353.
assignment in species delimitation: Can integrative data increase efficacy? Proc.
R. Soc. Lond. B 281, 20132765–20132765. doi: 10.1098/rspb.2013.2765 Pedersen, H., and Faurholdt, N. (2007). Ophrys: The bee orchids of Europe.London: Royal Botanic Gardens.
Fay, M. (2018). Orchid conservation: How can we meet the challenges in the
twenty-first century? Bot. Stud. 59, 1–6. doi: 10.1186/s40529-018-0232-z Pessoa, E., Alves, M., Alves-Araújo, A., Palma-Silva, C., and Pinheiro, F. (2012).Integrating different tools to disentangle species complexes: A case study in
Foelsche, G., Foelsche, W., Gerbaud, M., and Gerbaud, O. (2000). Ophrys Epidendrum (Orchidaceae). Taxon 61, 721–734. doi: 10.1002/tax.614002
peraiolae spec. nov. und die taxa der Ophrys fusca gruppe in Korsika (Ophrys
peraiolae spec. nov. and the taxa of the Ophrys fusca group in Corsica.). J. Eur. Pillon, Y., and Chase, M. (2007). Taxonomic exaggeration and its effects on
Orchideen 32, 403–455. orchid conservation: Research note. Conserv. Biol. 21, 263–265. doi: 10.1111/j.1523-1739.2006.00573.x
Fujita, M., Leaché, A., Burbrink, F., McGuire, J., and Moritz, C. (2012).
Coalescent-based species delimitation in an integrative taxonomy. Trends Ecol. Pires, A., and Marinoni, L. (2010). DNA barcoding and traditional taxonomy
Evol. 27, 480–488. doi: 10.1016/j.tree.2012.04.012 unified through integrative taxonomy: A view that challenges the debatequestioning both methodologies. Biota Neotrop. 10, 339–346. doi: 10.1590/S1676-
Gibert, A., Louty, F., Buscail, R., Baguette, M., Schatz, B., and Bertrand, J. (2022). 06032010000200035
Extracting quantitative information from images taken in the wild: A case study
of two vicariants of the Ophrys aveyronensis species complex. Diversity 14:400. Plants of the World Online [POWO] (2022). Facilitated by the royal botanic
doi: 10.3390/d14050400 gardens. Available online at: http://www.plantsoftheworldonline.org/ (accessedSeptember 22, 2022).
Gögler, J., Zitari, A., Paulus, H., Cozzolino, S., and Ayasse, M. (2016). Species Puillandre, N., Lambert, A., Brouillet, S., and Achaz, G. (2012). ABGD,
boundaries in the Ophrys iricolor group in Tunisia: Do local endemics always Automatic barcode gap discovery for primary species delimitation. Mol. Ecol. 21,
matter? Plant Syst. Evol. 302, 481–489. doi: 10.1007/s00606-015-1271-5 1864–1877. doi: 10.1111/j.1365-294X.2011.05239.x
Gouy, M., Guindon, S., and Gascuel, O. (2010). SeaView version 4: A Pyron, R., Hsieh, F., Lemmon, A., Lemmon, E., and Hendry, C. (2016).
multiplatform graphical user interface for sequence alignment and phylogenetic Integrating phylogenomic and morphological data to assess candidate species-
tree building. Mol. Biol. Evol. 27, 221–224. doi: 10.1093/molbev/msp259 delimitation models in brown and red-bellied snakes (Storeria). Zool. J. Linn. Soc.
Guillot, G., Renaud, S., Ledevin, R., Michaux, J., and Claude, J. (2012). A 177, 937–949. doi: 10.1111/zoj.12392
unifying model for the analysis of phenotypic, genetic, and geographic data. Syst. R Development Core Team. (2008). R: A language and environment for statistical
Biol. 61, 897–911. doi: 10.1093/sysbio/sys038 computing. Vienna: R Foundation for Statistical Computing.
Hey, J. (2006). On the failure of modern species concepts. Trends Ecol. Evol. 21, Rakosy, D., Cuervo, M., Paulus, H., and Ayasse, M. (2017). Looks matter:
447–450. doi: 10.1016/j.tree.2006.05.011 Changes in flower form affect pollination effectiveness in a sexually-deceptive
Huang, J., and Knowles, L. (2016). The species versus subspecies conundrum: orchid. J. Evol. Biol. 30, 1978–1993. doi: 10.1111/jeb.13153
Quantitative delimitation from integrating multiple data types within a single Rankou, H. (2011). Ophrys fusca. The IUCN red list of threatened species 2011:
Bayesian approach in hercules beetles. Syst. Biol. 65, 685–699. doi: 10.1093/sysbio/ e.T175999A7166962. Gland: IUCN.
syv119 Rannala, B., and Yang, Z. (2003). Bayes estimation of species divergence times
IUCN, MNHN, FCBN, and SFO. (2010). “Chapitre orchidées de france and ancestral population sizes using DNA sequences from multiple loci. Genetics
métropolitaine,” in La liste rouge des espèces menacées de France (Paris: IUCN). 164, 1645–1656. doi: 10.1093/genetics/164.4.1645
Joffard, N., Arnal, V., Buatois, B., Montgelard, C., and Schatz, B. (2020). Ronquist, F., and Huelsenbeck, J. (2003). MrBayes 3: Bayesian phylogenetic
Floral scent evolution in the section Pseudophrys: Phylogenetic constraints inference under mixed models. Bioinformatics 19, 1572–1574. doi: 10.1093/
or pollinator-mediated selection? Plant Biol. 22, 881–889. doi: 10.1111/plb.1 bioinformatics/btg180
3104 Schatz, B., Gauthier, P., Debussche, M., and Thompson, J. (2014). A decision
Joffard, N., Buatois, B., and Schatz, B. (2016). Integrative taxonomy of the fly tool for listing species for protection on different geographic scales and
orchid group: Insights from chemical ecology. Sci. Nat. 103, 1–5. doi: 10.1007/ administrative levels. J. Nat. Conserv. 22, 75–83. doi: 10.1016/j.jnc.2013.09.003
s00114-016-1403-y Schatz, B., Genoud, D., Claessens, J., and Kleynen, J. (2020). Orchid-pollinator
Joffard, N., Massol, F., Grenié, M., Montgelard, C., and Schatz, B. (2019). network in Euro-Mediterranean region: What we know, what we think we know,
Effect of pollination strategy, phylogeny and distribution on pollination niches of and what remains to be done. Acta Oecol. 107:103605. doi: 10.1016/j.actao.2020.
Euro-Mediterranean orchids. J. Ecol. 107, 478–490. doi: 10.1111/1365-2745.13013 103605
Knowles, L., and Carstens, B. (2007). Delimiting species without monophyletic Schatz, B., Genoud, D., Escudié, P., Geniez, P., Wünsch, K., and Joffard,
gene trees. Syst. Biol. 56, 887–895. doi: 10.1080/10635150701701091 N. (2021). Is Ophrys pollination more opportunistic than previously thought?
Frontiers in Ecology and Evolution 12 frontiersin.org
57
fevo-10-1058550 November 21, 2022 Time: 10:9 # 13
Joffard et al. 10.3389/fevo.2022.1058550
Insights from different field methods of pollinator observation. Bot. Lett. 168, Triponez, Y., Arrigo, N., Pellissier, L., Schatz, B., and Alvarez, N. (2013).
333–347. doi: 10.1080/23818107.2021.1959394 Morphological, ecological and genetic aspects associated with endemism in the
Schiestl, F., and Ayasse, M. (2002). Do changes in floral odor cause speciation fly orchid group. Mol. Ecol. 22, 1431–1446. doi: 10.1111/mec.12169
in sexually deceptive orchids? Plant Syst. Evol. 234, 111–119. doi: 10.1007/s00606- Véla, E., Rebbas, K., Martin, R., de Premorel, G., and Tison, J.-M. (2015).
002-0187-z Waiting for integrative taxonomy: Morphospecies as an operational proxy for the
Schiestl, F., Ayasse, M., Paulus, H., Löfstedt, C., Hansson, B., Ibarra, F., et al. radiative and reticulate genus Ophrys L. (Orchidaceae)? Eur. J. Environ. Sci. 5,
(1999). Orchid pollination by sexual swindle. Nature 399, 421–421. doi: 10.1038/ 153–157. doi: 10.14712/23361964.2015.89
20829 Véla, E., Tirard, A., Renucci, M., Suehs, C., and Provost, E. (2007). Floral
Schlick-Steiner, B., Steiner, F., Seifert, B., Stauffer, C., Christian, E., and chemical signatures in the genus Ophrys L. (Orchidaceae): A preliminary test of
Crozier, R. (2010). Integrative taxonomy: A multisource approach to exploring a new tool for taxonomy and evolution. Plant Mol. Biol. Rep. 25, 83–97. doi:
biodiversity. Annu. Rev. Entomol. 55, 421–438. doi: 10.1146/annurev-ento- 10.1007/s11105-007-0009-0
112408-085432 Vereecken, N., Dafni, A., and Cozzolino, S. (2010). Pollination syndromes in
Schlüter, P., Kohl, G., Stuessy, T., and Paulus, H. (2007). A screen of low- mediterranean orchids–implications for speciation, taxonomy and conservation.
copy nuclear genes reveals the LFY gene as phylogenetically informative in Bot. Rev. 76, 220–240. doi: 10.1007/s12229-010-9049-5
closely related species of orchids (Ophrys). Taxon 56, 493–504. doi: 10.1002/tax.56 Vereecken, N., Streinzer, M., Ayasse, M., Spaethe, J., Paulus, H., Stökl, J., et al.
2019 (2011). Integrating past and present studies on Ophrys pollination–a comment
Sedeek, K., Scopece, G., Staedler, Y., Schönenberger, J., Cozzolino, S., Schiestl, on Bradshaw et al. Bot. J. Linn. Soc. 165, 329–335. doi: 10.1111/j.1095-8339.2011.
F., et al. (2014). Genic rather than genome-wide differences between sexually 01112.x
deceptive Ophrys orchids with different pollinators. Mol. Ecol. 23, 6192–6205. Will, K., Mishler, B., and Wheeler, Q. (2005). The perils of DNA barcoding
doi: 10.1111/mec.12992 and the need for integrative taxonomy. Syst. Biol. 54, 844–851. doi: 10.1080/
Solís-Lemus, C., Knowles, L., and Ané, C. (2015). Bayesian species delimitation 10635150500354878
combining multiple genes and traits in a unified framework. Evolution 69, 492– Xu, S., and Schlüter, P. (2015). Modelling the two-locus architecture of
507. doi: 10.1111/evo.12582 divergent pollinator adaptation: How variation in SAD paralogs affects fitness
Soliva, M., and Widmer, A. (2003). Gene flow across species boundaries in and evolutionary divergence in sexually deceptive orchids. Ecol. Evol. 5, 493–502.
sympatric, sexually deceptive Ophrys (Orchidaceae) species. Evolution 57, 2252– doi: 10.1002/ece3.1378
2261. doi: 10.1554/02-442 Yang, L., Kong, H., Huang, J., and Kang, M. (2019). Different species
Soliva, M., Kocyan, A., and Widmer, A. (2001). Molecular phylogenetics of or genetically divergent populations? Integrative species delimitation of the
the sexually deceptive orchid genus Ophrys (Orchidaceae) based on nuclear and Primulina hochiensis complex from isolated karst habitats. Mol. Phylogenet. Evol.
chloroplast DNA sequences. Mol. Phylogenet. Evol. 20, 78–88. doi: 10.1006/mpev. 132, 219–231. doi: 10.1016/j.ympev.2018.12.011
2001.0953 Yang, Z. (2015). The BPP program for species tree estimation and species
Stökl, J., Paulus, H., Dafni, A., Schulz, C., Francke, W., and Ayasse, M. (2005). delimitation. Curr. Zool. 61, 854–865. doi: 10.1093/czoolo/61.5.854
Pollinator attracting odour signals in sexually deceptive orchids of the Ophrys fusca Yang, Z., and Rannala, B. (2010). Bayesian species delimitation using multilocus
group. Plant Syst. Evol. 254, 105–120. doi: 10.1007/s00606-005-0330-8 sequence data. Proc. Natl. Acad. Sci. U.S.A. 107, 9264–9269. doi: 10.1073/pnas.
Stökl, J., Schlüter, P., Stuessy, T., Paulus, H., Fraberger, R., Erdmann, D., 0913022107
et al. (2009). Speciation in sexually deceptive orchids: Pollinator-driven selection Yeates, D., Seago, A., Nelson, L., Cameron, S., Joseph, L., and Trueman, J.
maintains discrete odour phenotypes in hybridizing species. Biol. J. Linn. Soc. 98, (2011). Integrative taxonomy, or iterative taxonomy? Syst. Entomol. 36, 209–217.
439–451. doi: 10.1111/j.1095-8312.2009.01279.x doi: 10.1111/j.1365-3113.2010.00558.x
Sukumaran, J., and Knowles, L. (2017). Multispecies coalescent delimits Zhang, C., Low, S., Song, Y., Kozlowski, G., Li, L., Zhou, S., et al. (2020).
structure, not species. PNAS 114, 1607–1612. doi: 0.1073/pnas.1607921114 Shining a light on species delimitation in the tree genus Engelhardia Leschenault
Tison, J.-M., and de Foucault, B. (2014). Flora gallica: Flore de France. Mèze: ex Blume (Juglandaceae). Mol. Phylogenet. Evol. 152:106918. doi: 10.1016/j.ympev.
Biotope editions, 1200. 2020.106918
Frontiers in Ecology and Evolution 13 frontiersin.org
58
fevo-10-1047267 November 23, 2022 Time: 16:39 # 1
TYPE Original Research
PUBLISHED 29 November 2022
DOI 10.3389/fevo.2022.1047267
Shade and drought increase
OPEN ACCESS fungal contribution to partially
EDITED BY
Isabel Marques, mycoheterotrophic terrestrial
University of Lisbon, Portugal
REVIEWED BY orchids Goodyera pubescens
César Marín,
Universidad Santo Tomás, Chile
Gerhard Gebauer, and Tipularia discolor
University of Bayreuth, Germany
*CORRESPONDENCE
Melissa K. McCormick Melissa K. McCormick1*, Kerry L. Good1,
mccormickm@si.edu
Thomas J. Mozdzer1,2 and Dennis F. Whigham1
SPECIALTY SECTION
This article was submitted to 1Smithsonian Environmental Research Center, Smithsonian Institution, Edgewater, MD,
Conservation and Restoration Ecology, United States, 2Department of Biology, Bryn Mawr College, Bryn Mawr, PA, United States
a section of the journal
Frontiers in Ecology and Evolution
RECEIVED 17 September 2022
ACCEPTED 10 November 2022 Many photosynthetic plants supplement photosynthetic carbon with fungal
PUBLISHED 29 November 2022
carbon, but the mechanisms that govern dependence on mycoheterotrophic
CITATION
McCormick MK, Good KL, Mozdzer TJ carbon are poorly understood. We used exclusion shelters to manipulate
and Whigham DF (2022) Shade and water and light availability to plants of the terrestrial orchids Goodyera
drought increase fungal contribution
to partially mycoheterotrophic pubescens and Tipularia discolor. We tracked changes in δ
13C from
terrestrial orchids Goodyera photosynthesis and δ15N acquired from soil-derived inorganic nitrogen
pubescens and Tipularia discolor.
Front. Ecol. Evol. 10:1047267. versus mycoheterotrophy, along with direct measures of photosynthesis
doi: 10.3389/fevo.2022.1047267 in T. discolor. We hypothesized that shade would increase dependence
COPYRIGHT on mycoheterotrophy compared to reference plants, while drought would
© 2022 McCormick, Good, Mozdzer decrease both photosynthesis and the abundance of potential mycorrhizal
and Whigham. This is an open-access
article distributed under the terms of fungi. Drought and shade enriched
13C and 15N in both G. pubescens
the Creative Commons Attribution and T. discolor, compared to control plants, indicating increased fungal
License (CC BY). The use, distribution
or reproduction in other forums is contribution to orchid tissues. Physiological measurements of T. discolor
permitted, provided the original leaves showed that dark respiration, water use efficiency, and relative
author(s) and the copyright owner(s)
are credited and that the original electron transport rate did not vary significantly, but shaded plants had
publication in this journal is cited, in greater quantum efficiency, suggesting they were light-limited. Light saturated
accordance with accepted academic photosynthesis of T. discolor leaves was lower in both shaded and drought-
practice. No use, distribution or
reproduction is permitted which does treated plants, indicating lower photosynthetic capacity, and likely greater
not comply with these terms. dependence on mycoheterotrophy and corresponding enrichment in 13C
and 15N. This study documented changes in orchid dependence on fungal
carbon in response to manipulated environmental conditions. Both shade and
drought increased the dependence of both orchids on mycoheterotrophically
derived carbon and nitrogen.
KEYWORDS
Goodyera pubescens, Tipularia discolor, mycoheterotrophy, orchid, mycorrhizae,
stable isotope
Frontiers in Ecology and Evolution 01 frontiersin.org
59
fevo-10-1047267 November 23, 2022 Time: 16:39 # 2
McCormick et al. 10.3389/fevo.2022.1047267
Introduction Hynson et al., 2009; Preiss et al., 2010). Indeed, Preiss et al.
(2010) and Schweiger et al. (2019) demonstrated a correlation
Between 85 and 92% of land plants obtain nutrients and between the light quantity and the proportion of carbon that
water from the soil through mycorrhizal associations (Wang orchids derived from fungi.
and Qiu, 2006; Brundrett and Tedersoo, 2018). For most plants, Almost all herbaceous species in forests are highly or
this association is a two-way exchange, with the plant providing obligately dependent on mycorrhizal fungi (Brundrett and
carbon to the fungus in exchange for other resources. Some non- Kendrick, 1988; Whigham, 2004), and the loss of plant-
photosynthetic plants, termed fully mycoheterotrophic, obtain fungal interactions has negative consequences for physiological
carbon from their mycorrhizal fungi (Gebauer and Meyer, 2003; processes, including nutrient and water uptake (Hale et al.,
Lallemand et al., 2019), while other mycoheterotrophic plants 2011). Plant-fungal interactions can be disrupted, and the
only initially rely on fungal carbon until they produce green quantity and direction of benefits altered. McCormick et al.
leaves (Leake and Cameron, 2010; Tĕšitel et al., 2018). However, (2006), for example, found that individuals of Goodyera
recently many green photosynthetic plants have been shown pubescens lost their mycorrhizal fungi during a drought.
to supplement photosynthetic carbon with fungal carbon, and Surviving plants subsequently associated with the same or
are termed partially mycoheterotrophic (Gebauer and Meyer, different mycorrhizal fungi, but also suffered higher mortality.
2003) or mixotrophic (e.g., Selosse and Roy, 2008; Hynson The presence of mycorrhizal fungi is also important in non-
et al., 2009; Merckx et al., 2010; Selosse and Martos, 2014). The orchids. Bitterlich et al. (2019) found that mycorrhizal fungi
mechanisms that govern the extent to which plants depend on supported increased photosynthesis in tomatoes, but only when
mycoheterotrophically derived carbon are poorly understood there was sufficient light and moisture. In contrast, Zhu et al.
but may be important for understanding the evolution of (2011) and Cabral et al. (2016) found that mycorrhizal plants
mycorrhizal associations, especially mycoheterotrophy (Selosse maintained higher photosynthetic rates and yield in response
and Roy, 2008; Leake and Cameron, 2010; Wang et al., 2021). to heat stress. These findings suggest that stress responses by a
All orchids are mycoheterotrophic at the protocorm life wide range of plants are affected by fungal interactions, but there
history stage and depend entirely on fungi for carbon and have been few instances where that hypothesis has been directly
other resources that are required for transition to later tested.
life history stages. During the later life history stages, the Stable isotope natural abundance analysis is a useful
amount of carbon they derive mycoheterotrophically varies approach to the study of mycoheterotrophic nutrient pathways
(e.g., Leake, 1994; Rasmussen and Rasmussen, 2007). Most (e.g., Gebauer and Meyer, 2003; Trudell et al., 2003; Ogura-
orchids photosynthesize at maturity and are not obligate Tsujita et al., 2009). Heterotrophically derived nutrients reflect
mycoheterotrophs. However, nearly all orchids continue to the isotopic composition of their source and, because fungal
associate with mycorrhizal fungi, and largely autotrophic C and sometimes nitrogen (N) are typically enriched in heavy
orchids can be partially mycoheterotrophic (e.g., Gebauer and isotopes relative to photosynthetically fixed carbon and soil-
Meyer, 2003; Liebel et al., 2010; Yagame et al., 2012; Selosse and derived inorganic nitrogen, C and N isotopes can be used
Martos, 2014; Hynson, 2016; Schiebold et al., 2018; Schweiger to estimate the degree of mycoheterotrophy (e.g., Gebauer
et al., 2018). Studies of albino and variegated variants of green and Meyer, 2003; Zimmer et al., 2007; Suetsugu et al., 2019).
orchids have also been used to demonstrate the importance of The carbon and nitrogen isotopic distinctiveness of nutrient
resource movement from mycorrhizal fungi to orchids (Selosse contributions from saprotrophic fungi is far less than for
et al., 2004; Lallemand et al., 2019; Suetsugu et al., 2019) and to fungi that simultaneously form ectomycorrhizal associations.
demonstrate a linear relationship between leaf chlorophyll and Hydrogen isotopes are now being used to overcome the limited
fungal contributions to plant carbon (Stöckel et al., 2011). Until power of C and N isotopes to quantify mycoheterotrophy
recently, orchids were solely assumed to be the beneficiaries and have demonstrated greater fungal contribution to plant
of a non-mutualistic association, obtaining carbon from fungi nutrition than previously suspected (Gebauer et al., 2016),
but not providing anything in return (Alexander and Hadley, but these methods are still not widely applied, and they
1985; Smith and Read, 1997). However, recent studies have were unavailable when the reported study was conducted. If
shown that species within the genus Goodyera provide carbon the amount of mycoheterotrophy changes with environmental
(C) to mycorrhizal fungi under specific laboratory conditions conditions or stress, then plant isotopic enrichment would
(Cameron et al., 2006, 2008; Hynson et al., 2009). While be expected to reflect that change. Such changes in the
the circumstances that dictate the direction of C flow are relative contribution of fungi to plant nutrition and isotopic
unclear, it has been speculated that stressors that reduce a composition could result from increasing fungal contribution,
plant’s photosynthetic ability (e.g., limited light and moisture) decreasing photosynthetic contribution, or both (Jacquemyn
may prevent autotrophic carbon acquisition to the extent that et al., 2021). Additionally, isotopic composition can shift
orchids will increase the level of resources gained through with direct effects of environmental conditions on stomatal
mycoheterotrophy (Gebauer, 2005; McCormick et al., 2006; conductance and photosynthesis.
Frontiers in Ecology and Evolution 02 frontiersin.org
60
fevo-10-1047267 November 23, 2022 Time: 16:39 # 3
McCormick et al. 10.3389/fevo.2022.1047267
We tested the hypothesis that photosynthetic orchids growth. After becoming photosynthetic, the species associates
increase reliance on mycorrhizal fungi for C and N acquisition with a wide range of fungi that belong to several distantly related
during periods of resource limitation. We used C and N Tulasnella clades (McCormick et al., 2004).
isotope analysis to determine whether two orchids with
different life history characteristics, G. pubescens and Tipularia
discolor, relied more on mycorrhizal fungi for N and C under Study location
conditions of light limitation and decreased water availability.
Light availability has been previously shown to influence The experiments were conducted in six deciduous
photosynthesis in Tipularia discolor (Tissue et al., 1995; Hughes forest stands, three for each species of orchid, at the
et al., 2019) and carbon acquisition in a species of Goodyera Smithsonian Environmental Research Center (SERC) in
(Liebel et al., 2015). Both direct effects of drought on plants Edgewater, Maryland, USA. For both orchids, in each site
and increased fungal contribution to plant carbon would be we located 12 mature plants. For G. pubescens, the plants
expected to increase plant δ13C, but only increased fungal had rosettes that were ≥4 cm diameter and ≥five leaves.
contribution would be expected to increase enrichment in Tipularia discolor plants, which produce a single leaf per
15N. We hypothesized that increased shade would decrease year, had leaves that were ≥3 cm wide. The orchids that
photosynthesis without directly affecting fungi, resulting in were selected in each forest stand were separated by 1–
orchid leaves enriched in 13C, reflecting increased fungal 2 meters to prevent sampling multiple plants associated with
contribution to plant carbon. In contrast, we hypothesized a single fungal organism. McCormick et al. (2006) found
that drought would affect photosynthesis through decreased that orchids separated by more than 50 cm associated with
stomatal conductance but would potentially decrease fungal different fungal individuals. We randomly assigned four
contribution to plant carbon. We expected the two species to individuals of each species to the shade and drought treatments
differ in mean isotopic enrichment, because they associated with (described below) and controls. We also selected 12 Fagus
different fungi and were active at different times of the year, grandifolia Ehrh. seedlings (10–20 cm tall) associated with
but we expected isotopic enrichment to increase or decrease each selected orchid to serve as autotrophic reference plants.
similarly in response to treatment conditions. Fagus grandifolia was the only autotrophic species present
across all study sites and <20 cm away from each selected
orchid.
Materials and methods
Study species Experimental set-ups
Goodyera pubescens R.Br is an evergreen orchid occurring Individual orchids subjected to drought or shade treatments
in mid and late successional forests throughout eastern were covered by exclosure shelters, each constructed of 1.9 cm
United States. Individual plants have a basal rosette of leaves diameter PVC pipe to form a 50 cm × 50 cm canopy with 28 cm
with new leaves produced in the spring. Flowering occurs in legs. Shade structures used black 95% shade cloth as per Gorchov
mid-summer and the inflorescence emerges from the center et al. (2011). Drought shelters were covered with UV-permeable
of the basal rosette. After flowering, rhizomes may branch, rain barrier plastic (2-mil ACLAR 22A, Honeywell Specialty
allowing limited asexual reproduction, but clones remain small Films, Linden, NJ, USA), and were bordered on their uphill
and did not extend beyond the experimental treatments. edge by a 50 cm length of landscape edging to divert surface
Pelotons of mycorrhizal fungi are present year-round in older runoff. Shelters for G. pubescens remained in place from June-
roots and colonize newly produced roots (Rasmussen and October 2009 (5 months) and shelters over T. discolor remained
Whigham, 2002). Plants associate exclusively with a single clade in place from October 2009–February 2011 (16 months). Leaves
of Tulasnella spp. (McCormick et al., 2004) that decompose of both species were collected for isotope analysis, described
organic matter as their primary form of nutrition, and can below, at the end of the study. In addition to the two orchids
switch fungi following drought (McCormick et al., 2006). and reference F. grandifolia seedlings, we also collected leaves
Tipularia discolor (Pursh) Nutt. is a winter-green orchid that for isotope analysis from two seedlings of F. grandifolia that were
produces a single leaf that appears in early autumn, typically growing beneath two of the T. discolor exclosures, one beneath
September-October, and senesces in the spring, typically May. a shade and the other beneath a drought treatment. While just
Flowering occurs at the end of July or beginning of August two F. grandifolia seedlings make for a very small sample size,
when leaves are not present. The species occurs primarily in we had hoped to have far more F. grandifolia individuals, as
hardwood forests throughout Eastern US. Fungal pelotons are well as other species, but few other plants grow in the shaded
present in roots throughout the year, with two fungi in the genus understory locations where the study took place and no other
Protomerulius that support seed germination and protocorm plants survived the treatments.
Frontiers in Ecology and Evolution 03 frontiersin.org
61
fevo-10-1047267 November 23, 2022 Time: 16:39 # 4
McCormick et al. 10.3389/fevo.2022.1047267
Environmental data with the instrument set to the following conditions: block
temperature: 25◦C, PAR: 1,000 micromols, as these represent
Soil moisture data associated with each shelter were optimum light and temperature conditions for this species
collected at the beginning and end of each experiment to (Tissue et al., 1995). After a minimum of 5 min under light
assess relative differences among sites, and to verify treatment saturated condition, we began to log data.
efficacy. Percent soil moisture between 0 and 12 cm depth We then changed PAR to 0 to estimate dark respiration (Rd)
was measured at eight locations in each site to account for and waited a minimum of 3 minutes until steady state conditions
site differences in water availability (HydrosenseTM Moisture were achieved prior to logging data. Water use efficiency
Meter, Campbell Scientific Australia Pty. Ltd., Garbutt, QLD, (WUE) under light saturated conditions was calculated as
Australia). Photosynthetically active radiation (PAR) was Asat/transpiration. We also performed rapid light curves using
measured using an AccuPAR PAR-80 1 m Sunfleck ceptometer pulse amplitude modulated fluorometry (Mini PAM, Walz,
(Decagon Devices, Pullman, WA, USA) at 50 cm above the Hamburg, Germany) to estimate quantum efficiency (α) and the
ground, a height that reflected light levels that plants would maximum relative electron transport rate (rETR). Briefly, leaves
have been experiencing prior to manipulation, at each plant were exposed to eight increasing levels of PAR for 10 seconds,
location, and at random locations within each forest stand. followed by a 0.6 saturation pulse of light. Both α and rETR were
PAR was highly variable (range 1–46% available PAR) among fit in non-linear models in SAS (v 9.2) (proc nlin) as described
locations, sites, and beneath the light shelters and was never by Ralph and Gademann (2005).
a significant factor in plant C, N, or isotopic enrichment (all Experimental treatment and site effects on Asat, Rd, α, and
P > 0.6), so it was not included in final analyses. Soil moisture rETR were analyzed using two-way ANOVAs in Systat v 12.0
values were compared among treatments (fixed independent with site and treatment as main effects and interactions. Where
variable) and sites (random independent variable) using an treatment effects were significant, we conducted a post-hoc
ANOVA. Experimental treatment and site effects were analyzed comparison among the treatment means using Tukey’s honestly
using two-way ANOVAs in Systat v 12.0 with site and treatment significant difference test in Systat (12.0).
as main effects and interactions. Where treatment effects were
significant, we conducted a post-hoc comparison among the
treatment means using Tukey’s honestly significant difference Stable isotope abundances and
test in Systat (12.0). nitrogen and carbon concentrations
Goodyera pubescens leaves were collected after 17 weeks
Plant growth and ecophysiology of treatment and analyzed for relative isotopic abundance.
We collected the youngest full-sized leaf from the center
For G. pubescens, initial plant size was measured by taking of each rosette. This ensured that we were collecting a leaf
a photograph of each plant with a ruler for scale. Photos were that had formed after the treatment was initiated, hence
printed, using the ruler to check for scale, and each leaf was minimizing dilution of treatment effects through averaging
cut out and area measured using a LI-3100 area meter (LiCor, over the lifespan of leaves that were already present when
Lincoln, NE, USA). At the end of the experiment, plants were treatments began (e.g., Hynson et al., 2012). At the same time,
again photographed, and area measured as before. Growth we harvested the youngest full-sized leaves from the nearby
of each plant was calculated as the change in area from the F. grandifolia seedlings. After each leaf harvest, scissors used
beginning of the experiment until the end. to cut each leaf were cleaned with 95% ethanol to prevent
For T. discolor, we measured the length and width of each cross-contamination. Leaves were placed directly into sterile
leaf (each plant produced a single leaf per year) using a ruler micro-centrifuge tubes, returned to the laboratory within 3 h,
and converted to area using: Leaf Area = 2/3 (length × width). and stored at 20◦− C until they were prepared for isotope mass
Growth of each leaf was based on area at the beginning and end spectrometry (below).
of the experiment. For this species, we took advantage of the Fully expanded leaves of T. discolor were collected in
availability of a LiCor instrument that was not available when February 2011, 16 months after the experiment began. This
we conducted the study with G. pubescens. We measured the ensured we were collecting leaves that initiated after treatment
effects of drought and shade on T. discolor light saturated rates onset. The youngest full-sized leaves of the nearby F. grandifolia
of photosynthesis (Asat) using a LiCor 6400 (LiCor Biosciences, near the T. discolor shelters were collected earlier, October 2010.
Lincoln, NE, USA) on a warm (air temperature was 19.5◦C) Fagus grandifolia leaves were collected at a different time because
winter day, February 18, 2011. Hughes et al. (2019) found by the time T. discolor leaves were fully developed, the reference
that February was when T. discolor had the highest rates of plants would not have had leaves. Leaves of the two F. grandifolia
photosynthesis, reflecting a combination of increased light in the seedlings growing beneath T. discolor shelters (described above)
forest understory and physiological activity. Asat was measured were also sampled in October 2010.
Frontiers in Ecology and Evolution 04 frontiersin.org
62
fevo-10-1047267 November 23, 2022 Time: 16:39 # 5
McCormick et al. 10.3389/fevo.2022.1047267
Frozen leaf samples from the orchids and F. grandifolia TABLE 1 Percent soil moisture (volumetric ± se) at each of the three
were lyophilized, ground, and weighed (3–4 mg of ground sites and three treatments (control, drought, and shade) for Goodyerapubescens and Tipularia discolor.
foliar tissue) into 5 × 9 mm tin capsules (Costech Analytical
Technologies, Valencia, CA, USA). Stable isotope ratio gas Site 1 Site 2 Site 3
chromatography mass spectrometry (EA-IRMS) analyses were
Goodyera pubescens
completed at the Smithsonian Museum Conservation Institute
Control 23.7 ± 5.4 21.3 ± 1.2 31.5 ± 1.8
Stable Isotope/Mass Spectrometry Lab in Suitland, MD,
Drought 12.2 ± 1.6 11.9 ± 1.1 18.0 ± 1.1
using an Elemental Analyzer Model 4010 (Costech Analytical
Shade 19.5 ± 1.5 22.3 ± 2.5 29.3 ± 1.8
Technologies) coupled to a Delta V Advantage Isotopic Ratio
Tipularia discolor
Mass Spectrometer with Isodat NT Software (Thermo Fisher
Control 20.6 ± 1.2 19.4 ± 0.5 17.0 ± 0.2
Scientific, Waltham, MA, USA).
Drought 11.6 ± 0.2 12.0 ± 0.2 11.2 ± 0.2
Measured isotope abundances are presented as δ-values and
Shade 20.3 ± 1.2 18.9 ± 0.1 16.4 ± 0.5
calculated using the equation: δ15N or δ13C = (Rsample/Rstandard
−1) × 1000 [h] (where Rsample and Rstandard are ratios of
heavy:light isotope of each element in the sample or standard)
(Gebauer and Meyer, 2003). Because stable isotope composition
is affected by local climatic conditions, relative isotope ratios
were normalized to site-specific enrichment factors for each
species using the equation: ε = δxS − δxR, where δxS is the
individual δ15N or δ13C value of a sample, and δxR is the mean
δ15N or δ13C of the 12 autotrophic reference plants at the site
in question (Preiss and Gebauer, 2008). Total %N and %C were
measured on all leaf samples with the same instrument used for
the isotope analysis.
Percent N and C data were analyzed for correlations with
isotopic composition, and to determine whether they varied
with sample date or experimental treatment. We compared the
isotopic enrichment of drought, shade, and control plants using
ANOVAs with ε15N and ε13C as dependent variables. Treatment
(control, shade, drought) and species were fixed independent
variables, and site (nested within species) was a random variable.
After considering the effect of treatment overall, we compared
drought and shade treatments in a second set of identical
ANOVAs. All calculations were conducted using Systat 12 for
Windows (Systat Software Inc., San Jose, CA, USA).
Results
FIGURE 1
Relative leaf area growth (mean ± 1 SE) for (A) Goodyera
Environmental data pubescens and (B) Tipularia discolor control and treatment plots
at each site. The three bars for the two treatments and controls
represent different sites where the experiment was conducted,
Soil moisture varied among sites and treatments for both as described in the section “Materials and methods.”
species (Table 1). The treatment effect was significant (F = 61.3,
P < 0.001). Soil moisture differed among the species, which was
expected, since the studies were carried out in different years
and seasons, and was significantly lower inside the precipitation Orchid growth
exclosures for both species, compared to shade or control
locations (ANOVA: Species: F = 37.5, P < 0.001; Treatment: Relative leaf growth was significantly different between
F = 61.3, P < 0.001; Species × Treatment: F = 2.05, P = 0.15, species (Figures 1A,B; ANOVA: Species: F = 13.5, P = 0.001)
Species (Site): F = 15.6, P < 0.001; Supplementary Figure 1). but the treatment and interaction effects were not significant
Shaded and control soils did not differ significantly (F = 0.700, (Treatment: F = 0.038, P = 0.96; Species × Treatment: F = 0.022,
P = 0.41); Species × Treatment: F = 2.05, P = 0.15, Species (Site): P = 0.98, Species (Site): F = 3.41, P = 0.014). Relative leaf growth
F = 15.6, P < 0.001. was positive in all of the controls and the majority of treatments
Frontiers in Ecology and Evolution 05 frontiersin.org
63
fevo-10-1047267 November 23, 2022 Time: 16:39 # 6
McCormick et al. 10.3389/fevo.2022.1047267
TABLE 2 Carbon (%C) and nitrogen (%N) concentrations (each Treatment: F = 1.57, P = 0.22), and the interactions between
mean ± se) of G. pubescens and T. discolor across all treatments (top
two rows) and concentrations across treatments (bottom three rows). Species × Treatment (F = 0.808, P = 0.45) and Species
(Site) (F = 0.890, P = 0.48) were not significant. Leaf N
%C %N concentrations were significantly higher for T. discolor (Species:
F = 16.0, P < 0.001) and drought-exposed plants had lower
Species
%N than control or shaded plants (F = 3.35, P = 0.042). The
Goodyera pubescens 44.14 ± 0.20 2.08 ± 0.10
Species × Treatment (F = 0.011, P = 0.99) and Species (Site)
Tipularia discolor 44.46 ± 0.14 2.59 ± 0.09
(F = 1.61, P = 0.18) interactions were not significant for %N.
Treatment
Drought and shade treated plants of both orchids were
Control 44.23 ± 0.20 2.37 ± 0.15
enriched in both 13C and 15N compared to controls (Figure 2),
Shade 44.59 ± 0.18 2.52 ± 0.10
but the species differences were only significant for ε13C
Drought 44.09 ± 0.23 2.13 ± 0.11
(Table 3). For both species, there were between plot differences
in ε13C and ε15N (Figure 3), but the difference between shade
but there were noticeable differences in relative growth between and drought treated plants was not significant (both P > 0.56).
sites within the two treatment sites (Figure 1A). Relative leaf Leaves of both orchids differed from the control F. grandifolia
growth was not different from zero, based on the size of the error leaves (Figure 2). As described in the Methods, we were able to
bars, in one of the drought treatment plots, and negative in two sample a single F. grandifolia seeding in a shaded and drought
T. discolor plot. The shaded seedling had ε13of the shaded plots (Figure 1A). C = −0.51 and
ε15N = 0.022, and the drought-exposed seedling had ε13C = 0.53
and ε15N = −0.06 (Figure 2).
Stable isotope abundances and
nitrogen and carbon concentrations Ecophysiology
There were no significant species, treatment, or sites effects As expected, photosynthesis of T. discolor leaves differed
for leaf C (Table 2; ANOVA: Species: F = 1.81, P = 0.18; between treatments, as demonstrated by significantly different
FIGURE 2
Isotopic enrichment factors for 13C and 15N orchids (G. pubescens: shades of green/yellow) and (T. discolor: shades of blue) exposed to control
(medium green or blue), drought (yellow or light blue), and shade (dark green or dark blue) treatments. Within each species, values for orchids
from the three sites are presented as different symbols, though overlapping symbol shapes for the two species do not indicate shared sites. The
green box around 0,0 indicates the mean (defined as 0) and standard error of enrichment factors for autotrophic reference plants (Fagus
grandifolia). The isotopic enrichment factors for the single surviving shaded and drought-treated F. grandifolia are indicated as S and D,
respectively.
Frontiers in Ecology and Evolution 06 frontiersin.org
64
fevo-10-1047267 November 23, 2022 Time: 16:39 # 7
McCormick et al. 10.3389/fevo.2022.1047267
TABLE 3 The results of ANOVA tests for differences in carbon a previously difficult to discern connection between light levels
concentration (%C), nitrogen concentration (%N), and enrichment in
13C and 15N among orchid species, treatments (drought, shade, and and partial mycoheterotrophy (Schweiger et al., 2019). In
control), sites within species, and species × treatment interactions. contrast, Tĕšitel et al. (2018) provided an argument for the
retention of photosynthesis in partially mycoheterotrophic
df F P
plants, as a support for seed set. None of these studies
%C demonstrated the shifting of photosynthetic contributions with
Species 1 1.81 0.18 a manipulative experiment. We found that drought and shade
Treatment 2 1.57 0.22 stresses caused declines in parameters related to photosynthesis
Species(site) 4 0.89 0.48 and resulted in increased reliance on mycoheterotrophy. We
Species × Treatment 2 0.808 0.45 propose that the increased mycoheterotrophic contribution to
%N orchid nutrition facilitated the maintenance of similar growth
Species 1 16.0 <0.001 rates in both species and dark respiration in T. discolor despite
Treatment 2 3.35 0.042 the stress conditions.
Species(site) 4 1.61 0.18 The increased reliance on mycoheterotrophy was
Species × Treatment 2 0.011 0.99 interpreted from increased enrichment in 13C and 15N,
ε13C but isotopic composition can also be affected by other factors.
Species 1 394 <0.001 Both drought and shade resulted in enrichment in 13C.
Treatment 2 30.1 <0.001
However, both stressors can also have direct effects on 13C
Species(site) 4 0.872 0.48
composition. In the absence of a changed fungal contribution
Species × Treatment 2 10.3 <0.001
15 to carbon, we would have expected that a direct effect of shadeε N
on plant carbon cycling would led to depleted 13C, as shown
Species 1 2.13 0.15
Treatment 2 49.2 0.001 for a wide range of non-orchid autotrophs (Preiss et al., 2010;<
Species(site) 4 0.379 0.28 Liebel et al., 2015; Lallemand et al., 2018). This also appears to
Species Treatment 2 2.00 0.14 be borne out by the single shaded F. grandifolia seedling, which×
had a lower ε13C than the untreated autotrophic reference
plants. Importantly, we found that the actual difference was
quantum efficiency, α (F = 4.01, P = 0.04), and marginally in the opposite direction, demonstrating that the orchids were
significantly different rate of photosynthesis under light- more enriched in 13C, not less. The results suggest that our
saturating conditions, Asat (F = 3.48, P = 0.09). Dark respiration measurements may have underestimated the true increases in
(Rd) was highly variable and not statistically different across sites mycoheterotrophy that occurred as a result of shading.
or treatments (Table 4). Control plants had higher mean rates In contrast to the shade treatment, we hypothesized that
of light-saturated photosynthesis (Asat) than shaded or drought- drought-treated orchids would be unable to increase the fungal
treated plants (F = 8.93, P = 0.015), perhaps reflecting their lower contribution to C. However, we measured enrichment in 13C
stress level. Shaded plants had higher quantum efficiencies (α) in drought-treated orchids that was almost the same as for
than control and drought-treated plants (F = 7.07, P = 0.015), shade. Although we do not know the extent to which OMF are
demonstrating the effects of light limitation. There were also able to translocate water, a possible reason for not seeing the
significant differences among sites in rETR (F = 8.37, P = 0.002) expected decrease in ε13C could be that the drought imposed
and WUE (F = 4.25, P = 0.06) (Table 4). by the shelters was too localized to affect the fungi, which
might have been able to translocate water from outside the
shelters. For example, Ruth et al. (2011) found that arbuscular
Discussion mycorrhizal fungi were able to translocate water from one
chamber to an associated plant in another chamber. Another
This study offers experimental data supporting the possible explanation for the increase in ε13C could be the
hypothesis that light and drought stress can increase orchid direct effects of drought. The δ13C of plants experiencing
dependence on fungal carbon. Selosse and Roy (2008) drought has been found to increase 1–2h as a result of
and Motomura et al. (2010) hypothesized that increasing stomatal closure to minimize water loss and increased internal
mycoheterotrophy leads to the evolution of achlorophyllous, recycling of CO2 (e.g., Pollastrini et al., 2010). However, the
totally mycoheterotrophic plants, and that this might be single F. grandifolia seedling that was exposed to drought
triggered by very low light conditions. Others have shown was only slightly enriched in 13C compared to untreated
a correlation between light availability and plant isotopic autotrophic control plants, suggesting that while our isotopic
concentrations, indicative of fungal contribution to plant measurements might have led to an overestimate of increases in
carbon (e.g., Gebauer, 2005; Liebel et al., 2010; Preiss et al., mycoheterotrophy in response to drought, they do not seem to
2010), and a recent study used multiple isotopes to demonstrate negate them.
Frontiers in Ecology and Evolution 07 frontiersin.org
65
fevo-10-1047267 November 23, 2022 Time: 16:39 # 8
McCormick et al. 10.3389/fevo.2022.1047267
FIGURE 3
Isotopic enrichment factors for 13C and 15N for G. pubescens (A,B) and T. discolor (C,D) exposed to control, drought, and shade treatments.
Mean enrichment factor ± se is shown. Different colored bars indicate different sites.
TABLE 4 Mean ecophysiological parameters (±se) for T. discolor in three sites under control conditions or exposed to drought or shade.
Site Treatment rETR α WUE Asat Rd
1 Control 77.14 ± 5.87 0.28 ± 0.03 9.89 ± 0.98 6.78 ± 0.47 −0.71 ± 0.13
2 Control 58.55 ± 6.87 0.28 ± 0.02 34.73 5.87 −0.29
3 Control 92.23 ± 7.63 0.28 ± 0.01 14.6 ± 4 6.44 ± 1.75 −0.81 ± 0.08
1 Drought 71.72 ± 8.24 0.27 ± 0.01 9.22 ± 0.39 4.85 ± 0.29 −0.49 ± 0.08
2 Drought 47.31 ± 7.65 0.26 ± 0.02 22.69 ± 10.54 3.79 ± 0.3 −0.54 ± 0.05
3 Drought 103.20 0.23 11.36 ± 3.08 4.9 ± 1.59 −0.63 ± 0.11
1 Shade 83.03 ± 7.73 0.32 ± 0.01 9.00 3.32 −0.45
2 Shade 70.27 ± 6.83 0.3 ± 0.02 17.69 ± 0.56 3.68 ± 1.21 −0.87 ± 0.08
3 Shade 66.5 ± 10.62 0.31 ± 0.02 8.35 ± 0.21 4.51 ± 1.07 −0.54 ± 0.08
rETR, maximum relative electron transport rate; α, quantum efficiency; WUE, water use efficiency; Asat , light saturated photosynthesis; Rd , dark respiration. For some plots, only one
plant was available to measure, so the values for that plant are given with no standard error reported.
Another possible cause for the observed enrichment of 13C fungi, leading to shifts in orchid isotopic composition without
is that drought could disrupt the orchid-fungus relationship corresponding changes in quantitative contributions to plant
and cause orchids to associate with different, perhaps more carbon. This is only a realistic possibility for T. discolor because
drought-resistant, fungi. McCormick et al. (2006) found that it associates with diverse fungi that can include fungi that
G. pubescens in locations near the present study sites switched are ectomycorrhizal with surrounding trees, and so potentially
to associate with different fungal genets following a drought. isotopically very different (e.g., Gebauer and Meyer, 2003).
If such a switch happened in this experiment, the new fungi However, in many T. discolor individuals sampled over the
could have had different isotopic compositions than the original course of 10 years, we have not found any that were
Frontiers in Ecology and Evolution 08 frontiersin.org
66
fevo-10-1047267 November 23, 2022 Time: 16:39 # 9
McCormick et al. 10.3389/fevo.2022.1047267
sufficiently enriched in 15N to suggest a considerable input from can increase reliance on carbon derived from fungi when
ectomycorrhizal fungi. In G. pubescens, fungal associates belong photosynthetic capability was encumbered. Although it has
to a very narrow clade that we would expect to be isotopically previously been postulated that limited light availability could
very similar. determine reliance on fungi for carbon (Gebauer, 2005; Hynson
Patterns for 15N enrichment mirrored those for 13C, et al., 2009; Liebel et al., 2010; Preiss et al., 2010), this study
suggesting that both nitrogen and carbon were taken up provides experimental evidence that reduced light availability
from fungi, and that the uptake was affected by drought and drought can both increase mycorrhizal dependence in a
and shade stress. The strongest differences in 15N enrichment partially mycoheterotrophic orchid. These results, combined
that reflect the extent of mycoheterotrophy have been seen with the evidence that photosynthesis contributes primarily
in plants associating with ectomycorrhizal fungi, which are to above-ground parts of partially mycoheterotrophic plants
typically enriched in 15N relative to N obtained from inorganic (Tĕšitel et al., 2018), provide the framework on which to build
nutrients in the soil (e.g., Hynson et al., 2013). Saprotrophic a more detailed understanding of the evolution of mycorrhizal
fungi often have less enriched 15N abundance, depending on associations. Further, with investigation of other green orchids
what they decompose to obtain nutrients, and so the nitrogen and partially mycoheterotrophic plants, this promises to
“signal” for fungal contribution to plant nutrition is weaker advance understanding of the dynamics of mycoheterotrophy
(Bidartondo et al., 2004; Martos et al., 2009; Schweiger et al., in forest understories, and may help to explain why nearly
2019). Indeed, the observed increase in ε15N was far less all forest understory herbs are mycorrhizal (Whigham, 2004),
than what has been observed for orchids associated with fungi despite very low light availability, which would be expected to
that are simultaneously ectomycorrhizal with other plants, limit the availability of photosynthetic carbon to contribute to a
but it was still a detectable increase (e.g., Schiebold et al., bidirectional mycorrhizal association.
2018). This is particularly important for interpreting the direct
effects of drought and shade on plant photosynthesis, and
thus ε13C, because ε15N is expected to be unaffected by direct Data availability statement
effects of drought (Peuke et al., 2006) and shade, yet we saw
a similar increase in δ15N in drought-exposed and shaded The raw data supporting the conclusions of this
plants. Accordingly, the two surviving F. grandifolia plants in article will be made available by the authors, without
the drought and shade treatments both had ε15N of nearly undue reservation.
0. However, drought-treated orchids also had lower overall
nitrogen concentration, suggesting that less total nitrogen Author contributions
was taken up by the plants and that perhaps the fungal
contribution to drought-treated plants, while a proportional MM and KG conceived the bulk of the study with additions
increase compared to unstressed control orchids, nevertheless by DW and TM. KG (2009–2011), MM (2009–2022), and TM
could have been less than for shaded plants. (2011–2022) carried out the study. MM analyzed the data.
The differences in photosynthetic parameters for T. discolor MM and KG wrote the manuscript. DW and TM edited the
shed some light on how drought and shade affected plant manuscript. All authors contributed to the article and approved
physiology and complemented isotopic results. In particular, the submitted version.
we found lower light saturated photosynthesis (Asat) in shade
and drought stressed orchids; suggesting that photosynthetic
rates were lower in stressed plants. Shaded plants were able Funding
to photosynthesize more efficiently (greater α), a common
adaptation to low light levels, but this was likely not enough KG was funded by the Environmental Leadership Center
to make up for the greatly decreased light availability in at Warren Wilson College. MM was funded in part by the US
this treatment. Other physiological parameters, WUE, and National Park Service (PMIS 120082).
Rd, were unchanged, suggesting that T. discolor was able to
maintain many aspects of their physiology, despite decreased
photosynthetic ability. Both orchid species were also able Acknowledgments
to maintain similar growth rates despite shade and drought
stress, perhaps pointing to the importance of increased We thank Amy Boyd, Lou Weber, and Phil Jamison for
mycoheterotrophy for surviving stressful conditions. their advice on study design and analysis and Jay O’Neill,
Our results suggest that increased mycoheterotrophy may Anne Chamberlain, Christine France, and Gary Peresta for
be triggered by stresses that limit photosynthetic ability their advice and technical support during the experiment.
and not just by limited light. The effects of experimental Isotopic samples were analyzed at the Smithsonian’s Museum
treatments on plant ε13C and ε15N indicated that the orchids Conservation Institute.
Frontiers in Ecology and Evolution 09 frontiersin.org
67
fevo-10-1047267 November 23, 2022 Time: 16:39 # 10
McCormick et al. 10.3389/fevo.2022.1047267
Conflict of interest organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
The authors declare that the research was conducted in the claim that may be made by its manufacturer, is not guaranteed
absence of any commercial or financial relationships that could or endorsed by the publisher.
be construed as a potential conflict of interest.
Supplementary material
Publisher’s note
The Supplementary Material for this article can be
All claims expressed in this article are solely those of the found online at: https://www.frontiersin.org/articles/10.3389/
authors and do not necessarily represent those of their affiliated fevo.2022.1047267/full#supplementary-material
References
Alexander, C., and Hadley, G. (1985). Carbon movement between host and Hynson, N. A. (2016). The carbon and nitrogen ecophysiologies of two
mycorrhizal endophyte during the development of the orchid Goodyera repens. endemic tropical orchids mirrors those of their temperate relatives and the local
New Phytol. 101, 657–665. doi: 10.1111/j.1469-8137.1985.tb02871.x environment. R. Soc. Open Sci. 3:160427. doi: 10.1098/rsos.160427
Bidartondo, M., Burghardt, B., Gebauer, G., Bruns, T., and Read, D. (2004). Hynson, N. A., Madsen, T. P., Selosse, M.-A., Adam, I. K. U., Ogura-Tsujita,
Changing patterns in the dark, isotopic and molecular evidence of ectomycorrhizal Y., Roy, M., et al. (2013). “The physiological ecology of mycoheterotrophy,” in
liaisons between forest orchids and trees. Proc. Biol. Sci. 271, 1799–1806. doi: Mycoheterotrophy: The biology of plants living on fungi, ed. V. S. F. T. Merckx
10.1098/rspb.2004.2807 (New York, NY: Springer), 297–342. doi: 10.1007/978-1-4614-5209-6_8
Bitterlich, M., Franken, P., and Graefe, J. (2019). Atmospheric drought and Hynson, N. A., Mambelli, S., Amend, A., and Dawson, T. E. (2012). Measuring
low light impede mycorrhizal effects on leaf photosynthesis–a glasshouse study carbon gains from fungal networks in understory plants from the tribe Pyroleae
on tomato under naturally fluctuating environmental conditions. Mycorrhiza 29, (Ericaceae): A field manipulation and stable isotope approach. Oecologia 169,
13–28. doi: 10.1007/s00572-018-0872-6 307–317. doi: 10.1007/s00442-011-2198-3
Brundrett, M. C., and Kendrick, B. (1988). The mycorrhizal status, root Hynson, N. A., Preiss, K., and Gebauer, G. (2009). Is it better to give than
anatomy, and phenology of plants in a sugar maple forest. Can. J. Bot. 66, receive? A stable isotope perspective to orchid-fungal carbon transport in the
1153–1173. doi: 10.1139/b88-166 green orchid species Goodyera repens and G. oblongifolia. New Phytol. 182, 8–11.
Brundrett, M. C., and Tedersoo, L. (2018). Evolutionary history of mycorrhizal doi: 10.1111/j.1469-8137.2009.02778.x
symbioses and global host plant diversity. New Phytol. 220, 1108–1115. doi: 10. Jacquemyn, H., Brys, R., Waud, M., Evans, A., Figura, T., and Selosse, M.-
1111/nph.14976 A. (2021). Mycorrhizal communities and isotopic signatures in two partially
Cabral, C., Ravnskov, S., Tringovska, I., and Wollenweber, B. (2016). Arbuscular mycoheterotrophic orchids. Front. Plant Sci. 12:618140. doi: 10.3389/fpls.2021.
mycorrhizal fungi modify nutrient allocation and composition in wheat (Triticum 618140
aestivum L.) subjected to heat-stress. Plant Soil 408, 385–399. doi: 10.1007/s11104-
016-2942-x Lallemand, F., Figura, T., Damesin, C., Fresneau, C., Griveau, C., Fontaine,N., et al. (2019). Mixotrophic orchids do not use photosynthates for perennial
Cameron, D. D., Johnson, I., Read, D. J., and Leake, J. R. (2008). Giving and underground organs. New Phytol. 221, 12–17. doi: 10.1111/nph.15443
receiving, measuring carbon cost of mycorrhizas in the green orchid, Goodyera
repens. New Phytol. 180, 176–184. doi: 10.1111/j.1469-8137.2008.02533.x Lallemand, F., Robionek, A., Courty, P.-E., and Selosse, M.-A. (2018). The13C content of the orchid Epipactis palustris (L.) Crantz responds to light as in
Cameron, D. D., Leake, J. R., and Read, D. J. (2006). Mutualistic mycorrhiza in autotrophic plants. Bot. Lett. 165, 265–273. doi: 10.1080/23818107.2017.1418430
orchids, evidence from plant-fungus carbon and nitrogen transfers in the green-
leaved terrestrial orchid Goodyera repens. New Phytol. 171, 405–416. doi: 10.1111/ Leake, J. (1994). The biology of mycoheterotrophic (saprotrophic) plants. New
j.1469-8137.2006.01767.x Phytol. 127, 171–216. doi: 10.1111/j.1469-8137.1994.tb04272.x
Gebauer, G. (2005). “Partnertausch im dunklen wald-stabile isotope geben Leake, J. R., and Cameron, D. D. (2010). Physiological ecology of
neue einblicke in das ernährungsverhalten von orchideen,” in Rundgespräche der mycoheterotrophy. New Phytol. 185, 601–605. doi: 10.1111/j.1469-8137.2009.
komission für ökologie, Vol. 30, ed. Bayerische Akademie der Wissenschaften 03153.x
(München: Verlag Dr. Friedrich Pfeil), 55–67. Liebel, H. T., Bidartondo, M. I., and Gebauer, G. (2015). Are carbon and
Gebauer, G., and Meyer, M. (2003). 15N and 13C natural abundance of nitrogen exchange between fungi and the orchid Goodyera repens affected by
autotrophic and mycoheterotrophic orchids provides insight into nitrogen and irradiance? Ann. Bot. London 115, 251–261. doi: 10.1093/aob/mcu240
carbon gain from fungal association. New Phytol. 160, 209–216. doi: 10.1046/j. Liebel, H. T., Bidartondo, M. I., Preiss, K., Segreto, R., Stöckel, M., Rodda, M.,
1469-8137.2003.00872.x et al. (2010). C and N stable isotope signatures reveal constraints to nutritional
Gebauer, G., Preiss, K., and Gebauer, A. C. (2016). Partial mycoheterotrophy modes in orchids from the Mediterranean and Macaronesia. Am. J. Bot. 97,
is more widespread among orchids than previously assumed. New Phytol. 211, 903–912. doi: 10.3732/ajb.0900354
11–15. doi: 10.1111/nph.13865 Martos, F., Dulormne, M., Pailler, T., Bonfante, P., Faccio, A., Fournel, J., et al.
Gorchov, D. L., Thompson, E., O’Neill, J., Whigham, D., and Noe, D. A. (2011). (2009). Independent recruitment of saprotrophic fungi as mycorrhizal partners by
Treefall gaps required for establishment, but not survival, of invasive Rubus tropical achlorophyllous orchids. New Phytol. 18, 668–681. doi: 10.1111/j.1469-
phoenicolasius in deciduous forest, Maryland, USA. Plant Spec. Biol. 26, 221–234. 8137.2009.02987.x
doi: 10.1111/j.1442-1984.2011.00317.x McCormick, M. K., Whigham, D. F., and O’Neill, J. (2004). Mycorrhizal
Hale, A. N., Tonsor, S. J., and Kalisz, S. (2011). Testing the mutualism disruption diversity in photosynthetic terrestrial orchids. New Phytol. 163, 425–438. doi:
hypothesis, physiological mechanisms for invasion of intact perennial plant 10.1111/j.1469-8137.2004.01114.x
communities. Ecosphere 9:110. doi: 10.1890/ES11-00136.1 McCormick, M. K., Whigham, D. F., Sloan, D., O’Malley, K., and Hodkinson,
Hughes, N. M., Gigantino, G. M., Grace, M. H., Hoffman, K. M., Lila, M. A., B. (2006). Orchid-fungus fidelity, a marriage meant to last? Ecology 87, 903–911.
Willans, B. N., et al. (2019). Photosynthetic profiles of green, purple, and spotted- doi: 10.1890/0012-9658(2006)87[903:OFAMMT]2.0.CO;2
leaf morphotypes of Tipularia discolor (Orchidaceae). Southeast. Nat. 18, 641–658. Merckx, V., Stöckel, M., Fleischmann, A., Bruns, T. D., and Gebauer, G. (2010).
doi: 10.1656/058.018.0415 15N and 13C natural abundance of two mycoheterotrophic and a putative partially
Frontiers in Ecology and Evolution 10 frontiersin.org
68
fevo-10-1047267 November 23, 2022 Time: 16:39 # 11
McCormick et al. 10.3389/fevo.2022.1047267
mycoheterotrophic species associated with arbuscular mycorrhizal fungi. New Selosse, M.-A., and Roy, M. (2008). Green plants that feed on fungi, facts and
Phytol. 188, 590–596. doi: 10.1111/j.1469-8137.2010.03365.x questions about mixotrophy. Trends Plant Sci. 14, 64–70. doi: 10.1016/j.tplants.
Motomura, H., Selosse, M.-A., Martos, F., Kagawa, A., and Yukawa, T. (2010). 2008.11.004
Mycoheterotrophy evolved from mixotrophic ancestors, evidence in Cymbidium Selosse, M.-A., Faccio, A., Scappaticci, G., and Bonfante, P. (2004).
(Orchidaceae). Ann. Bot. 106, 573–581. doi: 10.1093/aob/mcq156 Chlorophyllous and ahlorophyllous specimens of Epipactis microphylla (Neotieae,
Ogura-Tsujita, Y., Gebauer, G., Hashimoto, T., Umata, H., and Yukawa, T. Orchidaceae) are associaed with ectomycorrhizal septomycetes, including truffles.
(2009). Evidence for novel and specialised mycorrhizal parasitism, the orchid Microb. Ecol. 47, 416–426. doi: 10.1007/s00248-003-2034-3
Gastrodia confusa gains carbon from saprotrophic Mycena. Proc. Biol. Sci. 276, Smith, S., and Read, D. (1997). Mycorrhizal symbiosis, 2nd Edn. San Diego, CA:
761–767. doi: 10.1098/rspb.2008.1225 Academic Press.
Peuke, A. D., Gessler, A., and Rennenberg, H. (2006). The effect of drought on C Stöckel, M., Meyer, C., and Gebauer, G. (2011). The degree of
and N stable isotopes in different fractions of leaves, stems, and roots of sensitive mycoheterotrophic carbon gain in green, variegated and vegetative albino
and tolerant beech ecotypes. Plant Cell Environ. 29, 823–835. doi: 10.1111/j.1365- individuals of Cephalanthera damasonium is related to leaf chlorophyll
3040.2005.01452.x concentrations. New Phytol. 189, 790–796. doi: 10.1111/j.1469-8137.2010.03510.x
Pollastrini, M., Desotgiu, R., Cascio, C., Bussotti, F., Cherubini, P., Saurer, M., Suetsugu, K., Yamato, M., Masubayashi, J., and Tayasu, I. (2019). Comprative
et al. (2010). Growth and physiological responses to ozone and mild drought study of nutritional mode and mycorrhizal fungi in green and albino variants of
stress of tree species with different ecological requirements. Trees Struct. Funct. Goodyera velutina, an orchid mainly utilizing saprotrophic rhizoctonia. Mol. Ecol.
24, 695–704. doi: 10.1007/s00468-010-0439-4 28, 4290–4299. doi: 10.1111/mec.15213
Preiss, K., Adam, I. K. U., and Gebauer, G. (2010). Irradiance governs Tĕšitel, J., Tĕšitelová, T., Minasiewicz, J., and Selosse, M.-A. (2018). Mixotrophy
exploitation of fungi, fine-tuning of carbon gain by two partially myco- in land plants, Why to stay green? Trends Plant Sci. 23:8. doi: 10.1016/j.tplants.
heterotrophic orchids. Proc. Biol. Sci. 277, 1333–1336. doi: 10.1098/rspb.2009. 2018.05.010
1966 Tissue, D. T., Skillman, J. B., McDonald, E. P., and Strain, B. R. (1995).
Preiss, K., and Gebauer, G. (2008). A methodologial approach to improve Photosynthesis and carbon allocation in Tipularia discolor (Orchidaceae), a
estimates of nutrient gains by partially myco-heterotrophic plants. Isotopes wintergreen understory herb. Am. J. Bot. 82, 1249–1256. doi: 10.1002/j.1537-2197.
Environ. Health. Stud. 44, 393–401. doi: 10.1080/10256010802507458 1995.tb12658.x
Ralph, P. J., and Gademann, R. (2005). Rapid light curves, a powerful tool to Trudell, S., Rygiewicz, P., and Edmonds, R. L. (2003). Nitrogen and carbon
assess photosynthetic activity. Aquat. Bot. 82, 222–237. doi: 10.1016/j.aquabot. stable isotope abundances support the myco-heterotrophic nature and host-
2005.02.006 specificity of certain achlorophyllous plants. New Phytol. 160, 391–401. doi: 10.
Rasmussen, H. N., and Rasmussen, F. N. (2007). Trophic relationships in 1046/j.1469-8137.2003.00876.x
orchid mycorrhiza – diversity and implications for conservation. Lankesteriana Wang, B., and Qiu, Y.-L. (2006). Phylogenetic distribution and evolution of
7, 334–341. doi: 10.15517/lank.v7i1-2.19560 mycorrhizas in land plants. Mycorrhiza 16, 299–363. doi: 10.1007/s00572-005-
Rasmussen, H. N., and Whigham, D. F. (2002). Phenology of roots and 0033-6
mycorrhiza in orchid species differing in phototrophic strategy. New Phytol. 154, Wang, D., Jacquemyn, H., Gomes, S. I. F., Vos, R. A., and Merckx, V. S. F. T.
797–807. doi: 10.1046/j.1469-8137.2002.00422.x (2021). Symbiont switching and trophic mode shifts in Orchidaceae. New Phytol.
Ruth, B., Khalvati, M., and Schmidhalter, U. (2011). Quantification of 231, 791–800. doi: 10.1111/nph.17414
mycorrhizal water uptake via high-resolution on-line water content sensors. Plant Whigham, D. F. (2004). The ecology of woodland herbs. Annu.
Soil 342, 459–468. doi: 10.1007/s11104-010-0709-3 Rev. Ecol. Syst. 35, 583–621. doi: 10.1146/annurev.ecolsys.35.021103.10
Schiebold, J. M. I., Bidartondo, M., Lenhard, F., Makiola, A., and Gebauer, 5708
G. (2018). Exploiting mycorrhizas in broad daylight, partial mycoheterotrophy Yagame, T., Orihara, T., Selosse, M.-A., Yamato, M., and Iwase, K. (2012).
is a common nutritional strategy in meadow orchids. J. Ecol. 106, 168–178. doi: Mixotrophy of Platanthera minor, an orchid associated with ectomycorrhiza-
10.1111/1365-2745.12831 forming Ceratobasidiaceae fungi. New Phytol. 193, 178–187. doi: 10.1111/j.1469-
Schweiger, J. M.-I., Bidartondo, M. I., and Gebauer, G. (2018). Stable isotope 8137.2011.03896.x
signatures of underground seedlings reveal the organic matter gained by adult
orchids from mycorrhizal fungi. Funct. Ecol. 32, 870–881. doi: 10.1111/1365-2435. Zhu, X. C., Song, F. B., Liu, S. Q., and Liu, F. L. (2011). Effects of
13042 arbuscular mycorrhizal fungus on photosynthesis and water status of maizeunder high temperature stress. Plant Soil 346, 189–199. doi: 10.1007/s11104-011-
Schweiger, J. M.-I., Kemnade, C., Bidartondo, M. I., and Gebauer, G. (2019). 0809-8
Light limitation and partial mycoheterotrophy in rhizoctonia-associated orchids.
Oecologia 189, 375–383. doi: 10.1007/s00442-019-04340-0 Zimmer, K., Hynson, N. A., Gebauer, G., Allen, E. B., Allen, M. F., andRead, D. J. (2007). Wide geographical and ecological distribution of nitrogen
Selosse, M. -A., and Martos, F. (2014). Do chlorophyllous orchids and carbon gains from fungi in pyroloids and monotropoids (Ericaceae)
heterotrophically use mycorrhizal fungal carbon?. Trends Plant Sci. 19, 683–685. and in orchids. New Phytol. 175, 166–175. doi: 10.1111/j.1469-8137.2007.
doi: 10.1016/j.tplants.2014.09.005 02065.x
Frontiers in Ecology and Evolution 11 frontiersin.org
69
fevo-10-1059136 December 13, 2022 Time: 8:14 # 1
TYPE Original Research
PUBLISHED 13 December 2022
DOI 10.3389/fevo.2022.1059136
Host tree species effects on
OPEN ACCESS long-term persistence of
EDITED BY
Dennis Whigham, epiphytic orchid populations
Smithsonian Institution, United States
REVIEWED BY
Spyros Tsiftsis, Adriana Ramírez-Martínez1, Tamara Ticktin2 and
International Hellenic University, Demetria Mondragon1
Greece *
Bertrand Schatz, 1Instituto Politécnico Nacional, Centro Interdisciplinario de Investigación para el Desarrollo Integral
Centre National de la Recherche Regional (CIIDIR), Unidad Oaxaca, Oaxaca, México, 2School of Life Sciences, University of Hawai’i at
Scientifique (CNRS), France Mānoa, Honolulu, HI, United States
*CORRESPONDENCE
Demetria Mondragon
dmondragon@ipn.mx
SPECIALTY SECTION The destinies of epiphytic orchids (about 70% of all orchids) are linked to their
This article was submitted to
Conservation and Restoration Ecology, host trees. However, there is little information on if differences in host trees
a section of the journal characteristics can affect the long-term persistence of orchid populations,
Frontiers in Ecology and Evolution
and how this might vary under different climatic conditions. We compared
RECEIVED 30 September 2022
ACCEPTED 21 November 2022 the population dynamics of two epiphytic orchid species, Alamania punicea
PUBLISHED 13 December 2022 and Oncidium brachyandrum growing on two host trees with contrasting
CITATION leaf phenologies: the deciduous Quercus martinezii and the semideciduous
Ramírez-Martínez A, Ticktin T and Q. rugosa, over 3 years with varying levels of rainfall, in a montane tropical
Mondragon D (2022) Host tree
species effects on long-term oak forest in Oaxaca, Mexico. Using data from > 500 individuals growing
persistence of epiphytic orchid on 63 host trees, we applied linear mixed effects models, Integral Projection
populations.
Front. Ecol. Evol. 10:1059136. Models, and Life Table Response Experiments to identify the effects of host
doi: 10.3389/fevo.2022.1059136 tree on orchid vital rates and population growth rates. For both orchid species,
COPYRIGHT survival and growth did not differ between host species during wettest year.
© 2022 Ramírez-Martínez, Ticktin and
Mondragon. This is an open-access However, during the driest year both vital rates were higher on the semi-
article distributed under the terms of deciduous host Q. rugosa than on the deciduous Q. martinezii. Host species
the Creative Commons Attribution did not affect fecundity for A. punicea, but for O. brachyandrum fecundity was
License (CC BY). The use, distribution
or reproduction in other forums is higher on the deciduous host. For A. punicea, λ values were similar between
permitted, provided the original hosts during the wettest and intermediate years, but significantly lower (1
author(s) and the copyright owner(s)
are credited and that the original λ = 0.28) on the deciduous than on the semi-deciduous host during the
publication in this journal is cited, in driest year. This was due primarily to lower survival on the deciduous host.
accordance with accepted academic
practice. No use, distribution or For O. brachyandrum, λ was slightly higher (1 λ = 0.03) on the deciduous
reproduction is permitted which does than the semideciduous host during the wettest year, due to higher growth
not comply with these terms.
and reproduction. However, during the intermediate and driest years, λ values
were significantly higher on the semi-deciduous than on the deciduous host
(1 λ = 0.13 and 0.15, respectively). This was due to higher survival and
growth. A. punicea populations appear more vulnerable to dry conditions
than O. brachyandrum, likely due to its smaller pseudobulbs, and hence lower
water-storing capacity. Our results show that host tree species can both
influence the vital rates and the long-term dynamics of orchid populations,
Frontiers in Ecology and Evolution 01 frontiersin.org
70
fevo-10-1059136 December 13, 2022 Time: 8:14 # 2
Ramírez-Martínez et al. 10.3389/fevo.2022.1059136
and these effects vary across orchids species and over time. Our results
highlight the importance of maintaining a diversity of host trees to ensure
long-term population persistence.
KEYWORDS
Orchidaceae, population dynamics, host preference, integral projection models, life
table response experiment, Quercus, Oncidium, Alamania
Introduction Although there is a large literature on the effects of host traits
on orchid germination, establishment and survival, there is little
The orchid family is comprised of 850 genera with more information on whether differences in vital rates may scale up
than 30,000 species, and 50% of these species are concentrated across the life cycle to differentially affect population persistence.
in tropical areas of the world (Chase et al., 2015). In Mexico, Similarly, if and how these differences may shift with climatic
there are approximately 1,200 species of orchids. Nearly 40% of conditions is largely unexplored. For example, annual variation
Mexican orchids are endemic. Some species of Mexican orchids, in climatic conditions can influence the demographic behavior
including epiphytes, are very attractive to horticulturists, of epiphytes (Mondragón et al., 2004; Ticktin et al., 2016).
collectors, and public, and are extracted from their natural Populations of epiphytes growing on host species that allow for
habitats and sold illegally, which can lead to them being higher humidity due to their phenology, architecture or bark
threatened or extirpated (Halbinger and Soto, 1997; Merritt water holding capacity might perform better during dry years,
et al., 2014). but not during average or wettest years. The one epiphyte study
For vascular epiphytes, the presence of their host trees that has assessed this showed that the vital rates and population
is essential for the establishment and permanence of their growth rates of an epiphytic bromeliad were different when
populations. However, due to differences in host traits, not growing on perennial pine vs. deciduous oak host trees (Ticktin
all trees offer the same conditions for the establishment and et al., 2016). Populations on oaks had higher fecundity, but
development of epiphytic orchids (Wagner et al., 2015). For those on pines had higher survival and growth. Growth rates
example, rough bark texture can affect the capture of seeds of populations on both host genera increased with increasing
[rugose and scaly barks favor seed adherence compared to dry season rainfall, but the effect was larger for populations on
smooth barks (Adhikari and Fischer, 2011; Gowland et al., oaks. The authors concluded that the presence of both pine and
2013; Timsina et al., 2016)], while an ability to retain and oak trees is very important for long-term conservation of these
release water can favor the germination of seeds [barks with bromeliad populations.
higher water retention capacity and slower release rates favor Although we are unaware of other studies that have
seed germination (Callaway et al., 2002; Einzmann et al., addressed this question for orchids, it is of great importance
2015)]. Similarly, the presence of allelopathic compounds in for developing conservation and management plans. Like many
the bark of trees can limit seed germination and establishment other wild species, populations of orchids are threatened by
of epiphytes (Callaway et al., 2002; Harshani et al., 2014); the habitat loss and conversion to monocultures of timber species
rate of bark exfoliation and the fragility of branches can lead to or other types of plantations (Boelter et al., 2011; Mondragón
differential mortality rates as a result of epiphyte falls (López- Chaparro et al., 2015), where diversity of host tree species is
Villalobos et al., 2008); and the nutrient quality of stemflows considerably diminished, for such as reforestation with pine
and throughfalls could affect growth and fertility of the epiphytic species only (Jiménez-Bautista et al., 2014), or substitution of
orchids (higher amounts of nutrients could increase growth and native shade trees from coffee or cocoa plantations by Inga spp.
fecundity rates). Finally, the leaf phenology of the host trees trees (Peeters et al., 2003; Valencia et al., 2016). These changes,
can affect the demography of epiphytes (Einzmann et al., 2015; in combination with changes in distribution of preferred hosts
Ticktin et al., 2016). In addition, these and other host trees of species of epiphytes due to climate change (Hsu et al.,
characteristics can not only influence epiphytes directly, but also 2012), could present problems for the long-term persistence of
indirectly by providing different microclimatic conditions for epiphytic orchid populations.
the mycorrhizal fungal community, which are indispensable for We carried out a demographic study to provide the first
the germination of the orchid seeds. Different trees may possess test of whether and how host species can affect the population
different communities of fungi that may or may not favor the dynamics of epiphytic orchids. We focus on two epiphytic
germination of orchids (Otero et al., 2007; Rasmussen et al., orchid species Alamania punicea Lex. in La Llave and Lex.
2015). and Oncidium brachyandrum Lindl, growing on two congeneric
Frontiers in Ecology and Evolution 02 frontiersin.org
71
fevo-10-1059136 December 13, 2022 Time: 8:14 # 3
Ramírez-Martínez et al. 10.3389/fevo.2022.1059136
FIGURE 1
Geographic location of the study area within a seasonal oak forest in the municipality of Yanhuitlán, Oaxaca, Mexico.
host trees: the fully deciduous Q. martinezii Neé and the semi populations (Mondragón Chaparro et al., 2015; Cortes-
deciduous Quercus rugosa Neé. We addressed the following Anzures et al., 2017), would be lower on Q. rugosa due to
questions: its thicker branches.
(2) Higher survival of individuals on the semideciduous
i) Do orchid vital rates (survival, growth, reproduction) vary Q. rugosa would translate into higher population growth
between host tree species? rates (λ values), since population growth rates of long-
ii) Do differences in vital rates translate into differences in lived species are highly sensitive to differences in survival
population growth rates? (Franco and Silvertown, 2004). We also expected that the
iii) Do differences in vital rates and population growths rates difference in λ values may be greater in drier years than in
vary with inter-annual climatic variation? wettest years.
We hypothesized that:
Materials and methods
(1) Survival and reproduction would be higher for both
orchids on semi-deciduous Q. rugosa than fully deciduous Study area and species
Q. martinezii, due to lower light penetration and higher
humidity in their treetops of the former, which can This study was carried out in an oak forest in Tooxi,
help orchids avoid photoinhibition (de la Rosa-Manzano municipality of Yanhuitlán, Oaxaca, Mexico located in the
et al., 2014; Einzmann et al., 2015) and increase flowering Sierra Madre del Sur physiographical province (17◦33′57.34′′
(Cervantes et al., 2005). In addition, we expected that N and 97◦22′19.28′′ W, elevation 2,579 m a.s.l; Figure 1)
branch fall, one of the main causes of mortality in epiphyte that encompasses the Mixteca Alta UNESCO Global Geopark
Frontiers in Ecology and Evolution 03 frontiersin.org
72
fevo-10-1059136 December 13, 2022 Time: 8:14 # 4
Ramírez-Martínez et al. 10.3389/fevo.2022.1059136
TABLE 1 Precipitation patterns in Tooxi, Yanhuitlán, Oaxaca from 2018 to 2020.
Year Annual precipitation Dry season (November–April) total/ Range:
(mm) average dry season monthly precipitation minimum/month-maximum/month
(mm/month) during dry season (mm/month)
2018 1058 264.8/44.1± 24.2 14.9–84.7
2019 947 103.1/17.28± 4.3 10–22.3
2020 1041 125.2/20.9± 15.1 4.3–48.2
(García-Sánchez et al., 2021). Mean annual precipitation is
753 ± 152 mm, with a dry season average of 86 ± 67 mm. Alamania punicea Lex. is an epiphytic perennial orchid, 3–
The mean temperature of the dry season is 14.4 ± 1.3◦C, 6 cm high including the inflorescence; ovoid pseudobulbs,
and the average maximum and minimum temperatures slightly elongated, covered by translucent papyrus sheaths,
are 23.3 ± 1.8◦C and 7.0 ± 2.9, respectively (INIFAP, 7–10 mm long; leaves 2, rarely 3, at the apex of the
2021; Table 1). Tree vegetation includes deciduous, semi- pseudobulb, elliptic to oblanceolate sheets, 1–4 cm long, 5–
deciduous and evergreen trees and is comprised mainly 10 mm wide; flowers 7–14, red to pinkish reddish. Fruits
of Quercus candicans Neé, Q. castanea Neé, Q. crassifolia are capsules with dust-like seeds. There is no report of
Humb. and Bonpl., Q. rugosa Neé, Q. martinezii C.H. Müll., A. punicea breeding system. Stpiczyńska et al. (2005) suggest
Juniperus flaccida Schltdl., and Arbutus xalapensis Kunth. that given its floral morphology, it could be pollinated by
The epiphytic vegetation includes Tillandsia bourgaei Baker, hummingbirds. Alamania punicea is an endemic species
T. macdougallii, T. plumosa Baker, T. prodigiosa (Lem.) Baker, prevalent in cool and seasonally dry Quercus-Pinus forests
T. recurvata (L.) L., T. usneoides (L.) L., Pleopeltis conzatti on the Trans-Mexican Volcanic Belt and the Sierra Madre
(Weath.) R. M. Tryon and A.F.Tryon, Polypodium martensii Oriental above 1,900 m a.s.l. (where oaks are dominant;
Mett., Echeveria nodulosa (Baker) Otto, O. brachyandrum García-Cruz et al., 2003; Soto Arenas, 2005; UNEP-WCMC,
and A. punicea. 2020).
We selected two species of tree hosts, Q. rugosa and
Q. martinezii (Table 2) for this study. Quercus rugosa is
distributed from Texas and Arizona, in the USA, to the Oncidium brachyandrum is an epiphytic perennial orchid,
Sierra Madre de Chiapas in Guatemala. Its populations up to 20 cm high with clustered pseudobulbs, ovoid to
are abundant in mountainous areas of western and central ellipsoid or subglobose, somewhat laterally compressed,
Mexico in temperate sub-humid climate, between the 1,800 2–3 cm long; 2 or 3 lateral leaves; flowers 2 or 3,
and 2,800 m a.s.l. Quercus martinezii is endemic to Mexico simultaneous, showy, 25–30 mm in diameter, sepals, and
and distributed in the states of Aguascalientes, Guanajuato, petals brown or yellow with irregular brown spots and
Hidalgo, Guerrero, Jalisco, Nuevo Leon, Nayarit, Puebla, yellow lip. Fruits are capsules with dust-like seeds. This
Queretaro, San Luis Posoti, Tamaulipas, Veracruz, and species probably is pollinated by oil-collecting or bombini
Oaxaca from 2,000 to 2,500 m a.s.l (Valencia-A, 2004). bees and might be self-incompatible as reported for other
The other species that hosted orchids (Q. crassifolia and members of the genera (Damon and Cruz-López, 2006;
Q. candicans, J. flaccida and A. xalapensis) had few individuals Pemberton, 2008). This orchid is distributed in Honduras,
on them. Guatemala, and in Mexico in the states of Durango,
We focused on the two most abundant orchid species Guerrero, Jalisco, Michoacán de Ocampo, Morelos, Nayarit,
out of the three present at the study site. In addition, Sinaloa, Zacatecas, and Oaxaca. It grows mainly in oak
Alamania punicea is listed in CITES Appendix II, is the forests elevations of 2,000–2,500 m a.s.l. (Jiménez et al.,
only species of a monotypic genus endemic to Mexico, 1998).
and nothing is known about the demography of this
species that could help to establish management and Precipitation, the most limiting factor for epiphytes (Zotz
conservation strategies. Oncidium brachyandrum is harvested and Hietz, 2001; Laube and Zotz, 2003), varied across our three
commercially species in Tlaxiaco, Oaxaca which is close study years, especially during the dry season (Table 1). Although
to the study site (Ticktin et al., 2020). Nothing is known precipitation in all 3 years was above the mean of the past
about the demography of this species; there is only one
demographic study on another species of the genus (Oncidium 20 years (Supplementary material), we refer to these years as
poikilostalix Kraenzl.) M.W. Chase and N.H. Williams but the wettest year (2018), driest year (2019) where dry season
in coffee plantations. We describe each orchid species as precipitation was less than half that of the wettest year–and
follows: intermediate year (2020).
Frontiers in Ecology and Evolution 04 frontiersin.org
73
fevo-10-1059136 December 13, 2022 Time: 8:14 # 5
Ramírez-Martínez et al. 10.3389/fevo.2022.1059136
TABLE 2 Host traits of two Quercus species in a seasonal oak forest in of individuals of each host species and tested for differences
Yanhuitlán, Oaxaca, Mexico. between hosts using ANOVAs (Table 2; Ramírez-Martínez,
Trait Quercus martinezii Quercus rugosa 2022).
Leaf phenology Deciduous Semi-deciduous
Leaf area (cm2)* 39.96± 4.61b 59.70± 11.94a
Tree height (m) 9.69± 1.43a 7.74± 2.05b Data analyses
(n = 21) (n = 42)
Diameter at breast height (cm) 36.11± 12.59ns 29.09± 17.72ns Host effects on demographic rates (vital rates)
(n = 21) (n = 42) We tested differences in survival, growth, and reproduction
Bark roughness (cm)† 5.87± 0.39ns 5.53 0.59ns± of orchid individuals on the different Quercus species using
(n = 10) (n = 10) generalized linear mixed models (GLMMs). Initial size (log-
Water holding capacity of bark 0.28 ns± 0.07 0.33± 0.09ns transformed), host species (Q. martinezii vs. Q. rugosa), and
(ml/cm3) (n = 5) (n = 5)
year were fixed effects and individual orchid nested within
Canopy openness (%)
individual host tree were random effects. We used regression
Dry season 30.74 6.74ns 33.10 ns± ± 8.91
ns analyses to identify which measure of size (e.g., numberWettest season 38.07± 5.81 38.07± 2.24ns
of pseudobulbs, size of the pseudobulb, etc.) was the best
Relative humidity (%)
ns ns predictor of growth, reproduction, and survival. We usedDry season 75.49± 15.96 80.38± 14.04
ns ns Akaike’s information criterion (AICc) to compare model fit.Wettest season 71.44± 10.32 76.99± 9.30
We found that for A. punicea, the best predictor was an
Mean temperature (◦C)
ns ns index of number of leaves times area of the longest leafDry season 14.17± 0.57 14.52± 0.95
(calculated with the formula for an oval). For O. brachyandrum,
Wettest season 13.76 1.56ns± 13.47 ns± 1.79
ns ns the best predictor was pseudobulb area (calculated with theConcentration of phosphorus in 0.16± 0.03 0.15± 0.05
throughfalls (mg/l) n = 5 n = 5 formula for an oval).
Concentration of potassium in 2.56 1.06ns 2.76 1.16ns To model the probability of survival, reproduction, and± ±
throughfalls (mg/l) n = 5 n = 5 probability of mortality due to desiccation, we used binomial
distributions, and to model the number of capsules, we used
Values are means ± SD for each tree species. Superscript lower-case letters indicate
significant differences (ANOVA: P < 0.05, Tukey HSD). a negative binomial model. To model growth (size at t + 1),
Significant differences across tree species [F(1,62) = 15.30, p = 0.0002]. we used Gaussian error structure with an exponential variance
*Significant differences across tree species [F(1,39) = 47.52, p = 0.0001].
†Information taken from Hernández-Álvarez (2021). structure, where the variance increases as an exponential
function of initial size (Zuur et al., 2009). We modeled the
probability of reproduction with the minimum size observed
We selected 21 Q. martinezii and 42 Q. rugosa trees within a for plant reproduction (sizes: A. punicea ≥ 1.44 cm2, and
1 ha plot and tagged and measured all the orchids of our study O. brachyandrum ≥ 0.79 cm2). We selected the best fit
species on them. We selected a higher number of Q. rugosa trees model based on the lowest Akaike (AICc). All analyses
since they had lower densities of orchids: on average there were were performed using the glmmTMB package in R v.
14 ± 11 O. brachyandrum plants/tree on Q. martinezii versus 4.1.1.
6± 6 on Q. rugosa; these values were 8± 6 plants/tree vs. 7± 6,
respectively, for A. punicea. For each of 3 years (2017–2020), we Host effects on population growth dynamics
recorded plant status (alive, dead), size, and fecundity (number Integral projection models
of capsules), and recorded the number of new seedlings. For We used integral projection models [IPMs (Easterling
both species of orchid, we measured height and width of the et al., 2000)] to project the long-term (asymptotic) population
largest pseudobulb and counted the number of pseudobulbs. growth rates (λ values) of each orchid species growing on
We also recorded causes of mortality distinguishing broadly each of the two host species. The IPMs are constructed from
between desiccation (the entire dead plant was still attached continuous functions that describe size-dependent growth,
to the tree) and falling (the plant was missing). Although survivorship, and fecundity. The IPM kernel is the sum of
individuals that die due to desiccation are susceptible to falling, two functions. One describes the survival, probability and
our monthly checks ensured that we were able to ascertain growth (or shrinkage) of survivors (p kernel), and the second
the correct cause of mortality. We did not include herbivory is the reproductive contribution of each individual and the size
as a cause of death since we did not observe orchid plants distribution of the new seedlings (f kernel). Our IPM took the
attacked by herbivores at our study site. This is consistent with form:
findings from other studies that rates of herbivory are mostly ∫
low in epiphytes (Benzing, 1990; Zotz, 1998; Winkler et al., Un (y, t + 1) = [p (x, y)+ f (x, y)]n (x, t) dx
2005). Finally, we measured host characteristics on a subsample L
Frontiers in Ecology and Evolution 05 frontiersin.org
74
fevo-10-1059136 December 13, 2022 Time: 8:14 # 6
Ramírez-Martínez et al. 10.3389/fevo.2022.1059136
For both orchid species, the p(x,y) kernel was represented TABLE 3 Estimated coefficients from mixed-effect models of the
by the survival probability of individuals of size x to probability of survival, growth, reproduction, and probability ofproducing capsules for Alamania punicea plants growing on two
size y attributable to size-dependent survival, s(x), and Quercus species.
growth g(x,y): p(x,y) = s(x) g(x,y). The fertility kernel
Fixed effects Estimate SE Z value P-value
f (x,y) denotes the production of new seedlings of size
(x) produced from plants of size (y). This was calculated Probability of surviving to t + 1†
for plants of reproductive size as: f (x,y) = s(x) fn(x) Intercept 1.6513 0.4654 3.548 0.000388
pEfd(y), where s(x) is the survival of plants of size (x), Size at start 0.5076 0.1049 4.837 1.32e− 06
fn(x) is the probability of producing capsules for plant Year 2 (2018–2019) −1.3216 0.4157 −3.179 0.001477
size x times the number of capsules per plant size x; pE Year 3 (2019–2020) −1.0069 0.4871 −2.067 0.038704
is the number of new seedlings produced per capsule, Host species (Q. rugosa) 0.4296 0.6309 0.681 0.495931
and fd(y) is the size distribution of new seedlings. For Year 2× Q. rugosa 1.1342 0.6219 1.824 0.068194
each host species, pE was calculated as the number of Year 3× Q. rugosa −0.5589 0.6066 −0.921 0.356817
seedlings observed in the field divided by the total number Size at t + 1 of surviving individuals (growth)
of capsules produced. We calculated the asymptotic Intercept 0.30439 0.10874 2.80 0.00512
projected population growth rate (λ) for each IPM using Size at start 0.83544 0.02640 31.65 < 2e-16
the popbio 2.7 package in R (Stubben and Milligan, 2007) Year 2 (2018–2019) −0.52526 0.11704 −4.49 7.2e− 06
and obtained 95% confidence intervals by bootstrapping Year 3 (2019–2020) 0.39822 0.12676 3.14 0.00168
(N = 100). Host species (Q. rugosa) −0.06609 0.12641 −0.52 0.60110
We used life table response experiments (LTREs) (Caswell, Year 2× Q. rugosa 0.33882 0.16004 2.12 0.03425
2001) to identify which vital rate contributed most to the Year 3× Q. rugosa −0.11524 0.17310 −0.67 0.50557
observed differences in population growth rates between hosts, Probability of producing capsules at time t (for individuals≥ 35 cm)†
for each orchid species and year. Intercept −4.8208 0.7344 −6.564 5.23e− 11
Size at start 0.5040 0.1692 2.980 0.00289
Year 2 (2018–2019) 0.5833 0.6006 0.971 0.33141
Results Year 3 (2019–2020) 0.4047 0.7060 0.573 0.56647
Host species (Q. rugosa) 0.9169 0.6554 1.399 0.16183
Host effects on vital rates Year 2× Q. rugosa 0.0587 0.7462 0.079 0.93730
Year 3× Q. rugosa −20.3430 7179.93 −0.003 0.99774
Alamania punicea Capsules produced per reproductive plant at time t*
Survival Intercept −0.7825 0.4918 −1.591 0.111585
For individuals on both host species, survival increased as Size at start 0.4162 0.1160 3.590 0.000331
a function of size and was highest during the wettest year and Host species (Q. rugosa) 0.5830 0.3702 1.575 0.115327
lowest during the driest year (Table 3 and Figure 2A). The †Binomial (logit) GLMM.
best fit model included an interaction between host species and GLMM with Gaussian error structure and an exponential variance structure.
year such that difference in survival between individuals on *Negative binomial GLMM.
semideciduous Q. rugosa and the deciduous Q. martinezii was
least during the wettest year. The probability of mortality due did the number of capsules produced by reproducing plants
to desiccation decreased significantly with (log) size (β =−0.69, (Table 3).
SE = 0.09, z = −7.35, p < 0.001) but did not differ significantly
between host trees. Oncidium brachyandrum
Survival
Growth For plants on both host species, survival increased as a
For individuals on both species, growth was lowest in the function of plant size. There was no difference in survival
driest year and highest in the intermediate year. During the dry between host species during the wettest year, but survival on
year, growth was higher on the semi-deciduous Q. martinezii the semi-deciduous Q. rugosa was higher than on the deciduous
than on the deciduous Q. martinezii (Table 3 and Figure 2B). Q. martinezii during the dry and intermediate years (Table 4
and Figure 3A). The probability of mortality due to desiccation
Reproduction decreased significantly with (log) size (β = −0.64, SE = 1.07,
Plants began to reproduce once they reached 1.08 cm2 size z = −6.00, p < 0.001) and was significantly higher in the
(number of leaves times area of the longest leaf). Only 1.4% intermediate year than in the wettest year (β =−1.63, SE = 0.57,
of plants flowered and 100% those produced capsules. The z =−2.89, p = 0.004). It did not differ significantly between host
probability of reproduction increased as a function of size, as trees.
Frontiers in Ecology and Evolution 06 frontiersin.org
75
fevo-10-1059136 December 13, 2022 Time: 8:14 # 7
Ramírez-Martínez et al. 10.3389/fevo.2022.1059136
FIGURE 2
(A) Survival and (B) growth as function of size, for Alamania punicea plants growing on two oak species Quercus martinezii (blue) and Q. rugosa
(green). Solid lines indicate year 2017–2018, dashed lines year 2018–2019, and dotted lines year 2019–2020.
TABLE 4 Estimated coefficients from mixed-effect models of the Reproduction
probability of survival, growth, reproduction, and probability of 2
producing capsules for Oncidium brachyandrum plants growing on Plants began to reproduce once they reached 0.78 cm
two Quercus species. (pseudobulb area); 14% of plants flowered and 100% of these
produced capsules. The probability of reproduction and the
Fixed effects Estimate SE Z value P-value
number of capsules produced per reproductive plant both
Probability of surviving to t + 1† increased as a function of size (Table 4). Host species was
Intercept 2.021234 0.363086 5.567 2.59e− 08 included in the best fit model for the probability of reproduction,
Size at start 0.404343 0.078653 5.141 2.74e− 07 with reproduction higher on the deciduous Q. martinezii than
Year 2 (2018–2019) −0.500072 0.275850 −1.813 0.0699 the semideciduous Q. rugosa (Table 4).
Year 3 (2019–2020) −0.357534 0.331662 −1.078 0.2810
Host species (Q. rugosa) −0.009813 0.598163 −0.016 0.9869
Year 2× Q. rugosa 1.320310 0.661457 1.996 0.0459 Host tree effect on population
Year 3× Q. rugosa 2.523453 1.119749 2.254 0.0242 dynamics
Size at t + 1 of surviving individuals (growth)
Intercept −0.27854 0.05545 −5.02 6.08e− 07 Population growth rates of A. punicea on the two host
Size at start 0.74661 0.02133 35.01 < 2e-16 species were similar in the wettest year and the intermediate
Year 2 (2018–2019) 0.26731 0.08206 3.26 0.00112 years (1 λ between hosts = 0.01). However, population growth
Year 3 (2019–2020) 0.82965 0.08549 9.70 < 2e-16 rates were significantly higher on the semideciduous host than
Host species (Q. rugosa) −0.06476 0.11067 −0.59 0.55847 the deciduous host during the driest year ((1 λ = 0.28;
Year 2 Q. rugosa 0.42201 0.159992 2.64 0.00832 Figure 4A). For O. brachyandrum, λ was slightly lower on×
Year 3 Q. rugosa 0.45812 0.16220 2.82 0.00474 Q. martinezii than on Q. rugosa during the wettest year (1× −
Probability of producing capsules at time t (for individuals 35 cm)† λ = 0.03). However, the reverse was true during the driest and≥
Intercept −3.4918 0.4120 8.476 < 2e-16 intermediate years, where λ values were significantly higher−
Size at start 1.5336 0.2978 5.150 2.62e 07 on Q. rugosa than on Q. martinezii (1 λ = 0.15 and 0.13,−
Host species (Q. rugosa) −0.8107 0.4413 −1.837 0.0662 respectively) (Figure 4B). For O. brachyandrum, λ values were
Capsules produce per reproductive plant at time t* higher in the dry and intermediate years, than in the wettest
Intercept −0.02442 0.27662 −0.088 0.9297 year.
Size at start 0.36621 0.19401 1.888 0.0591
LTRE analyses
†Binomial (logit) GLMM. For A. punicea, the higher λ value observed for populations
GLMM with Gaussian error structure and an exponential variance structure.
*Negative binomial GLMM. on Q. rugosa in the driest year was mainly due to higher survival
on that host tree. For the wettest and the intermediate rainfall
year, there was little differences in λ values across hosts, but
Growth differences were due to higher survival on Q. rugosa during the
For plants growing on both host species, growth was higher wettest year growth, and higher growth on Q. martinezii during
in dry and intermediate years than in wettest year. Growth was the intermediate year (Figures 5A–C). For O. brachyandrum
higher on the semi-deciduous Q. rugosa than on the deciduous higher growth and reproduction contributed the most to the
Q. martinezii only during the dry year (Table 4 and Figure 3B). higher λ value of Q. martinezii in wettest year. The higher λ
Frontiers in Ecology and Evolution 07 frontiersin.org
76
fevo-10-1059136 December 13, 2022 Time: 8:14 # 8
Ramírez-Martínez et al. 10.3389/fevo.2022.1059136
FIGURE 3
(A) Survival and (B) growth as function of size, for Oncidium brachyandrum plants growing on two oak species Quercus martinezii (blue) and
Q. rugosa (green). Solid lines indicate year 2017–2018, dashed lines year 2018–2019, and dotted lines year 2019–2020.
FIGURE 4
Population growth rates (λ values) of (A) Alamania punicea and (B) Oncidium brachyandrum growing on two Quercus species, from 2018 to
2020 in an oak forest in Oaxaca, Mexico. Error bars represent 95% confidence intervals.
values observed for populations in Q. rugosa during the two sunnier and drier microclimates during dry season on the latter,
drier years was due to higher growth (driest year), and higher which increased mortality due to desiccation. The same has been
survival (intermediate year) (Figures 5D–F). reported for an epiphytic bromeliad, where individuals growing
on perennial pines had higher survival and growth rates that
those growing on deciduous oak (Ticktin et al., 2016). Similarly,
Discussion Callaway et al. (2002) reported higher growth rates of epiphytic
individuals growing on their preferred host and attributed it to
The aim of our paper was to test if and how host species the higher water holding capacity its bark. Species with greater
can affect the population dynamics of epiphytic orchids. Our water holding capacity provide a more humid environment, and
results show that rates of survival, growth and reproduction of allow epiphytes access to water for a longer period. Humidity
epiphytic orchids can vary across host tree species, and this can is recognized as the more limiting factor within epiphytism
translate into difference in population growth rates. They also
(Benzing, 1990; Laube and Zotz, 2003; Zotz, 2013).
suggest that the direction and magnitude of these differences
appears to depend on climatic conditions. Our results suggest potential lag effects of drought, since for
Our finding that both orchid species had higher growth the semi-deciduous host, a higher number of orchid individuals
and survival on the semi-deciduous host than on the deciduous died the year after the driest year (intermediate year), than
host, during the driest year, but not during the wettest year, during the driest year itself. According to Zotz and Tyree (1996)
is consistent with other studies. Einzmann et al. (2015) found there can be long-term drought stress effects on the physiology
higher growth and survival rates of two epiphytic species of orchids that can be perceived afterward forcing plants to show
growing on perennial trees than on deciduous trees, due to the die back on some of their parts (like leaves), and finally die.
Frontiers in Ecology and Evolution 08 frontiersin.org
77
fevo-10-1059136 December 13, 2022 Time: 8:14 # 9
Ramírez-Martínez et al. 10.3389/fevo.2022.1059136
FIGURE 5
Life table response experiment contributions for Alamania punicea (A–C) and Oncidium brachyandrum (D–F) plants growing on two Quercus
species. Darker colors represent life-history transitions that make greater contributions to higher λ values observed. Values across the diagonal
represent contributions from survival, and those below diagonal represent contributions from growth. Fecundity contributions are represented
in the top right corner.
Our finding that fecundity of O. brachyandrum was related with drought tolerance. A. punicea has thick leaves
also higher on the deciduous host could be related to the and cuticles, and small pseudobulbs (7–10 mm long), while
low photosynthetic efficiency of epiphytic orchids due to O. brachyandrum has thin leaves and pseudobulbs 2–3 cm long.
drought adaptation (Sahagun-Godinez, 1996); they therefore These represent two of the main strategies of orchids for drought
need higher light (as in deciduous trees in our case) to tolerance (Stancato et al., 2001; Yang et al., 2016): thick cuticles
perform photosynthesis efficiently and produce photosynthates avoid water loss, while pseudobulbs store water. In our study,
for flower production. The lack of effect that we observed for the species with pseudobulbs better buffered the effect of host
A. punicea could be related to sample size, since very few tree on its growth and survival rates. Pseudobulbs play an
individuals flowered during our study. The low probability important role in the growth and survival of epiphytic orchids
of flowering is typical for many epiphytic orchids (Tremblay, since they not only store water but are also responsible for
2006). the partition of assimilates, carbohydrate and minerals and can
The differences we found in vital rates scaled up to perform photosynthesis (Ng and Hew, 2000). Further research
differences in population growth rates, with λ values much is needed to identify how morphological and physiological
lower on the deciduous host during the dry years than in the variation in epiphytic orchids (Dressler, 1993; Yang et al., 2016;
wettest year. Our finding that 1 λ between the wettest and dry Zhang et al., 2018) shapes demographic responses. In addition,
year was greater for A. punicea than for O. brachyandrum is translocation experiments could further disentangle differences
likely related to differences between the species in adaptation in demographic rates across hosts.
Frontiers in Ecology and Evolution 09 frontiersin.org
78
fevo-10-1059136 December 13, 2022 Time: 8:14 # 10
Ramírez-Martínez et al. 10.3389/fevo.2022.1059136
Our results highlight the importance of preserving tree Funding
species diversity to foster the long-term persistence of
populations of vascular epiphytes. This is especially true if we We are grateful for the financial support provided by
consider that vascular epiphytes function as metapopulations the Instituto Politécnico Nacional (projects SIP-20170823, SIP-
(Winkler et al., 2009; Valverde and Bernal, 2010), where 20180674, SIP-20195449, and SIP-20200025). AR-M received a
individuals growing on one host tree represent a sub-population studentship from Consejo Nacional de Ciencia y Tecnología
interconnected with other sub-populations (on other host (CONACYT) to carry out her Ph.D. studies.
trees) by seed dispersal, and where trees that support growing
sub-populations determine the growth of the metapopulation
(Winkler et al., 2009). Thus, metapopulation growth may be Acknowledgments
maximized when there is a diversity of host species that allow
for growth under varying climatic conditions. In addition,
We acknowledged the contributions of specific colleagues,
variation among genotypes of the same tree species could
institutions, or agencies that aided the efforts of the authors.
potentially affect demographic rates, given that Zytynska et al.
(2011) found a positive correlation between the genetic distances
among host trees and similarity among the community vascular
epiphytes growing on them. Further research is needed to Conflict of interest
better understand how host tree traits shape the persistence
of populations of epiphytic orchids. On the ground, working The authors declare that the research was conducted in the
with local communities to help identify land-use options that absence of any commercial or financial relationships that could
maintain tree diversity will be key. be construed as a potential conflict of interest.
Data availability statement Publisher’s note
The raw data supporting the conclusions of this article will All claims expressed in this article are solely those of the
be made available by the authors, without undue reservation. authors and do not necessarily represent those of their affiliated
organizations, or those of the publisher, the editors and the
reviewers. Any product that may be evaluated in this article, or
Author contributions claim that may be made by its manufacturer, is not guaranteed
or endorsed by the publisher.
AR-M conducted fieldwork and wrote most of the
manuscript. DM conceived the study, provided the advice,
financial support, conducted fieldwork, and contributed to Supplementary material
the manuscript. TT provided the advice, helped with data
analyses, contributed to the manuscript, and reviewed the The Supplementary Material for this article can be
English. All authors contributed to the article and approved the found online at: https://www.frontiersin.org/articles/10.3389/
submitted version. fevo.2022.1059136/full#supplementary-material
References
Adhikari, Y. P., and Fischer, A. (2011). Distribution pattern of the epiphytic Caswell, H. (2001). Matrix population models: construction, analysis, and
orchid Rhynchostylis retusa under strong human influence in Kathmandu valley, interpretation, 2nd Edn. New York, NY: Oxford University Press.
Nepal. Bot. Orient. J. Plant Sci. 8, 90–99. doi: 10.3126/botor.v8i0.5956 Cervantes, S. E., Graham, E. A., and Andrade, J. L. (2005). Light microhabitats,
Benzing, D. H. (1990). Vascular epiphytes: general biology and related biota, growth and photosynthesis of an epiphytic bromeliad in a tropical dry forest. Plant
1st Edn. Cambridge: Cambridge University Press, doi: 10.1017/CBO978051152 Ecol. 179, 107–118. doi: 10.1007/s11258-004-5802-3
5438 Chase, M. W., Cameron, K. M., Freudenstein, J. V., Pridgeon, A. M., Salazar, G.,
Boelter, C. R., Zartman, C. E., and Fonseca, C. R. (2011). Exotic tree Van den Berg, C., et al. (2015). An updated classification of Orchidaceae. Bot. J.
monocultures play a limited role in the conservation of Atlantic Forest epiphytes. Linn. Soc. 177, 151–174. doi: 10.1111/boj.12234
Biodivers. Conserv. 20, 1255–1272. doi: 10.1007/s10531-011-0026-z Cortes-Anzures, B. O., Corona-López, A. M., Toledo-Hernández, V. H.,
Callaway, R. M., Reinhart, K. O., Moore, G. W., Moore, D. J., and Pennings, S. C. Valencia-Díaz, S., and Flores-Palacios, A. (2017). Branch mortality influences
(2002). Epiphyte host preferences and host traits: mechanisms for species-specific phorophyte quality for vascular epiphytes. Botany 95, 709–716. doi: 10.1139/cjb-
interactions. Oecologia 132, 221–230. doi: 10.1007/s00442-002-0943-3 2017-0023
Frontiers in Ecology and Evolution 10 frontiersin.org
79
fevo-10-1059136 December 13, 2022 Time: 8:14 # 11
Ramírez-Martínez et al. 10.3389/fevo.2022.1059136
Damon, A. A., and Cruz-López, L. (2006). Fragrance in Relation to Pollination plantations in Southern Mexico. Agric. Ecosyst. Environ. 95, 481–493. doi: 10.1016/
of Oncidium sphacelatum and Trichocentrum oerstedii (Orchidaceae) in the S0167-8809(02)00204-9
Soconusco Region of Chiapas, Mexico. Selbyana 27, 186–194. Pemberton, R. W. (2008). Pollination of the ornamental orchid Oncidium
de la Rosa-Manzano, E., Andrade, J. L., Zotz, G., and Reyes-García, C. (2014). sphacelatum by the naturalized oil-collecting bee (Centris nitida) in Florida.
Epiphytic orchids in tropical dry forests of Yucatan, Mexico – Species occurrence, Selbyana 29, 87–91. doi: 10.1653/0015-4040(2008)091[0101:NOTOCB]2.0.CO;2
abundance and correlations with host tree characteristics and environmental
conditions. Flora 209, 100–109. doi: 10.1016/j.flora.2013.12.002 Ramírez-Martínez, A. (2022). Host tree effect on demography and phenology ofepiphytic species. Doctoral dissertation. Oaxaca: Instituto Politécnico Nacional.
Dressler, R. (1993). Phylogeny and classification of orchid family. Cambridge:
Cambridge University Press. Rasmussen, H. N., Dixon, K. W., Jersáková, J., and Těšitelová, T. (2015).Germination and seedling establishment in orchids: a complex of requirements.
Easterling, M. R., Ellner, S. P., and Dixon, P. M. (2000). Size-specific sensitivity: Ann. Bot. 116, 391–402. doi: 10.1093/aob/mcv087
applying a new structured population model. Ecology 81, 694–708. doi: 10.1890/
0012-9658(2000)081[0694:SSSAAN]2.0.CO;2 Sahagun-Godinez, E. (1996). Trends in the phenology of flowering in theOrchidaceae of Western Mexico. Biotropica 28, 130–136. doi: 10.2307/2388778
Einzmann, H. J. R., Beyschlag, J., Hofhansl, F., Wanek, W., and Zotz, G. (2015).
Host tree phenology affects vascular epiphytes at the physiological, demographic Soto Arenas, M. A. (2005). “Alamania,” in Genera orchidacearum.
and community level. AoB Plants 7:lu073. doi: 10.1093/aobpla/plu073 Epidendroideae (Part one), Vol. 4, eds A. M. Pridgeon, P. J. Cribb, M. W.Chase, and F. N. Rasmussen (Oxford: Oxford University Press).
Franco, M., and Silvertown, J. (2004). A comparative demography of plants
based upon elasticities of vital rates. Ecology 85, 531–538. Stancato, G. C., Mazzafera, P., and Buckeridge, M. S. (2001). Effect of a droughtperiod on the mobilisation of non-structural carbohydrates, photosynthetic
García-Cruz, P. J., Saldaña, S. L. M., Jiménez, M. R., and Solano, G. R. (2003). efficiency and water status in an epiphytic orchid 1Part of the Ph.D. thesis of
Orchidaceae. Tribu epidendreae. Flora Bajío Regiones Adyacentes 119, 1–173. Giulio Cesare Stancato (Instituto de Biologia, UNICAMP). Plant Physiol. Biochem.
García-Sánchez, L., Canet, C., Cruz-Pérez, M. Á, Morelos-Rodríguez, L., 39, 1009–1016. doi: 10.1016/S0981-9428(01)01321-3
Salgado-Martínez, E., and Corona-Chávez, P. (2021). A comparison between local Stpiczyńska, M., Davies, K. L., and Gregg, A. (2005). Comparative account of
sustainable development strategies based on the geoheritage of two post-mining nectary structure in Hexisea imbricata (Lindl.) Rchb.f. (Orchidaceae). Ann. Bot.
areas of Central Mexico. Int. J. Geoheritage Parks 9, 391–404. 95, 749–756. doi: 10.1093/aob/mci081
Gowland, K. M., van der Merwe, M. M., Linde, C. C., Clements, M. A., and Stubben, C., and Milligan, B. (2007). Estimating and analyzing demographic
Nicotra, A. B. (2013). The host bias of three epiphytic Aeridinae orchid species is models using the popbio package in R. J. Stat. Softw. 22, 1–23. doi: 10.18637/jss.
reflected, but not explained, by mycorrhizal fungal associations. Am. J. Bot. 100, v022.i11
764–777. doi: 10.3732/ajb.1200411 Ticktin, T., Mondragón, D., and Gaoue, O. G. (2016). Host genus and rainfall
Halbinger, F., and Soto, M. (1997). Laelias of Mexico, 15. México, DF: Orquídea, drive the population dynamics of a vascular epiphyte. Ecosphere 7:e01580. doi:
1–160. 10.1002/ecs2.1580
Harshani, H. C., Senanayake, S. P., and Sandamali, H. (2014). Host tree Ticktin, T., Mondragón, D., Lopez-Toledo, L., Dutra-Elliott, D., Aguirre-
specificity and seed germination of Dendrobium aphyllum (Roxb.) C.E.C. Fisch. León, E., and Hernández-Apolinar, M. (2020). Synthesis of wild orchid trade
in Sri Lanka. J. Natl. Sci. Found. 42, 71–86. doi: 10.4038/jnsfsr.v42i1.6682 and demography provides new insight on conservation strategies. Conserv. Lett.
Hernández-Álvarez, G. R. (2021). Efecto del forofito sobre la germinación 13:e12697. doi: 10.1111/conl.12697
de alamania punicea y oncidium brachyandrum. Master dissertation. Oaxaca: Timsina, B., Rokaya, M. B., Münzbergová, Z., Kindlmann, P., Shrestha, B.,
Instituto Politécnico Nacional. Bhattarai, B., et al. (2016). Diversity, distribution and host-species associations
Hsu, R. C.-C., Tamis, W. L. M., Raes, N., de Snoo, G. R., Wolf, J. H. D., of epiphytic orchids in Nepal. Biodivers. Conserv. 25, 2803–2819. doi: 10.1007/
Oostermeijer, G., et al. (2012). Simulating climate change impacts on forests and s10531-016-1205-8
associated vascular epiphytes in a subtropical island of East Asia. Divers. Distrib. Tremblay, R. L. (2006). The effect of population structure, plant size, herbivory
18, 334–347. doi: 10.1111/j.1472-4642.2011.00819.x and reproductive potential on effective population size in the temperate epiphytic
INIFAP (2021). Sistema de base de datos meteorológicos. Available online orchid, Sarcochilus australis. Cunninghamia 9, 529–535.
at: https://clima.inifap.gob.mx/lnmysr/DatosIndirectos/Essenger (accessed UNEP-WCMC, (2020). CITES Secretariat, Geneva, Switzerland. Compiled by
September 30, 2020). UNEP-WCMC, Cambridge, UK. The checklist of CITES species website. Available
Jiménez, R. M., Sánchez, L. S., and García-Cruz, J. (1998). Orchidaceae. Flora del online at: https://efloramex.ib.unam.mx/cdm_dataportal/taxon/35a4cfa8-6d5e-
bajío y de regiones adyacentes. Fascículo 67, 38. 48cd-a5f7-b3dd3b380c81#distribuci%C3%B3n (accessed September 15, 2022).
Jiménez-Bautista, L., Damon, A., Ochoa-Gaona, S., and Tapia, R. C. (2014). Valencia, V., Naeem, S., García-Barrios, L., West, P., and Sterling, E. J. (2016).
Impact of silvicultural methods on vascular epiphytes (ferns, bromeliads and Conservation of tree species of late succession and conservation concern in coffee
orchids) in a temperate forest in Oaxaca, Mexico. For. Ecol. Manage. 329, 10–20. agroforestry systems. Agric. Ecosyst. Environ. 219, 32–41. doi: 10.1016/j.agee.2015.
doi: 10.1016/j.foreco.2014.05.053 12.004
Laube, S., and Zotz, G. (2003). Which abiotic factors limit vegetative growth Valencia-A, S. (2004). Diversidad del género Quercus (Fagaceae) en México. Bot.
in a vascular epiphyte? Funct. Ecol. 17, 598–604. doi: 10.1046/j.1365-2435.2003.00 Sci. 75, 33–53. doi: 10.17129/botsci.1692
760.x Valverde, T., and Bernal, R. (2010). ¿Hay asincronía demográfica entre
López-Villalobos, A., Flores-Palacios, A., and Ortiz-Pulido, R. (2008). The poblaciones locales de Tillandsia recurvata?: Evidencias de su funcionamiento
relationship between bark peeling rate and the distribution and mortality of two metapoblacional. Bol. Soc. Bot. Mex. 23–36.
epiphyte species. Plant Ecol. 198, 265–274. doi: 10.1007/s11258-008-9402-5 Wagner, K., Mendieta-Leiva, G., and Zotz, G. (2015). Host specificity in vascular
Merritt, D. J., Hay, F. R., Swarts, N. D., Sommerville, K. D., and Dixon, K. W. epiphytes: a review of methodology, empirical evidence and potential mechanisms.
(2014). Ex situ conservation and cryopreservation of orchid germplasm. Int. J. AoB Plants 7:lu092. doi: 10.1093/aobpla/plu092
Plant Sci. 175, 46–58. doi: 10.1086/673370 Winkler, M., Hülber, K., and Hietz, P. (2009). Population dynamics of epiphytic
Mondragón, D., Durán, R., Ramírez, I., and Valverde, T. (2004). Temporal orchids in a metapopulation context. Ann. Bot. 104, 995–1004. doi: 10.1093/aob/
variation in the demography of the clonal epiphyte Tillandsia brachycaulos mcp188
(Bromeliaceae) in the Yucatán Peninsula. Mexico. J. Trop. Ecol. 20, 189–200. doi: Winkler, M., Hülber, K., Mehltreter, K., Franco, J. G., and Hietz, P.
10.1017/S0266467403001287 (2005). Herbivory in epiphytic bromeliads, orchids and ferns in a Mexican
Mondragón Chaparro, D. M., Valverde Valdés, M. T., and Hernández Apolinar, montane forest. J. Trop. Ecol. 21, 147–154. doi: 10.1017/S026646740400
M. (2015). Population ecology of epiphytic angiosperms: A review. Trop. Ecol. 56, 2081
1–39. Yang, S.-J., Sun, M., Yang, Q.-Y., Ma, R.-Y., Zhang, J.-L., and Zhang, S.-B. (2016).
Ng, C. K. Y., and Hew, C. S. (2000). Orchid pseudobulbs–false’bulbs with a Two strategies by epiphytic orchids for maintaining water balance: thick cuticles in
genuine importance in orchid growth and survival! Sci. Hortic. 83, 165–172. leaves and water storage in pseudobulbs. AoB Plants 8:lw046. doi: 10.1093/aobpla/
plw046
Otero, J. T., Aragón, S., and Ackerman, J. D. (2007). Site variation in spatial
aggregation and phorophyte preference in Psychilis monensis (Orchidaceae). Zhang, S., Yang, Y., Li, J., Qin, J., Zhang, W., Huang, W., et al. (2018).
Biotropica 39, 227–231. doi: 10.1111/j.1744-7429.2006.00258.x Physiological diversity of orchids. Plant Divers. 40, 196–208.
Peeters, L. Y. K., Soto-Pinto, L., Perales, H., Montoya, G., and Ishiki, M. Zotz, G. (1998). Demography of the epiphytic orchid, Dimerandra emarginata.
(2003). Coffee production, timber, and firewood in traditional and Inga-shaded J. Trop. Ecol. 14, 725–741. doi: 10.1017/S0266467498000534
Frontiers in Ecology and Evolution 11 frontiersin.org
80
fevo-10-1059136 December 13, 2022 Time: 8:14 # 12
Ramírez-Martínez et al. 10.3389/fevo.2022.1059136
Zotz, G. (2013). The systematic distribution of vascular epiphytes – a critical Zuur, A. F., Ieno, E. N., Walker, N. J., Saveliev, A. A., and Smith, G. M. (2009).
update. Bot. J. Linn. Soc. 171, 453–481. doi: 10.1111/boj.12010 Mixed effects models and extensions in ecology with R. 574. New York, NY: Springer.
Zotz, G., and Hietz, P. (2001). The physiological ecology of vascular epiphytes: doi: 10.1007/978-0-387-87458-6
current knowledge, open questions. J. Exp. Bot. 52, 2067–2078. doi: 10.1093/ Zytynska, S. E., Fay, M. F., Penney, D., and Preziosi, R. F. (2011). Genetic
jexbot/52.364.2067 variation in a tropical tree species influences the associated epiphytic plant and
Zotz, G., and Tyree, M. T. (1996). Water stress in the epiphytic orchid, invertebrate communities in a complex forest ecosystem. Philos. Trans. R. Soc. 366,
Dimerandra emarginata (G. Meyer) Hoehne. Oecologia 107, 151–159. 1329–1336.
Frontiers in Ecology and Evolution 12 frontiersin.org
81
TYPE Original Research
PUBLISHED 28 December 2022
DOI 10.3389/fenvs.2022.912428
Will Greenland be the last refuge
OPEN ACCESS for the continental European
EDITED BY
Dennis Whigham, small-white orchid?Niche
Smithsonian Institution, United States
REVIEWED BY modeling of future distribution of
Vladan Djordjević,
University of Belgrade, Serbia
Ole Bennike, Pseudorchis albida
Geological Survey of Denmark and
Greenland, Denmark
Jasmina Šinžar-Sekulić, Marta Kolanowska1,2, Sławomir Nowak3* and
University of Belgrade, Serbia
Agnieszka Rewicz1
*CORRESPONDENCE
Sławomir Nowak, 1Department of Geobotany and Plant Ecology, Faculty of Biology and Environmental Protection,
slawomir.nowak@ug.edu.pl University of Lodz, Lodz, Poland, 2Department of Biodiversity Research, Global Change Research
SPECIALTY SECTION Institute AS CR, Brno, Czech Republic,
3Department of Plant Taxonomy and Nature Conservation,
This article was submitted to Faculty of Biology, University of Gdańsk, Gdańsk, Poland
Conservation and Restoration Ecology,
a section of the journal
Frontiers in Environmental Science
RECEIVED 04 April 2022 Climate change affects populations of plants, animals, and fungi not only by
ACCEPTED 07 December 2022
28 December 2022 direct modifications of their climatic niches but also by altering their ecologicalPUBLISHED
interactions. In this study, the future distribution of suitable habitats for the
CITATION
Kolanowska M, Nowak S and Rewicz A small-white orchid (Pseudorchis albida) was predicted using ecological niche
(2022), Will Greenland be the last refuge modeling. In addition, the effect of global warming on the spatial distribution
for the continental European small-
and availability of the pollen vectors of this species was evaluated. Due to the
white orchid?Niche modeling of future
distribution of Pseudorchis albida. inconsistency in the taxonomic concepts of Pseudorchis albida, the differences
Front. Environ. Sci. 10:912428. in the climatic preferences of three proposed subspecies were investigated.
doi: 10.3389/fenvs.2022.912428
Due to the overlap of both morphological and ecological characters of
COPYRIGHT ssp. albida and ssp. tricuspis, they are considered to be synonyms, and the
© 2022 Kolanowska, Nowak and
Rewicz. This is an open-access article final analyses were carried out using ssp. albida s.l. and ssp. straminea. All of the
distributed under the terms of the models predict that with global warming, the number of suitable niches for
Creative Commons Attribution License
(CC BY). The use, distribution or these orchids will increase. This significant increase in preferred habitats is
reproduction in other forums is expected to occur in Greenland, but habitat loss in continental Europe will be
permitted, provided the original severe. Within continental Europe, Pseudorchis albida ssp. albidawill lose 44%–
author(s) and the copyright owner(s) are
credited and that the original 98% of its suitable niches and P. albida ssp. straminea will lose 46%–91% of its
publication in this journal is cited, in currently available habitats. An opposite effect of global warming was predicted
accordance with accepted academic
practice. No use, distribution or for pollinators of P. albida s.l., and almost all insects studied will be subject to
reproduction is permitted which does habitat loss. Still, within the predicted potential geographical ranges of the
not comply with these terms. orchid studied, some pollen vectors are expected to occur, and these can
support the long-term survival of the small-white orchid.
KEYWORDS
bioclimatic preferences, habitat loss, global warming, pollinators, Pseudorchis albida
ssp. tricuspis, Pseudorchis albida ssp. straminea
Frontiers in Environmental Science 01 frontiersin.org
82
Kolanowska et al. 10.3389/fenvs.2022.912428
1 Introduction In terms of anthropogenic threats, P. albida is endangered by
agricultural development and afforestation (Reinhammar et al.,
The sole representative of the genus Pseudorchis Ség., the 2002; Foley and Clarke, 2005; Forbes and Northridge, 2012).
small-white orchid (Pseudorchis albida (L.) Á. Löve and D. Löve), Reduction in traditional mowing and grazing has resulted in it
is a tuberous perennial geophyte growing in most of Europe and being overgrown by more competitive species (Reinhammar
northern Asia from Spain and Iceland to northwest Siberia et al., 2002; Holland et al., 2008). On the other hand, reduced
(Jersáková et al., 2011). This species is variable within its seed set and recruitment can result from over-grazing (Duffy
geographical range, and the morphological differences et al., 2009; Jersáková et al., 2011). The effect of global warming
prompted taxonomists to divide P. albida into three on this species is yet to be evaluated.
subspecies: ssp. albida, ssp. straminea (Fern.) Ä. Löve and D. According to the IUCNRed List, Pseudorchis albida is assessed
Löve, and ssp. tricuspis (Beck) Klein (Figure 1; Reinhammar, as a species of least concern because it is rather widespread
1998; Klein, 2000; Jersáková et al., 2011). However, the (Rankou, 2011). However, due to a considerable decline in its
recognition of these taxa is still debated. Reinhammar (1995, distribution, it is currently considered to be critically endangered in
1998) recognized two species in the genus Pseudorchis, with Greece (small population found by Tsiftsis and Antonopoulos,
moderate morphometric support. The studies on their allozymes 2011), vulnerable in Great Britain (Cheffings and Farrell, 2005)
also indicate that it is reasonable to accept the species status of the and Bulgaria (Petrova and Vladimirov, 2009), endangered in
lowland to subalpine P. albida s.s., and alpine P. straminea Ireland (Curtis and McGough, 1988), Czech Republic (Holub
(Reinhammar and Hedren, 1998). Klein (2000) accepted three and Procházka, 2000), Germany (Ludwig and Schnittler, 1996),
subspecies of P. albida: ssp. tricuspis (calcicolous, with an and Sweden (Gärdenfors, 2010), and near threatened in Norway
alpine–boreal distribution), ssp. albida (acidophilous, with (Artsdatabanken, 2010) and Poland (Kaźmierczakowa et al., 2016).
alpine–temperate–boreal distribution), and ssp. straminea It is also protected in many European countries (Reinhammar
(basiphilous, with west Arctic–north Atlantic distribution). et al., 2002; Bilz et al., 2011), e.g., Poland (Kaźmierczakowa et al.,
Jersáková et al. (2011) considered that the taxa of Pseudorchis 2016), Czech Republic (Danihelka et al., 2012), Denmark
characterized by differences in distribution are not well-defined (Damgaard et al., 2020), Romania (Sârbu et al., 2020), Ukraine
and accept the broad concept of P. albida s.l. On the other hand, (Kricsfalusy et al., 1999, 2010), Slovakia (Turis et al., 2014),
Bateman et al. (2017) recognized Pseudorchis albida and P. Norway (subordinate agency, 2022), Sweden
straminea as separate species, pointing out morphological (Naturva˚rdsverket, 2022), Austria (Zulka et al., 2001; Jersáková
features and molecular divergence (ITS) sufficient for species- et al., 2011), Germany (Jersáková et al., 2011), Switzerland
level distinction. The same authors rejected the separateness of P. (Jersáková et al., 2011), and Italy (Jersáková et al., 2011).
tricuspis due to overlap in supposedly taxonomically useful This study aimed to estimate the effect of global warming on
characters with P. albida and P. straminea (Bateman et al., the distribution of climatic niches suitable for P. albida s.l. Since
2017). Considering the differences in the taxonomic approach, this orchid relies mainly on sexual reproduction, the effect of
it was decided to accept all taxa as subspecies in this ecological climate change was also evaluated for the pollinators of this
study. The results of the analyses can be used in further orchid. To improve the estimates and because the taxonomic
taxonomic studies on Pseudorchis. separateness of P. albida ssp. tricuspis is questioned by some
Pseudorchis albida ssp. albida occurs in areas with a authors (Bateman et al., 2017), the differences in the preferred
boreal–montane climate and is found from United Kingdom climatic niches of the three-known subspecies of P. albida were
across Scandinavia to the northern Urals in the European part of evaluated in order to assess their ecological distinctiveness.
Russia as well as in mountain ranges from Spain across the Alps
to the Eastern Carpathians (Jersáková et al., 2011). Pseudorchis
albida ssp. straminea is restricted to areas with a west 2 Materials and methods
Arctic–north Atlantic climate (Iceland, Faroes, Greenland, and
Scandinavia; Jersáková et al., 2011), and Pseudorchis albida 2.1 List of localities
ssp. tricuspis is restricted to alpine–boreal areas (Swiss, Italian
and Austrian Alps, Tatra Mountains, and Eastern Carpathian; The databases of localities of Pseudorchis albida s.l. in
Jersáková et al., 2011). continental Europe as well as records of pollinators of this
Populations of small-white orchid reproduce mainly sexually orchid were compiled based on information in public facilities
(Jersáková et al., 2011), and vegetative propagation by tubers accessed through the Global Biodiversity Information Facility
contributes little to population growth (Summerhayes, 1951). As a (GBIF 2020; Supplementary Table S1). The information on
species that provides a nectar reward, several species of Lepidoptera pollen vectors was obtained from previous reports on
(Claessens andKleynen, 2011) are reported pollinators of Pseudorchis pollination of P. albida by Claessens and Kleynen (2011) and
albida.More recently, Jersáková et al. (2011) also reported species of Jersáková et al. (2011). There was an insufficient number of
Empis (Diptera) as diurnal pollen vectors. occurrences for Empis bistortae Meigen, 1822 for performing an
Frontiers in Environmental Science 02 frontiersin.org
83
Kolanowska et al. 10.3389/fenvs.2022.912428
FIGURE 1
Photographs of the small-white orchid in its natural habitat. Pseudorchis albida ssp. albida in Rhön, Germany (A), and Zillertal Alps, Austria [(B);
photographer: Marco Klüber/www.m-klueber.de], Pseudorchis albida ssp. straminea in Newfoundland, Canada [(C,D); photographer: James
Fowler], and Pseudorchis albida ssp. tricuspis on Mt. Mangart, Julian Alps, Slovenia [(E,F); photographer: Amadej Trnkoczy].
analysis for this insect. From a total of 4518 localities for 2.2 Principal component analysis
Pseudorchis albida (ssp. albida—316, ssp. straminea—4170,
and tricuspis—32) and 69424 for insects (Chrysoteuchia Principal components analysis (PCA) was used to evaluate
culmella (Linnaeus, 1758)—46299, Crambus ericella (Hübner, the differences between populations of P. albida ssp. straminea,
1813)—1098, Crambus pascuella L.—4032, Plutella xylostella P. albida ssp. albida, and P. albida ssp. tricuspis based on
(Linnaeus, 1758)—17643, and Udea uliginosalis (Stephens, 19 bioclimatic variables from WorldClim v. 2.1 (Table 1; Fick
1834)—352) available in the repositories, only records that and Hijmans, 2017). Calculations were carried out using the
were georeferenced with a minimum of 1 km precision were software package Statistica PL. ver. 13.3 (StatSoft Inc. 2011). The
selected. To reduce sampling bias, spatial thinning was carried data matrix was transformed (square root) before carrying out
out using SDMtoolbox 2.3 for ArcGIS (Kremen et al., 2008; the ordination analysis.
Brown, 2014). The data were rarified by designating a minimal
distance of 5 km for calculating climatic habitat heterogeneity.
The final database included 28 localities for P. albida ssp. albida, 2.3 Ecological niche modeling
414 for P. albida ssp. straminea, 11 for P. albida ssp. tricuspis
(Supplementary Data Sheet S1), and 3694 for its pollinators The modeling of the current and future distribution of the
(Chrysoteuchia culmella—1472, Crambus ericella—249, species studied was carried out using the maximum entropy
Crambus pascuella—707, Plutella xylostella—1244, and Udea method implemented in MaxEnt version 3.3.2 (Phillips et al.,
uliginosalis—22; Supplementary Data Sheet S2). 2004, 2006; Elith et al., 2011) based on presence-only
Frontiers in Environmental Science 03 frontiersin.org
84
Kolanowska et al. 10.3389/fenvs.2022.912428
TABLE 1 List of variables used in the PCA and modeling (with an asterisk).
Variable code Description
bio1* Annual mean temperature
bio2* Mean diurnal range (mean of monthly (max temp–min temp)
bio3* Isothermality (bio2/bio7) (×100)
bio4* Temperature seasonality (standard deviation ×100)
bio5 Max temperature in the warmest month
bio6 Min temperature in the coldest month
bio7 Temperature annual range (bio5–bio6)
bio8* Mean temperature in the wettest quarter
bio9* Mean temperature in the driest quarter
bio10 Mean temperature in the warmest quarter
bio11 Mean temperature in the coldest quarter
bio12* Annual precipitation
bio13 Precipitation in the wettest month
bio14* Precipitation in the driest month
bio15* Precipitation seasonality (coefficient of variation)
bio16 Precipitation in the wettest quarter
bio17 Precipitation in the driest quarter
bio18* Precipitation in the warmest quarter
bio19 Precipitation in the coldest quarter
observations. For the modeling, bioclimatic variables in 30 arc- 2.5 arc-minutes were re-scaled to fit bioclimatic variables. SSPs
seconds of the interpolated climate surface downloaded from are trajectories adopted by the Intergovernmental Panel on
WorldClim v. 2.1 were used (Fick and Hijmans, 2017). Nine of Climate Change (IPCC), which provide a broader view of a
19 variables were removed from the analyses due to their high “business as usual” world without a climate policy, with global
correlation with other variables as indicated by Pearson’s warming in 2100 ranging from a low of 3.1°C to a high of 5.1°C
correlation coefficient (Table 1; Supplementary Data Sheet S3) above pre-industrial levels (O’Neill et al., 2014).
computed using SDMtoolbox 2.3 for ArcGIS (Kremen et al., In all the analyses, the maximum number of iterations was set
2008; Brown, 2014). Because some previous studies (Barve et al., to 10000 and that of convergence threshold to 0.00001. The
2011) suggest that modeling based on data for a restricted area is neutral (= 1) regularization multiplier value and auto features
more reliable than calculating habitat suitability at a global scale, were used. All samples were added to the background. The
the area included in the analysis was restricted to 84.65–34.43˚N “random seed” option, which provided a random test
and 74.65˚W–45.43˚E. Since this study investigated the effect of partition and background subset for each run, was applied,
climate change on the distribution of the species and soil and 20% of the samples were used as test points. The run was
characteristics have little effect on models of Australian performed as a bootstrap with 100 replicates. The output was set
terrestrial orchid, Leporella fimbriata (Kolanowska et al., to logistic. In addition, the “fade by clamping” function in
2021a), we did not use these variables in the analyses. MaxEnt was enabled to preclude extrapolations outside the
Predictions of the future extent of the climatic niches of P. environmental range of the training data (Phillips et al.,
albida and its pollinator in 2080–2100 were made using climate 2006). All analyses of GIS data were carried out on ArcGIS
projections developed by the CNRM/CERFACS modeling group 10.6 (Esri, Redlands, CA, United States). The evaluation of the
for the coupled model intercomparison project (CNRM–CM6-1) models was conducted using the area under the curve (AUC;
for four shared socio-economic pathways (SSPs; O’Neill et al., Mason and Graham 2002; Evangelista et al., 2008) and True Skill
2014): SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5. The layers in Statistic (TSS; Allouche et al., 2006).
Frontiers in Environmental Science 04 frontiersin.org
85
Kolanowska et al. 10.3389/fenvs.2022.912428
FIGURE 2
PCA ordination diagram (principal component analysis) of the distributions of populations of P. albida ssp. straminea (red dots), P. albida
ssp. albida (black dots), and P. albida ssp. tricuspis (blue dots) based on 19 bioclimatic variables.
SDMtoolbox 2.3 for ArcGIS (Kremen et al., 2008; Brown, ssp. straminea. Our analyses indicate significant differences
2014) was used to visualize changes in the distribution of in the bioclimatic preferences of the subspecies. of P. albida.
suitable niches of the orchid studied and its pollinator due Along the gradient represented by the first axis, P. albida
to global warming. To compare the prediction of the model of ssp. straminea is correlated especially with precipitation in
the current distribution with future predictions, all SDMs were the warmest quarter (bio18) and the mean temperature in
converted into binary rasters and projected using the Goode the wettest quarter (bio8). The ordination diagrams of PCA
homolosine. The presence threshold was estimated based on the explained 68.96% of the total variance. The first component
values for grids in which the species studied were predicted to accounted for 51.77% of the total variance and the second for
occur using present-time data. Because about 70%–84% of 17.19% (Figure 2; Supplementary Table S2). Based on the
known localities of P. albida and its pollinators were located morphological similarities of the two latter orchids, a
in grids with values > 0.4, this threshold value was used to create broader concept of P. albida ssp. albida was used, which also
binary rasters. To determine the availability of pollinators for includes P. albida ssp. tricuspis.
the orchid, the overlap of the binary models of both organisms
was calculated.
3.2 Model evaluation and limiting factors
3 Results The models had high AUC (0.871–0.998) and TSS
(0.517–0.9924) scores, indicating their predictions are very
3.1 Ecological differences between reliable (Figure 3; Table 2). The most important variable
subspecies of Pseudorchis limiting the distribution of P. albida ssp. albida was
precipitation in the warmest quarter (bio18—47.5%). Much
The result of PCA analyses indicate that although the less significant for its occurrence were the annual precipitation
preferred niche of P. albida ssp. straminea differs from that (bio12—18.5%) and the annual mean temperature
of the two other taxa, P. albida ssp. albida and P. albida (bio1—14.9%). The latter factor was crucial (42.6%) for the
ssp. tricuspis occupy similar habitats. This is indicated by the distribution of P. albida ssp. straminea, followed by the mean
second axis, which separated P. albida ssp. albida and P. albida temperature in the wettest quarter (bio8—29.8%) and
ssp. tricuspis from most of the records of P. albida precipitation in the warmest quarter (bio18—7.9%).
Frontiers in Environmental Science 05 frontiersin.org
86
Kolanowska et al. 10.3389/fenvs.2022.912428
FIGURE 3
Current distribution of suitable niches for P. albida ssp. albida (A) and P. albida ssp. straminea (B) along with the localities included in the models
(marked by black dots).
3.3 Effect of climate change on P. albida its range; however, it could potentially extend its range to
and its pollinators Svalbard (only in the less severe scenarios, such as SSP1-
2.6 and SSP2-4.5, is its occurrence in Iceland not completely
The predictions of the present-time models are congruent threatened). Within continental Europe, Pseudorchis albida ssp.
with the known geographical ranges of P. albida ssp. albida and albida will lose 44% (SSP1-2.6)–99% (SSP5-8.5) of its suitable
P. albida ssp. straminea (Figure 3). Our analyses indicate the niches, and P. albida ssp. straminea will lose 46% (SSP1-2.6)–
critical changes in the distribution of small-white orchid (Figures 91% (SSP5-8.5) of its current habitat.
4–7). All models predict that the availability of suitable niches for While in the future P. albida is predicted to occupy different
the orchids studied will increase as a result of global warming areas, the situation is completely different for the pollinators of
(Table 3), but the significant increase in suitable niches is this species (Table 3). All models predict a significant loss of
expected to occur in Greenland, whereas habitat loss in habitat for them, which in the case of Udea uliginosalis could
continental Europe will be severe. Overall, the potential range result in its extinction (Table 3).
of P. albida ssp. albida will be 27%–88% greater than at present,
whereas that of P. albida ssp. straminea will be 88%–156%
greater. The unexpected result is that while SSP1-2.6 is 3.4 Availability of pollinators
expected to be the most advantageous climate change scenario
for the latter taxon, the same scenario is the least optimistic for P. Based on the analyses, Udea uliginosalis is currently present in
albida ssp. albida, which will mostly benefit from SSP5-8.5. ca. 10%of the potential range ofP. albida ssp. straminea, but will not
Pseudorchis albida ssp. albida is currently known to occur be present there by 2100 (Supplementary Data Sheet S4; Table 4).
only in continental Europe, but apparently its suitable habitats Plutella xylostella is predicted to be the most important
will be located mainly in Greenland in the future and will become pollinator of P. albida, with a range overlap of 75% (SSP5-8.5)–
extinct in continental Europe based on SSP5-8.5. P. albida ssp. 88% (SSP3-7.0) with P. albida ssp. albida and 70% (SSP2-4.5)–
straminea will also face significant loss of habitats in this part of 100% (SSP1-2.6) with P. albida ssp. straminea. Chrysoteuchia
Frontiers in Environmental Science 06 frontiersin.org
87
Kolanowska et al. 10.3389/fenvs.2022.912428
TABLE 2 TSS scores, average training AUC, and standard deviations (in culmella is currently present in 74% of the range of P. albida ssp.
brackets) for the replicate runs of the models. albida and 52% of that of P. albida ssp. straminea. The predicted
Species Scenario TSS AUC future distribution of this insect will overlap partially with both
subspecies of the small-white orchid, overlapping 66% (SSP1-2.6)–
P. albida ssp. albida Present 0.9388 0.994 (0.001)
91% (SSP3-7.0) of that of P. albida ssp. albida and 38% (SSP2-4.5)–
SSP1-2.6 0.9553 0.995 (0.001) 56% (SSP3-7.0) of that of P. albida ssp. straminea. The statistics for
SSP2-4.5 0.9410 0.994 (0.001) Crambus ericella and C. pascuella are similar (Table 4).
SSP3-7.0 0.9124 0.994 (0.001)
SSP5-8.5 0.9416 0.993 (0.001) 4 Discussion
P. albida ssp. straminea Present 0.9158 0.973 (0.001)
4.1 Implication for taxonomy
SSP1-2.6 0.9172 0.973 (0.001)
SSP2-4.5 0.9196 0.974 (0.001) The recognition of three taxa within the Pseudorchis albida
SSP3-7.0 0.9217 0.973 (0.001) group remains a topic of taxonomic discussion and concern in
terms of both their distinction and rank. This study indicates that
SSP5-8.5 0.9220 0.973 (0.001)
ssp. tricuspis occupies niches similar to those occupied by ssp. albida,
Chrysoteuchia culmella Present 0.6317 0.884 (0.002) even if ssp. tricuspis is considered to be an alpine taxon and ssp. albida
SSP1-2.6 0.6333 0.888 (0.002) associated with lowland to subalpine regions (Reinhammar et al.,
2002; Jersáková et al., 2011). On the other hand, Klein (2000) argues
SSP2-4.5 0.6412 0.884 (0.002)
that ssp. tricuspis should be considered to be a separate subspecies,
SSP3-7.0 0.6520 0.887 (0.002) and this concept is also accepted by other scientists (Moore, 1980;
SSP5-8.5 0.6341 0.885 (0.002) Reinhammar, 1998; Bournérias and Prat, 2005; Perazza, 2016).
Reinhammar (1995), Reinhammar (1998) based on the results of
Crambus ericella Present 0.7740 0.957 (0.003) amultivariatemorphometric study considering plants of ssp. tricuspis
SSP1-2.6 0.7740 0.960 (0.003) as conspecific with P. straminea. The position of “tricuspis” as a
SSP2-4.5 0.7740 0.957 (0.003) variety is proposed by Kreutz (2004), Delforge (2006), and Jersáková
et al. (2011). Landwehr (1977) believes that this taxon is just a form of
SSP3-7.0 0.7740 0.958 (0.004) P. albida. Unfortunately, no molecular studies have included
SSP5-8.5 0.7740 0.958 (0.003) ssp. tricuspis. The results presented indicate that their
morphological characteristics are very similar, which supports
Crambus pascuella Present 0.7208 0.921 (0.003)
merging them under ssp. albida.
SSP1-2.6 0.7112 0.922 (0.003) Unlike Pseudorchis albida ssp. tricuspis, ssp. straminea is
SSP2-4.5 0.6886 0.919 (0.002) more distinct. Only the rank of this taxon is debated. Analyses
presented in this paper reveal differences in the climatic
SSP3-7.0 0.7096 0.922 (0.003)
requirements of ssp. albida and ssp. straminea, which could
SSP5-8.5 0.7208 0.920 (0.002) be a potential argument and area for research on whether to
Plutella xylostella Present 0.5356 0.872 (0.003) elevate the latter taxon to a separate species. This is proposed
based on its morphology (Reinhammar 1995; Reinhammar 1998)
SSP1-2.6 0.5323 0.877 (0.003)
and differences in allozymes (Reinhammar and Hedren, 1998).
SSP2-4.5 0.5281 0.871 (0.003) According to Duffy et al. (2011) the AFLP markers for P. albida
SSP3-7.0 0.5170 0.875 (0.003) are very polymorphic, and there are significant differences both
within and among populations, and population genetic isolation
SSP5-8.5 0.5271 0.872 (0.003)
increases with distance but did not find any differences in plastid
Udea uliginosalis Present 0.9800 0.997 (0.001) microsatellites between Irish populations of ssp. albida and
SSP1-2.6 0.9425 0.997 (0.001) Swedish ssp. straminea. Based on molecular studies, Bateman
et al. (2003); Bateman et al. (2017) show that the differences in
SSP2-4.5 0.9843 0.997 (0.001)
DNA sequences (nrITS, rbcL, and trnL-F) of the two taxa are near
SSP3-7.0 0.9806 0.997 (0.001) the lowest level of acceptance for their being separate species.
SSP5-8.5 0.9924 0.998 (0.001) Bateman et al. (2017) also reported at least 14 morphometric
characters that can be used to identify these taxa. Based on
Frontiers in Environmental Science 07 frontiersin.org
88
Kolanowska et al. 10.3389/fenvs.2022.912428
FIGURE 4
Future distribution of suitable niches for P. albida ssp. albida predicted under SSP1-2.6 (A), SSP2-4.5 (B), SSP3-7.0 (C), and SSP5-8.5 (D) climates.
FIGURE 5
Future distribution of suitable niches for P. albida ssp. straminea predicted under SSP1-2.6 (A), SSP2-4.5 (B), SSP3-7.0 (C), and SSP5-8.5 (D)
climates.
previous studies and the results presented, it is proposed that case scenario (SSP1-2.6) both subspecies, ssp. albida and
“straminea” is a subspecies. ssp. straminea, will lose almost half of their current
suitable niches (44% and 46%, respectively). In the most
damaging SSP5-8.5, only 1%–9% of the currently available
4.2 Effect of global warming on habitats will still be suitable for small-white orchids.
occurrence of P. albida s.l. and its Global warming is one of the most important causes of
conservation changes in habitat (Opdam and Wascher, 2004; Troia et al.,
2019). This is particularly so for alpine species, the available
The effect of predicted climate change will adversely affect habitat for which is likely to significantly decrease (Freeman
populations of P. albida in continental Europe. In the best- et al., 2018; Lamprecht et al., 2018) and other species with
Frontiers in Environmental Science 08 frontiersin.org
89
Kolanowska et al. 10.3389/fenvs.2022.912428
FIGURE 6
Changes in the distribution of suitable niches for P. albida ssp. albida predicted under SSP1-2.6 (A), SSP2-4.5 (B), SSP3-7.0 (C), and SSP5-8.5 (D)
climates.
FIGURE 7
Changes in the distribution of suitable niches for P. albida ssp. straminea predicted under SSP1-2.6 (A), SSP2-4.5 (B), SSP3-7.0 (C), and SSP5-
8.5 (D) climates.
very specific ecological requirements (Tsiftsis et al., 2019). Greenland. Shifts in the ranges of species may enable
Geppert et al. (2020) indicated that ranges of some alpine them to access and colonize these areas (Kelly and
orchids are or will decrease, especially since they are also Goulden, 2008; Cannone and Pignatti, 2014; Geppert et al.,
threatened by other factors, i.e., habitat modification and 2020). However, as the populations of P. albida are usually
loss of specific ecological relationships. Similar results are very small (Jeřábková, 2006; Pearman et al., 2008; Jersáková
reported in a study on another orchid with a Scandinavian- et al., 2011), it is unlikely that ssp. albida will be able
alpine distribution in Europe, Nigritella nigra s.l. (Kolanowska to colonize and adapt to new habitats in Greenland in the
et al., 2021b). However, global warming will result in next few decades. These should be accessible for ssp.
the transformation of currently unsuitable habitats in Straminea, which is more likely to be able to colonize this
Frontiers in Environmental Science 09 frontiersin.org
90
Kolanowska et al. 10.3389/fenvs.2022.912428
TABLE 3 Changes in the coverage of suitable niches [km2] for P. albida and its pollinators.
Species Scenario Range expansion No change Range contraction Change
P. albida ssp. albida SSP1-2.6 66505.58 31020.44 45720.92 +27.08%
SSP2-4.5 96261.33 11635.83 65105.53 +40.60%
SSP3-7.0 104480.4 1910.023 74831.34 +38.64%
SSP5-8.5 144083.0 409.486 76331.88 +88.29%
P. albida ssp. straminea SSP1-2.6 508313.6 104807.2 134675.1 +156.02%
SSP2-4.5 485026.9 55570.46 183911.8 +125.74%
SSP3-7.0 451762.1 30816.04 208666.3 +101.51%
SSP5-8.5 436690.5 15737.52 223744.8 +88.92%
Chrysoteuchia culmella SSP1-2.6 329513.2 1025813 575486.4 −15.36%
SSP2-4.5 299677.4 903473 697826.3 −24.86%
SSP3-7.0 312333.2 771202.9 830096.4 −32.33%
SSP5-8.5 314846.2 616610.7 984688.7 −41.83%
Crambus ericella SSP1-2.6 423770.1 331805.3 239292.9 +32.30%
SSP2-4.5 362709.5 241766.3 329332 +5.84%
SSP3-7.0 348517.7 139041.3 432056.9 −14.63%
SSP5-8.5 247644.7 76063.9 495034.4 −43.32%
Crambus pascuella SSP1-2.6 239193.1 808203.6 341319.9 −8.88%
SSP2-4.5 345567.2 764014 385509.5 −3.47%
SSP3-7.0 398812 678053.4 471470.1 −6.32%
SSP5-8.5 442073.3 590718 558805.5 −10.15%
Plutella xylostella SSP1-2.6 977527.1 1318081 586601.3 +20.52%
SSP2-4.5 871332.8 1163715 740967.3 +6.84%
SSP3-7.0 748247.1 1067649 837033.1 −4.66%
SSP5-8.5 759111.1 1047278 857403.5 −5.16%
Udea uliginosalis SSP1-2.6 286.4351 8737.296 18757.06 −67.18%
SSP2-4.5 8.887008 1552.492 25941.86 −94.32%
SSP3-7.0 4.101696 2.734464 27491.62 −99.98%
SSP5-8.5 0 0 27494.35 −100.00%
area. That distributions of orchids can change as a result island’s coastal zone. Of course, this has implications for the future,
of global warming is unlikely, but is suggested in some when the area of the GrIS is expected to decrease significantly and
previous studies (van der Meer et al., 2016; Kolanowska thus there will be new areas for colonization by plants (Chambers
et al., 2017). et al., 2022; Greve and Chambers 2022; Yang et al., 2022).
An important aspect of the occurrence of Pseudorchis in While a similar decline in the availability of a pollinator
Greenland is that currently most of the island is covered by ice previously predicted for the Australian orchid Leporella fimbriata
(GrIS). Studies indicate that by 2100, the thickness of the GrIS will (Kolanowska et al., 2021a) is unlikely to affect P. albida,
decrease significantly, but the area occupied will not differ much changes in climate will probably not limit the long-term
(Muntjewerf et al., 2020;Greve andChambers 2022; Yang et al., 2022). survival of this species. According to data available in GBIF
This means that many areas predicted suitable by the models will still (Table 5), at the beginning of the flowering season (June–August)
be inaccessible to Pseudorchis, and its occurrence will be limited to the of both subspecies of Pseudorchis, their pollinators are active and
Frontiers in Environmental Science 10 frontiersin.org
91
Kolanowska et al. 10.3389/fenvs.2022.912428
TABLE 4 Overlap of potential ranges of P. albida and its pollinators.
Scenario C. culmella (%) C. ericella (%) C. pascuella (%) P. xylostella (%) U. uliginosalis (%)
P. albida ssp. albida Present 73.76 76.50 72.95 74.40 0.00
SSP1-2.6 65.52 72.26 83.42 80.38 0.00
SSP2-4.5 74.02 73.33 88.10 86.92 0.00
SSP3-7.0 90.63 77.19 89.52 88.10 0.00
SSP5-8.5 80.43 83.18 89.55 74.81 0.00
P. albida ssp. straminea Present 51.91 57.94 57.34 46.11 10.98
SSP1-2.6 44.55 61.87 47.92 100.00 0.00
SSP2-4.5 37.60 55.49 39.39 69.60 0.00
SSP3-7.0 56.03 64.79 48.07 74.74 0.00
SSP5-8.5 52.12 67.04 59.89 72.34 0.00
TABLE 5 Overlap of potential ranges of P. albida and its pollinators.
Species Month
I II III IV V VI VII VIII IX X XI XII
P. albida ssp. albida x x x x x
P. albida ssp. straminea x x x x x
Chrysoteuchia culmella x x x x
Crambus ericella x x x
Crambus pascuella x x x x x x
Plutella xylostella x x x x x x x x
Udea uliginosalis x x x
can transfer pollen. For September and October, there are no As the predicted changes in the ranges of the taxa studied
reports of Crambus ericella and Udea uliginosalis, so late- differ, their future need of conservation is also likely to
flowering populations are unlikely to reproduce. The effect of differ. Pseudorchis albida ssp. straminea is not threatened
climate change on the flowering time of orchids and activity of in the near future by changes in climate, whereas
their pollinators is poorly known; however, previous studies populations of P. albida ssp. albida are, especially in Central
indicate that global warming can lead to desynchronization and Eastern Europe. Nevertheless, Pfeifer et al. (2010) indicated
and decline in the fruiting process of plants (Robbirt et al., that relict areas are likely to occur in which this taxon
2014; Hutchings et al., 2018). Similar findings are reported by can survive much longer than in new areas, which could be
Tsiftsis and Djordjević (2020) for two deceptive species of the affected by various non-climate related factors. It is, therefore,
genusOphrys, and they highlight a disruption of plant–pollinator best to maintain current populations in the best possible
interactions due to climate change, resulting in serious condition. Reinhammar et al. (2002) studied the population
conservation consequences for these species. On the other dynamics of P. albida over 6 years in two permanent plots (3 ×
hand, Molnár et al. (2012) reported that the phenology of 3 m), one mown and the other left to succession revealed that
nectar-rewarding orchids or short-lived species with non- in the mown plot, the number of new individuals appearing
Mediterranean distributions is less affected by global warming annually was large and stable, whereas in the unmanaged plot,
than that of autogamous or deceptive, long-lived species with there was little or no recruitment. It is, therefore, important to
mainly Mediterranean distributions. Pseudorchis albida belongs maintain the stability of semi-natural habitats inhabited by P.
to the first group of species. albida.
Frontiers in Environmental Science 11 frontiersin.org
92
Kolanowska et al. 10.3389/fenvs.2022.912428
Data availability statement Publisher’s note
The original contributions presented in the study are All claims expressed in this article are solely those of
included in the article/Supplementary Material; further the authors and do not necessarily represent those of
inquiries can be directed to the corresponding author. their affiliated organizations, or those of the publisher,
the editors, and the reviewers. Any product that may be
evaluated in this article, or claim that may be made by
Author contributions its manufacturer, is not guaranteed or endorsed by the
publisher.
MK designed the research and collected data. AR performed
statistical analyses. MK, AR, and SN defined the methodology
and conducted the research, prepared figures, and wrote and Supplementary material
reviewed the manuscript.
The Supplementary Material for this article can be found
online at: https://www.frontiersin.org/articles/10.3389/fenvs.2022.
Acknowledgments 912428/full#supplementary-material
SUPPLEMENTARY DATA SHEET S1
We are grateful to Marco Klüber, Walter Ezell on behalf of Localities of P. albida used in analyses.
James Fowler and Amadej Trnkoczy for giving us permission to
SUPPLEMENTARY DATA SHEET S2
use their photographs in this paper. We are very thankful to the Localities of P. albida pollinators used in analyses.
prof. Anthony Dixon for all his suggestions and language
SUPPLEMENTARY DATA SHEET S3
corrections. We thank the reviewers for all their valuable Correlations between bioclimatic variables calculated using Pearson’s
comments on the first version of the text. correlation coefficient.
SUPPLEMENTARY DATA SHEET S4
Overlap of suitable niches for the orchids studied and their pollinators
Conflict of interest currently and in various climate change scenarios.
SUPPLEMENTARY TABLE S1
The authors declare that the research was conducted in the List of GBIF Occurrence Download used in the study.
absence of any commercial or financial relationships that could
SUPPLEMENTARY TABLE S2
be construed as a potential conflict of interest. Factor coordinates of the cases based on correlation.
References
Allouche, O., Tsoar, A., and Kadmon, R. (2006). Assessing the accuracy of species Cannone, N., and Pignatti, S. (2014). Ecological responses of plant species and
distribution models: Prevalence, kappa and the true skill statistic (TSS). J. Appl. Ecol. communities to climate warming: Upward shift or range filling processes? Clim.
43 (6), 1223–1232. doi:10.1111/j.1365-2664.2006.01214.x Change 123 (2), 201–214. doi:10.1007/s10584-014-1065-8
Artsdatabanken (2010). Red List Database (Informasjon om rødlistede arter er nå i Chambers, C., Greve, R., Obase, T., Saito, F., and Abe-Ouchi, A. (2022). Mass loss
Artsportalen). Trondheim: Artsdatabanken. of the Antarctic ice sheet until the year 3000 under a sustained late-21st-century
climate. J. Glaciol. 68 (269), 605–617. doi:10.1017/jog.2021.124
Barve, N., Barve, V., Jimenez-Valverde, A., Lira-Noriega, A., Maher, S. P.,
Peterson, A. T., et al. (2011). The crucial role of the accessible area in ecological Cheffings, C., and Farrell, L. (2005). The vascular plant red data list for Great
niche modeling and species distribution modeling. Ecol. Modell. 222, 1810–1819. Britain. Species Status 7. Peterborough: Joint Nature Conservation Committee.
doi:10.1016/j.ecolmodel.2011.02.011
Claessens, J., and Kleynen, J. (2011). The flower of the European orchid. Form and
Bateman, R. M., Hollingsworth, P. M., Preston, J., Yi-Bo, L. U. O., Pridgeon, A. function. Netherlands: Jean Claessens and Jacques Kleynen Publishers.
M., and Chase, M. W. (2003). Molecular phylogenetics and evolution of orchidinae
Curtis, T. G. F., and McGough, H. N. (1988). The Irish red data book. 1. Vascular
and selected habenariinae (orchidaceae). Bot. J. Linn. Soc. 142 (1), 1–40. doi:10.
plants. Dublin: The Stationery Office.
1046/j.1095-8339.2003.00157.x
Damgaard, C., Moeslund, J. E., and Wind, P. (2020). Changes in the
Bateman, R. M., Rudall, P. J., and Denholm, I. (2017). Morphometric
abundance of Danish orchids over the past 30 years. Diversity 12 (6), 244.
comparison of British Pseudorchis albida with Icelandic P. straminea
doi:10.3390/d12060244
(orchidaceae: Orchidinae). New J. Bot. 7 (2-3), 78–93. doi:10.1080/
20423489.2017.1408191 Danihelka, J., Chrtek, J., and Kaplan, Z. (2012). Checklist of vascular plants of the
Czech Republic. Preslia 84, 647–811.
Bilz, M., Kell, S. P., Maxted, N., and Lansdown, R. V. (2011). European red
list of vascular plants. Luxembourg: Publications Office of the European Delforge, P. (2006). Orchids of Europe, north africa and the Middle East. London:
Union. A. & C. Black.
Bournérias, M., and Prat, D. (Editors) (2005). Les orchidées de France, Belgique et Duffy, K. J., Fay, M. F., Smith, R. J., and Stout, J. C. (2011). Population genetics
Luxembourg. 2nd ed. (Biotope: Mezé). and conservation of the small white orchid, Pseudorchis albida, in Ireland. Biol.
Environ. 111B (2), 73–81. doi:10.1353/bae.2011.0004
Brown, J. L. (2014). SDMtoolbox: A python-based GIS toolkit for landscape
genetic, biogeographic, and species distribution model analyses.Methods Ecol. Evol. Duffy, K. J., Kingston, N. E., Sayers, B. A., Roberts, D. L., and Stout, J. C.
5 (7), 694–700. doi:10.1111/2041-210x.12200 (2009). Inferring national and regional declines of rare orchid species with
Frontiers in Environmental Science 12 frontiersin.org
93
Kolanowska et al. 10.3389/fenvs.2022.912428
probabilistic models. Conserv. Biol. 23 (1), 184–195. doi:10.1111/j.1523-1739. Kreutz, C. A. J. (2004). Kompendium der Europäischen orchideen. Landgraaf:
2008.01064.x Kreutz Publishers.
Elith, J., Phillips, S. J., Hastie, T., Dudík, M., Chee, Y. E., and Yates, C. J. (2011). A Kricsfalusy, V., Budnikov, G., and Lesio, I. (2010). Rare and protected plant
statistical explanation of MaxEnt for ecologists. Divers. Distrib. 17, 43–57. doi:10. species of the uzhansky national nature park (transcarpathia, Ukraine), taiszia -.
1111/j.1472-4642.2010.00725.x J. Bot. Košice 20, 115–125.
Evangelista, P. H., Kumar, S., Stohlgren, T. J., Jarnevich, C. S., Crall, A. W., Kricsfalusy, V., Budnikov, G., and Mihaly, A. (1999). Red list of transcarpathia:
Norman, J. B., III, et al. (2008). Modelling invasion for a habitat generalist and a Threatened plant species and plant communities. Uzhgorod: Zakarpattya
specialist plant species. Divers. Distrib. 14, 808–817. doi:10.1111/j.1472–4642.2008. Publishing.
00486.x
Lamprecht, A., Semenchuk, P. R., Steinbauer, K., Winkler, M., and Pauli, H.
Fick, S. E., and Hijmans, R. J. (2017). WorldClim 2: New 1km spatial resolution (2018). Climate change leads to accelerated transformation of high-elevation
climate surfaces for global land areas. Int. J. Climatol. 37 (12), 4302–4315. doi:10. vegetation in the central Alps. New Phytol. 220, 447–459. doi:10.1111/nph.15290
1002/joc.5086
Landwehr, J. (1977). Wilde orchideeën van Europa. Amsterdam: Landwehr
Foley, M., and Clarke, S. (2005). Orchids of the British isles. Cheltenham: Griffin Publishers.
Press.
Ludwig, G., and Schnittler, M. (1996). Red list of threatened plants in Germany
Forbes, R. S., and Northridge, R. H. (2012). The flora of county fermanagh. (rote liste gefährdeter pflanzen deutschlands). Bonn: Bundesamt für Naturschutz.
Holywood: National Museums Northern Ireland.
Mason, S. J., and Graham, N. E. (2002). Areas beneath the relative operating
Freeman, B. G., Lee-Yaw, J. A., Sunday, J. M., and Hargreaves, A. L. (2018). characteristics (ROC) and relative operating levels (ROL) curves statistical
Expanding, shifting and shrinking: The impact of global warming on species’ significance and interpretation. Q. J. R. Meteorol. Soc. 128, 2145–2166. doi:10.
elevational distributions. Glob. Ecol. Biogeogr. 27 (11), 1268–1276. doi:10.1111/geb.
1256/003590002320603584
12774
Molnár, V. A., Tökölyi, J., Végvári, Z., Sramkó, G., Sulyok, J., and Barta, Z. (2012).
Gärdenfors, U. (2010). Rödlistade arter i sverige - the 2010 red list of Swedish
Pollination mode predicts phenological response to climate change in terrestrial
species. Uppsala: ArtDatabanken, SLU.
orchids: A case study from central Europe. J. Ecol. 100, 1141–1152. doi:10.1111/j.
GBIF (2020). GBIF.org (07 june 2020) GBIF occurrence Download. Copenhagen, 1365-2745.2012.02003.x
Denmark: GBIF.
Moore, D. M. (1980). in Pseudorchis. Flora europaea. Editors T. G. Tutin,
Geppert, C., Perazza, G., Wilson, R. J., Bertolli, A., Prosser, F., Melchiori, G., et al. V. H. Heywood, N. A. Burges, D. M. Moore, D. H. Valentine, S. M. Walters,
(2020). Consistent population declines but idiosyncratic range shifts in Alpine et al. (Cambridge, U.K.: Cambridge University Press), 332.
orchids under global change. Nat. Commun. 11 (1), 5835. doi:10.1038/s41467-020-
Muntjewerf, L., Petrini, M., Vizcaino, M., Ernani da Silva, C., Sellevold, R.,
19680-2
Scherrenberg, M. D. W., et al. (2020). Greenland Ice Sheet contribution to 21st
Greve, R., and Chambers, C. (2022). Mass loss of the Greenland ice sheet until the century sea level rise as simulated by the coupled CESM2.1-CISM2.1. Geophys. Res.
year 3000 under a sustained late-21st-century climate. J. Glaciol. 68 (269), 618–624. Lett. 47, e2019GL086836. doi:10.1029/2019GL086836
doi:10.1017/jog.2022.9
Naturva˚rdsverket (2022). Fridlysta orkidéer. Available at https://www.naturvardsverket.
Holland, J. P., Pollock, M. L., and Waterhouse, A. (2008). From over-grazing to se/en/topics/species-protection/protected-species/(Access March 02, 2022).
under-grazing: Are we going from one extreme to another? Aspects Appl. Biol. 85,
O’Neill, B. C., Kriegler, E., Riahi, K., Ebi, K. L., Hallegatte, S., Carter, T. R., et al.
25–30.
(2014). A new scenario framework for climate change research: The concept of
Holub, J., and Procházka, F. (2000). Red List of the Flora of the Czech Republic shared socioeconomic pathways. Clim. change 122 (3), 387–400. doi:10.1007/
(state in the year 2000. Červený seznam květeny České repub. 72, 187–230. s10584-013-0905-2
Hutchings, M. J., Robbirt, K. M., Roberts, D. L., and Davy, A. J. (2018). Opdam, P., andWascher, D. (2004). Climate change meets habitat fragmentation:
Vulnerability of a specialized pollination mechanism to climate change revealed Linking landscape and biogeographical scale levels in research and conservation.
by a 356-year analysis. Botanical J. Linn. Soc. 186 (4), 498–509. doi:10.1093/ Biol. Conserv. 117, 285–297. doi:10.1016/j.biocon.2003.12.008
botlinnean/box086
Pearman, D. A., Preston, C. D., Rothero, G. P., andWalker, K. J. (2008). The flora
Jeřábková, K. (2006). Ecological demands and optimal management of of rum: An atlantic island reserve. Cornwall, UK: Truro.
Pseudorchis albida. Master thesis in Czech. Prague: Czech University of Life Perazza, G. (2016). inOrchidee d’ Italia. Editors P. Grünanger and B. Barsella. 2nd
Sciences.
ed. (Milan: GIROS/Castello), 80–81.
Jersáková, J., Malinová, T., Jeřábková, K., and Dötterl, S. (2011). Biological flora of
Petrova, A., and Vladimirov, V. (2009). Red list of Bulgarian vascular plants.
the British isles: Pseudorchis albida (L.) Á. & D. Löve. J.f Ecol. 99 (5), 1282–1298. Phytol. Balc. 15 (1), 63–94.
doi:10.1111/j.1365-2745.2011.01868.x
Pfeifer, M., Passalacqua, N. G., Bartram, S., Schatz, B., Croce, A., Carey, P. D., et al.
Kaźmierczakowa, R., Bloch-Orłowska, J., Celka, Z., Cwener, A., Dajdok, Z.,
(2010). Conservation priorities differ at opposing species borders of a European
Michalska-Hejduk, D., et al. (2016). Polska czerwona lista paprotników i roślin orchid. Biol. Conserv. 143 (9), 2207–2220. doi:10.1016/j.biocon.2010.06.005
kwiatowych. Polish red list of pteridophytes and flowering plants. Kraków: Instytut
Ochrony Przyrody Polskiej Akademii Nauk. Phillips, S. J., Anderson, R., and Schapire, R. E. (2006). Maximum entropy
modeling of species geographic distributions. Ecol. Modell. 190, 231–259. doi:10.
Kelly, A. E., and Goulden, M. L. (2008). Rapid shifts in plant distribution with 1016/j.ecolmodel.2005.03.026
recent climate change. Proc. Natl. Acad. Sci. U. S. A. 105, 11823–11826. doi:10.1073/
pnas.0802891105 Phillips, S. J., Dudík, M., and Schapire, R. E. (2004). “A maximum entropy
approach to species distribution modeling,” in ICML ’04. Proceedings of the twenty-
Klein, E. (2000). Pseudorchis albida subsp. tricuspis (Beck) Klein stat. nov., eine
first international conference on Machine learning (New York: ACM).
weitgehend übersehene, calcicole, alpisch-boreale Sippe (Orchidaceae – orchideae).
Phyton 40, 141–159. Rankou, H. (2011). Pseudorchis albida. The IUCN Red List of Threatened Species
2011: e.T176007A7169757. Accessed on 02 March 2022.
Kolanowska, M., Kras, M., Lipińska, M., Mystkowska, K., Szlachetko, D. L., and
Naczk, A. M. (2017). Global warming not so harmful for all plants-response of Reinhammar, L. G. (1995). Evidence for two distinctive species of Pseudorchis
holomycotrophic orchid species for the future climate change. Sci. Rep. 7 (1), (orchidaceae) in Scandinavia. Nord. J. Bot. 15 (5), 469–481. doi:10.1111/j.1756-
12704–12713. doi:10.1038/s41598-017-13088-7 1051.1995.tb00180.x
Kolanowska, M., Michalska, E., and Konowalik, K. (2021a). The impact of global Reinhammar, L. G., and Hedren, M. (1998). Allozyme differentiation
warming on the niches and pollinator availability of sexually deceptive orchid with a between lowland and alpine populations of Pseudorchis albida s.
single pollen vector. Sci. Total Environ. 795, 148850. doi:10.1016/j.scitotenv.2021. lat.(Orchidaceae) in Sweden. Nordic J. Bot. 18 (1), 7–14. doi:10.1111/j.1756-
148850 1051.1998.tb01090.x
Kolanowska, M., Rewicz, A., and Nowak, S. (2021b). Significant habitat loss of the Reinhammar, L. G., Olsson, E. G., and Sormerland, E. (2002). Conservation
black vanilla orchid (Nigritella nigra sl, Orchidaceae) and shifts in its pollinators biology of an endangered grassland plant species, Pseudorchis albida, with some
availability as results of global warming.Glob. Ecol. Conservation 27, e01560. doi:10. references to the closely related alpine P. straminea (Orchideceae). Botanical J. Linn.
1016/j.gecco.2021.e01560 Soc. 139, 47–66. doi:10.1046/j.1095-8339.2002.00041.x
Kremen, C. A., Cameron, A., Moilanen, S. J., Phillips, C. D., Thomas, H., Beentje, Reinhammar, L. G. (1998). Systematics of Pseudorchis albida s.1. (Orchidaceae) in
J., et al. (2008). Aligning conservation priorities across taxa in Madagascar with Europe and north America. Botanical J. Linn. Soc. 126, 363–382. doi:10.1111/j.
high-resolution planning tools. Science 320, 222–226. doi:10.1126/science.1155193 1095-8339.1998.tb01388.x
Frontiers in Environmental Science 13 frontiersin.org
94
Kolanowska et al. 10.3389/fenvs.2022.912428
Robbirt, K. M., Roberts, D. L., Hutchings, M. J., and Davy, A. J. (2014). Potential pollinator interactions. Sci. Rep. 10, 10623. doi:10.1038/s41598-020-
disruption of pollination in a sexually deceptive orchid by climatic change. Curr. 67491-8
Biol. 24 (23), 2845–2849. doi:10.1016/j.cub.2014.10.033
Tsiftsis, S., Djordjević, V., and Tsiripidis, I. (2019). Neottia cordata (orchidaceae)
Sârbu, A., Janauer, G. A., Exler, N., Sarbu, I., and Anastasiu, P. (2020). The at its southernmost distribution border in Europe: Threat status and effectiveness of
potential sensitivity to climate change of selected endangered and important Natura natura 2000 network for its conservation. J. Nat. Conserv. 48, 27–35. doi:10.1016/j.
2000 Habitats and plants from Bucegi Natural Park, Romania. Not. Bot. Horti jnc.2019.01.006
Agrobot. Cluj. Napoca. 48 (1), 456–479. doi:10.15835/nbha48111756
Turis, P., Kliment, J., Feráková, V., Dítě, D., Eliáš, P., Hrivnák, R., et al. (2014).
StatSoft Inc (2011). STATISTICA (data analysis software system), version 13.1. Red List of vascular plants of the Carpathian part of Slovakia. Thaiszia J. Bot. 24 (1),
Available at www.statsoft.com. 35–87.
subordinate agency (2022). Norwegian biodiversity information centre. https:// van der Meer, S., Jacquemyn, H., Carey, P. D., and Jongejans, E. (2016). Recent
www.biodiversity.no/Pages/166308/Publications (Access March, 2022).02. range expansion of a terrestrial orchid corresponds with climate-driven variation
in its population dynamics. Oecologia 181, 435–448. doi:10.1007/s00442-016-
Summerhayes, V. S. (1951). Wild orchids of Britain. London, UK: Collins.
3592-7
Troia, M. J., Kaz, A. L., Niemeyer, J. C., and Giam, X. (2019). Species traits and
Yang, H., Krebs-Kanzow, U., Kleiner, T., Sidorenko, D., Rodehacke, C. B., Shi, X.,
reduced habitat suitability limit efficacy of climate change refugia in streams. Nat.
et al. (2022). Impact of paleoclimate on present and future evolution of the
Ecol. Evol. 3, 1321–1330. doi:10.1038/s41559-019-0970-7
Greenland Ice Sheet. PLoS ONE 17 (1), e0259816. doi:10.1371/journal.pone.
Tsiftsis, S., and Antonopoulos, Z. (2011). Pseudorchis albida: An enigmatic orchid 0259816
for the Greek flora. J. Eur. Orch. 43 (4), 795–806.
Zulka, K. P., Eder, E., Höttinger, H., and Weigand, E. (2001). Grundlagen zur
Tsiftsis, S., and Djordjević, V. (2020). Modelling sexually deceptive orchid Fortschreibung der Roten Listen gefährdeter Tiere Österreichs. Wien:
species distributions under future climates: The importance of plant- Umweltbundesamt.
Frontiers in Environmental Science 14 frontiersin.org
95
fevo-10-1058334 January 6, 2023 Time: 11:34 # 1
TYPE Original Research
PUBLISHED 06 January 2023
DOI 10.3389/fevo.2022.1058334
Taxonomic revision of Sobralia
OPEN ACCESS section Racemosae Brieger
EDITED BY
Rusea Go, (Sobralieae, Orchidaceae)
Universiti Putra Malaysia, Malaysia
REVIEWED BY
1 1
Andrey Erst, Przemyslaw Baranow *, Dariusz Szlachetko and
Central Siberian Botanical Garden Pavel Kindlmann2
(RAS), Russia
Edlley Max Pessoa, 1Department of Plant Taxonomy and Nature Conservation, Faculty of Biology, University of Gdańsk,
Federal University of Mato Grosso do Gdańsk, Poland, 2Faculty of Science, Institute for Environmental Studies, Charles University, Prague,
Sul, Brazil Czechia
*CORRESPONDENCE
Przemyslaw Baranow
przemyslaw.baranow@ug.edu.pl
SPECIALTY SECTION Sobralia Ruiz & Pav. is a large and morphologically diverse neotropical orchid
This article was submitted to
Conservation and Restoration Ecology, genus. It can be divided into four sections and some informal groups of
a section of the journal species based mainly on the inflorescence architecture. While most of the
Frontiers in Ecology and Evolution
species have strongly abbreviated, compact raceme, the section Racemosae is
RECEIVED 30 September 2022
ACCEPTED 23 November 2022 characterized by an elongated inflorescence with distinct internodes between
PUBLISHED 06 January 2023 flowers. Although the group is well-defined and easily distinguishable in terms
CITATION of morphology, its species are often similar to each other and may be easily
Baranow P, Szlachetko D and misidentified. Identification is especially difficult when considering herbarium
Kindlmann P (2023) Taxonomic
revision of Sobralia section specimens. Here, a taxonomic revision of Sobralia section Racemosae is
Racemosae Brieger (Sobralieae, presented. Apart from particular species’ morphological characteristics, keys
Orchidaceae).
Front. Ecol. Evol. 10:1058334. for identification, ecological data, and distribution maps are presented.
doi: 10.3389/fevo.2022.1058334 Sobralia gambitana is described as a species new to science. A neotype for
COPYRIGHT S. hoppii Schltr. is proposed.
© 2023 Baranow, Szlachetko and
Kindlmann. This is an open-access
article distributed under the terms of KEYWORDS
the Creative Commons Attribution diversity, morphology, neotropics, new species, taxonomy
License (CC BY). The use, distribution
or reproduction in other forums is
permitted, provided the original
author(s) and the copyright owner(s)
are credited and that the original
publication in this journal is cited, in 1. Introduction
accordance with accepted academic
practice. No use, distribution or
reproduction is permitted which does Sobralia is a large orchid genus consisting of about 200 species distributed from
not comply with these terms. southern Mexico to Brazil and Bolivia. Its representatives can be found in various
habitats, from humid and shaded tropical forests to sunny, dry, open savannas or
roadsides. They grow from sea level to over 3,000 m a.s.l. They can occur as terrestrial
or lithophytic plants, but sometimes also as epiphytes (Pridgeon et al., 2006; Baranow,
2015).
Sobralia is a morphologically diverse group of species, especially when considering
the architecture of inflorescence and morphology of floral bracts and flower segments.
The differences allow distinguishing some groups of species, which served as the basis for
the description of infrageneric units (Lindley, 1854; Reichenbach, 1873; Brieger, 1983).
Frontiers in Ecology and Evolution 01 frontiersin.org
96
fevo-10-1058334 January 6, 2023 Time: 11:34 # 2
Baranow et al. 10.3389/fevo.2022.1058334
The nominal section was characterized by lateral or rarely herbarium material resulted in a discovery of the collection,
terminal inflorescences with branching, well-developed raceme which, in order of its distinctness, was recognized as a species
and relatively small floral bracts compared to the size of the new to science. Additionally, a neotype for S. hoppii Schltr.
ovary (Brieger, 1983). The section Racemosae Brieger, despite is selected. In addition, the first comprehensive key for the
terminal inflorescences, could be distinguished from the former identification of the species of section Racemosae is provided.
by its elongated and unbranched inflorescences with large
floral bracts. Section Globosae Brieger is composed of small
plants with narrow leaf blades, small flowers positioned in the 2. Materials and methods
terminal, and condensed inflorescences (shortened internodes
hidden under the floral bracts) that successively produce a single The presented revision was based on the morphological
flower at a time and elongate with successively produced floral study of the herbarium material deposited in the following
bracts. Species of section Abbreviatae Brieger share terminal herbaria AMES, BM, COAH, COL, CUVC!, F, K, K-L, MO, NY,
and condensed inflorescences with the previous section but, P, UGDA-DLSz, U, US, W, W-R (Thiers, 2022). In total, over 440
instead, present floral bracts forming a cone. The fifth section, herbarium specimens were examined within the study.
Intermediae Brieger, was established for a single taxon Sobralia Apart from the morphological data, the herbarium
fragrans Lindl. to emphasize its elongated basal internode of specimens were also a source of information concerning the
the inflorescences. Dressler (2002) enlarged this section, placing ecology of the studied species given under the morphological
other species with small flowers and inflorescences. descriptions. Moreover, the localities of the collections were
The present classification of Sobralia is based on Briegers’ used for the distribution presentation, and the geographical
1983 division of the genus into sections. However, the distribution maps were generated using the software QGIS
1 2
development of molecular methods revealed that the nominal version 3.22.12 and the Natural Earth data.
section of Sobralia is more closely related to other genera A conservation analysis was performed using the criteria
of Sobralieae than to the remaining groups of Sobralia. As from the International Union for the Conservation of Nature
the nominal section is also different in the morphological (IUCN, 2022). The Extent of Occurrence (EOO) and the Area of
characters, such as branching and often lateral inflorescence, it Occupancy (AOO) of each species were estimated using GeoCat
was elevated to the rank of a separate genus Brasolia (Baranow (Bachman et al., 2011).
et al., 2017, see also Dressler et al., 2011; Neubig et al., 2011).
Since then, the newly defined Sobralia consists of the species 3. Results
with terminal and unbranching inflorescences only. Most of the
species have abbreviated and compact raceme, hidden between
the floral bracts, forming a tight, cone-like structure, producing 3.1. Sobralia Ruiz & Pav. section
one or two flowers at a time. However, there is one group, Racemosae Brieger
section Racemosae Brieger, with elongated raceme, having
distinct internodes. The flowers of its representatives develop Orchideen 1 (13): 798. 1983; Type species: Sobralia rosea
from the nodes and are supported by distichous, large floral Poepp. & Endl., Nov. Gen. Sp. Pl. 1: 54, t. 93. 1836.
bracts. The inflorescence contains several flowers at various The group contains 15 species occurring in South America
stages of growth, with the youngest ones on its top. The distinct with the greatest species diversity in Northern Andes.
morphology is supported by the results of the molecular study,
3.1.1. Key to the species
which can be seen in the phylogenetic trees (Neubig et al., 2011;
Baranow et al., 2017). Also, the karyotype evolution analysis 1. Leaves less than 5 cm wide . . . 2
with the phylogenetic study as the background (Baranow et al., 2. Flowers deep rose–purple with bright yellow throat of
the lip, apical stelidia of gynostemium not exceeding anther
2022) as well as niche conservatism and ecological tolerance
apex . . . 1. S. paradisiaca
evolution study (Kolanowska et al., 2022) have confirmed the
2∗ Flowers yellow or white with yellow lip disk, apical
consistency of the group. Thus, the section appears to be well-
stelidia of gynostemium long, strongly exceeding anther
defined and distinct from the other groups of the genus. On the
apex . . . 3
other hand, the species of Racemosae are in many cases similar
3. Flowers yellow, stelidia rounded at apex . . . 2.
to each other and easy to misidentify. The only study devoted
S. chrysantha
to Racemosae was made by Romero-González (2003), but the
3∗ Flowers white or navy yellow with yellow or orange lip
author focused only on S. liliastrum Lindl. and its close allies.
disk or a dot on the apical part, stelidia acute at apex . . . 4
The study aims to present the results of the taxonomic
revision of all species of the section Racemosae with the
descriptions and illustrations of their morphology, with the 1 www.qgis.org/pl/
ecological data and maps of distribution. The revision of the 2 www.naturalearthdata.com
Frontiers in Ecology and Evolution 02 frontiersin.org
97
fevo-10-1058334 January 6, 2023 Time: 11:34 # 3
Baranow et al. 10.3389/fevo.2022.1058334
4. Lip divided into basal and apical parts by the distinct 3.1.1.1. Sobralia paradisiaca Rchb.f.
constriction just below the middle, gynostemium stelidia, Linnaea 22: 816. 1850. Type (designated by Baranow in
horn-like, falcate . . . 3. S. chrysoleuca Szlachetko et al., 2020: 248): Venezuela. Merida. Sine prec
4∗ Lip not constricted in the middle, gynostemium stelidia loc. Alt. 1600 m. March 1847. N. Funk and L.J. Schlim 1489
narrowly oblong 5 (Lectotype: W!, Isolectotypes: K!, P!).—Garay and Dunsterville.
5. Leaves ca 4 cm wide . . . 6. S. liliastrum Venezuelan Orchids Illustrated 404. 1959.—Szlachetko et al.
5∗ Leaves up to 2.5 cm wide . . . 6 Materials to the Orchid Flora of Colombia 3: 248. 2020.
6. Lip white with yellow throat and reddish orange elevated Plants up to 130 cm tall, caespitose, often leafy for all except
keels, floral segments 50–65 mm long . . . 4. S. elisabethae the basal quarter. Stem concealed in green tubular sheathing
6∗ Lip hyaline white with pale yellow, elevated keels, floral leaf bases which tend to become red or dark red when well
segments 40–45 mm long . . . 5. S. granitica exposed. Leaves up to 25 cm long and 4.5 cm wide, lanceolate,
1∗ Leaves 6–12 cm wide . . . 7 apex lightly attenuate, plicate, the sides of blades tend to be
7. Rachis fractiflex, bracts horizontally spreading, acute or revolute, making the upper surface convex, the uppermost leaves
obtuse . . . 8 smaller than the ones below, with spathe-like base subtending
8. Flowers white, lip red–purple on the lamina and the the rachis. Inflorescence producing 3–6 flowers developing in
throat, floral segments 70–75 mm long, lip furnished succession from 1 to 3 at a time; rachis terete, fractiflex. Sepals
with a pair of shallow ridges in the throat only . . . 7. and petals deep rose–purple paler right at base, lip deep rose–
S. luerorum purple with bright yellow throat. Dorsal sepal up to 65 mm
8∗ Flowers creamy white with purple striation on lip, floral long and 25 mm wide, oblanceolate to ligulate-oblanceolate,
segments up to 60 mm long, lip with two basal ridges and acute, moderately fleshy. Lateral sepals up to 70 mm long and
5–7 parallel lamellae running from the base to the apex . . . 33 mm wide, ligulate-lanceolate, somewhat oblique, moderately
8. S. gloriosa fleshy. Petals up to 70 mm long and 30 mm wide, elliptic-
7∗ Rachis sinuously flexuous, bracts suberectly spreading,
oblanceolate, acute, somewhat oblique. Lip 48–70 mm long,
acuminate . . . 9
33–50 mm wide when spread, elliptic-rectangular in general
9. Floral bracts leaf-like, up to 20 cm long, decreasing in size
outline, entire, apical margins truncate, strongly undulate and
toward the apex of inflorescence . . . 9. S. ruckeri
crispate, thin for the most part but axially much thickened at
9∗ Floral bracts up to 12 cm . . . 10
base where there are two ventral swellings about 10 mm long,
10. Floral segments not exceeding 90 mm in length, lip with
two basal ridges, additional lamella can be present too, but projecting from each side and almost touching each other, the
it is restricted to the middle of the lip only 11 rest of the axial part not thickened but with several raised veins. . .
11. Two basal thickenings fused together except their giving the impression of a thickening terminating in a small
margins 10. S. gambitana hollow point. Gynostemium up to ca 35 mm long, stelidia short,. . .
11∗ Two basal thickenings separate 12 obscure, subequal in length to the anther or shorter (Figure 1).. . .
12. Lip without any protuberances apart from the basal Ecology: Terrestrial. Flowering in March,
lamellae . . . 11. S. tamboana September, and December.
12∗. Lip with thickenings or lamellae running along one or Distribution: Colombia, Venezuela. Alt. 1600–2300 m.
more central veins . . . 13 Conservation status: EOO—CR, AOO—CR.
13. Lip base with 2 lamellae running to its middle and the Representative specimens (Supplementary Map 1)—
central vein in central part ornamented with lamella . . . 12. Venezuela. Merida. Between La Carbonera and La Azulita.
S. splendida 17 September 1966. J. de Bruijn 1134 (MO!); Sine loc. Alt.
13∗ Lip with 2 basal keels, median vein thickened, with two 2300 m. H. Wagener 124 (W! 21607, UGDA-DLSz!–drawing).
additional thickenings near the middle . . . 13. S. hoppii Colombia. Norte de Santander. Ocaña. Alt. 1830 m. 1846. L.J.
10∗ Floral segments 100 mm or more, lip disk with 3–7 Schlim 1203 (W-R!). Vaupés. Entre Wacaricuara y El Varador.
lamellae running from the base up to at least its middle . . . Al Río Yi. 9–12 December 1952, R. Romero Castañeda 3922
14 (COL!).
14. Lip white with a broad white margin, with purple Sobralia paradisiaca belongs to the group of species having
veins in the center, disk with 3 lamellae running from the relatively narrow leaves (up to 5 cm width) along with
base to the middle, the median-one high-carinate . . . 14. S. chrysantha, S. liliastrum, S. chrysoleuca, S. elisabethae, and
S. pulcherrima S. granitica. It can be easily separated from all of them by the
14∗ Lip dark purple–magenta with very narrow, white color of the flowers—it is the only taxon having deep rose–
margin, in center with fine, radiating, white veins, disk from purple tepals with a bright yellow throat of the lip. The species
the base to center transversed by 5–7 low, parallel lamellae differs from other S. liliastrum-complex representatives also by
. . . 15. S. rosea very short, rounded stelidia of gynostemium.
Frontiers in Ecology and Evolution 03 frontiersin.org
98
fevo-10-1058334 January 6, 2023 Time: 11:34 # 4
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 1
Sobralia paradisiaca Rchb. f. (A,B) apical part of stem with the inflorescence, (C) dorsal sepal, (D) petal, (E) lateral sepal, (F) lip, and (G)
gynostemium [redrawn by A. Król from Dunsterville and Garay (1959)].
In our opinion, there is a mistake in the protologue ISOTYPE: W!).—Szlachetko et al. Materials to the Orchid Flora
of the species. In W and K there are Funck and Schlim of Colombia 3: 249. 2020.
collections numbered 1489, and not 1749 as stated in the original Plants height unknown, probably well over 100 cm tall.
description. Apart from fragments of plants, the collection Leaves up to 25 cm long and 4 cm wide, lanceolate to elliptic-
includes also a hand drawing of a plant and floral parts (W-R lanceolate, acute, plicate. The leaf subtending the rachis up to
21609). It appears that collection 1489 should be indicated as 10 cm long. Inflorescence ca 10 cm long, rachis inconspicuously
the type specimen. flexuose. Floral bracts 15–50 mm long, narrowly lanceolate-
Dunsterville and Garay (1959) stated that S. paradisiaca may triangular, acute to acuminate. Flowers yellow, large. Dorsal
be only a juvenile form of S. liliastrum and treated as a synonym sepal 83 mm long, 12 mm wide, oblong-ligulate to linear,
of the latter species. Surprisingly, in the same publication, the subobtuse. Lateral sepals 83 mm long, 12 mm wide, obliquely
same authors listed S. paradisiaca as a valid species emphasizing linear-lanceolate, shortly acuminate. Petals 85 mm long, 13 mm
its distinctness observed during the study of the type specimen. wide, obliquely linear-lanceolate to ligulate-lanceolate, shortly
acuminate. Lip 70 mm long, up to 49 mm wide, broadly
3.1.1.2. Sobralia chrysantha Lindl. obovate to suborbicular-obovate in outline above cuneate
Fol. Orchid. 5 (Sobralia): 3. 1854. Type: Colombia. base, rounded at apex, indistinctly denticulate and undulate
(Santander). Socorro. Alt. 1220 m. L.J. Schlim 6 (Holotype: K-L!, along margins in the upper half, attenuate and canaliculate
Frontiers in Ecology and Evolution 04 frontiersin.org
99
fevo-10-1058334 January 6, 2023 Time: 11:34 # 5
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 2
Sobralia chrysantha Lindl. (A) apical part of stem with inflorescence, (B) dorsal sepal, (C) petal, (D) lateral sepal, (E) lip, (F) apical part of
gynostemium, side view, and (G) apical part of gynostemium, front view [drawn by P. Baranow; panels (A,E–G) from the isotype, lip, and
gynostemium redrawn from the drawings left on the herbarium sheet; panels (B–D) from Schlim 1188].
toward base, without thickenings except the still middle rib. Erect plant, height unknown, probably well above 100 cm
Gynostemium 52 mm long, clavate, staminodes oblong elliptic, tall. Leaves 30 cm long and 4 cm wide, oblong lanceolate
straight, much exceeding the anther, with a deep wing at to linear-lanceolate, acuminate, coriaceous, strongly plicate.
their back and an oblique emargination, apex falcate, blunt Inflorescence 12 cm long, ca. 15-flowered, rachis erect, nearly
(Figure 2). straight to somewhat flexuose. Floral bracts 25–30 mm long,
Ecology: Terrestrial. No data on flowering time. triangular-lanceolate, acuminate. Ovary 30 mm long. Flowers
Distribution: Colombia. Alt. 1220–2000 m. white or light yellow with distinct, deep yellow or orange dot
Conservation status: EOO—CR, AOO—CR. on the apical part of lip. Dorsal sepal 68 mm long, 18 mm
Representative specimens (Supplementary Map 2)– wide, lanceolate, acute. Lateral sepals 60 mm long, 15 mm wide,
Colombia. Santander. Socorro. Alt. 1300–2000 m. 1849. L.J. oblong-lanceolate, inconspicuously oblique, acute. Petals 57 mm
Schlim 1188 (W!, UGDA-DLSz!–drawing); Socorro. Alt. 1220 m. long, 25 mm wide, widely oblong or elliptic, somewhat oblique,
L.J. Schlim 6 (K-L!). acute. Lip 60 mm long, 45 mm wide, oblong, constricted below
It is interesting to note that Reichenbach’s drawing the middle and inconspicuously bilobed at the apex, margins in
accompanying the type specimen stored at W shows a very apical part irregularly crenate and crispate, disk with nine keels
massive stelidia which are apically bilobed, with the anterior running along the central veins from base almost to the apex,
lobe being somewhat longer and acute, and the posterior one base papillate. Gynostemium 36 mm long, slender but with large
shorter and rounded. In the materials examined we did not find and wide, massive, wing-like, triangular, oblique, acute apical
stelidia of this form. stelidia, which distinctly exceeding the anther apex (Figure 3).
Sobralia chrysantha resembles S. liliastrum-complex in habit Ecology: Terrestrial.
and with a very long stelidia much exceeding the anther apex but Distribution. Bolivia.
can be easily distinguished by the color of the flowers (yellow vs. Representative specimens–BOLIVIA. Sine loc. S.A. Pearce
white in S. liliastrum and its allies) and rounded apex of stelidia 777 (W!, UGDA-DLSz!–drawing).
(vs. acute in S. liliastrum complex). Unique characters of this species are lip constricted near
the middle, not found anywhere in the section Racemosae, and
3.1.1.3. Sobralia chrysoleuca Rchb. f. massive, horn-like, falcate stelidia.
Xenia Orchid. 2: 179. 1873. Type: BOLIVIA. Sine loc. S.A. According to the note on the herbarium label of the type
Pearce 777 (Holotype: W! 21594, UGDA-DLSz!–drawing). specimen, the flowers of the species may be white and yellow
Frontiers in Ecology and Evolution 05 frontiersin.org
100
fevo-10-1058334 January 6, 2023 Time: 11:34 # 6
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 3
Sobralia chrysoleuca Rchb. f. (A) apical part of stem with inflorescence, (B) dorsal sepal, (C) petal, (D) lateral sepal, (E) lip, and (F) gynostemium
(drawn by P. Baranow from the holotype).
colored. It may explain why the author compared the taxon light green. Leaves up to 22 cm long and 4 cm wide, lanceolate to
with Sobralia aurantiaca (a synonym of S. infundibuligera with narrowly lanceolate, long acuminate, prominently veined on the
a compact inflorescence, hidden between the bracts, not similar underside, plicate. Inflorescence terminal, elongate, racemose,
to those of the section Racemosae representatives)—the taxa are fractiflex, laxly few-flowered. Flowers produced in succession,
similar in order to the flower color. large, white, lip with yellow throat. Dorsal sepal 58–70 mm long,
11.5–15 mm wide, oblong oblanceolate, acuminate, somewhat
3.1.1.4. Sobralia liliastrum Lindl. fleshy, and thick. Lateral sepals 58–70 mm long, 11.5–15 mm
Gen. Sp. Orchid. Pl.: 177. 1833. Type (designated wide, oblong oblanceolate, acuminate, subfalcate, somewhat
by Baranow and Szlachetko, 2016: 339): Brazil. Bahia. fleshy and thick, lightly carinate dorsally. Petals 55–70 mm
P. Salzmann s.n. (Lectotype: K! 000293880–plant on the long, 16–21 mm wide, oblong elliptic, subobtuse to subacute,
right side of herbarium sheet; Isolectotypes: K!, MO!, W-R!, slightly falcate, thin, finely sulcate dorsally. Lip 60–62 mm long,
NY!—photograph, UGDA-DLSz!–drawing). ≡ Cattleya 43–54 mm wide, suborbicular-subflabellate in general outline,
liliastrum (Lindl.) Beer, Prakt. Stud. Orchid.: 212. 1854.— widest above the middle, obscurely 3-lobed, truncate at the
Garay and Dunsterville. Venezuelan Orchids Illustrated 322. apex, the apical margin erose, soft, thin, more or less undulate,
1959.–Baranow and Szlachetko, Pl Syst Evol. 302: 338. 2016.— sometimes with obscurely keeled lateral veins, often papillate
Szlachetko et al. Materials to the Orchid Flora of Colombia 3: at the base. Gynostemium 48 mm long with apical lanceolate-
250. 2020. subfalcate, acute stelidia much exceeding the anther apex.
= Sobralia liliastrum var. alba Lindl., Fol. Orchid. 5 Ecology: Terrestrial along lowland rivers, in savannas, and
(Sobralia): 4. 1854. Type: not designated on steep embankments with subxerophytic plants. Flowering
= Sobralia liliastrum var. rosea Lindl., Fol. Orchid. 5 throughout the year (Figure 4).
(Sobralia): 4. 1854. Type: not designated. Distribution: Venezuela, Guyana, Suriname, French Guiana,
= Sobralia liliastrum f. maior Hoehne, Relat. Commiss. Brazil. Alt. up to 2255 m.
Linhas Telegr. Estratég. Matto Grosso Amazonas 5, Bot. 4: 23, Conservation status: EOO—LC, AOO—EN.
pl. 74. 1912; Type: not designated. Representative specimens (Supplementary Map 3)—
Plants up to 300 cm tall, caespitose, erect, terete, the base Colombia. Amazonas. Araracuara. Sabana de la Angostura. Alt.
with the remains of sheaths, the apex leafy, perfectly smooth, 400 m. 21 December 1951. H. Garcia Barriga and R.E. Schultes
Frontiers in Ecology and Evolution 06 frontiersin.org
101
fevo-10-1058334 January 6, 2023 Time: 11:34 # 7
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 4
Sobralia liliastrum Lindl. (A) apical part of stem with inflorescence, (B) dorsal sepal, (C) petal, (D) lateral sepal, (E) lip, and (F) gynostemium, side
view (drawn by P. Baranow from the isolectotype kept at MO).
14143 (COL!); Río Caquetá. La Pedrera. Cerro de Cupati. Rangel, P. Palacios, and J. Betancur (COL!). Chocó. Alrededores
Alt. 240–580 m. 30 September 1952. H. Garcia Barriga 14529 de Coredo. 22 October 1946. R. Romero Castañeda 519 (COL!).
(COL!); Corregimiento La Pedrera, comunidad Bocas del Pira, Guainia. Poblacion el Remanes. Cerros de Mavicure y Pajarito
sabana “Vasewai,” margen derecha del Río Apaporis, approx. a orillas del Río Inírida, 40 km de Puerto Inírida. 1978.
10 min en bote de la comunidad Bocas del Pira Río arriba, F. Sarmiento 1084 (COL!); Correg. de San Felipe, Río Negro.
resguardo Yaigoje-Apaporis. 0◦27′05′′S, 70◦14′40′′. Alt. 240 m. Alrededores de la pista de aterrizaje. Alt. 100 m. 28 September
31 March 2009. J. Betancur, D. Cardenas, D. Tanimuca, and 1977. M. Pabon E., J. Espina, and C. Dominguez 228 (COL!);
E. Tanimulka 13995 (COL!); Araracuara. Río Caquetá. 1 April Caserio de Sta Rita, Río Guainia. Alt. 100 m. 15 October 1977.
1976. C. Sastre and H. Reichel D. 5182 (COL!). Caquetá. Sierra M. Pabon E., J. Espina, and C. Dominguez 337 (COL!). Guaviare.
de Chiribiqueta, Campamento Sur. Al. SW del Campamento, Mesa La Lindosa, Cerrito a 15–20 km al S de San José del
entre este y los primeros de la meseta. 0◦55′N, 72◦45′W. Alt. Guaviare. Alt. 400–600 m. 13–15 December 1950. J.M. Idrobo
350–400 m. 7 July 1990. P. Franco, J. Estrada, J. Fuertes and P. and R.E. Schultes 656 (COL!); Mpio. San José del Guaviare.
Palacios 3237 (COL!, US!); Sierra de Chiribiquete. Mesa encima Carretera de San José a Puerto Arturo, km 3, alrededores de
de la Cueva de las Pinturas, 1◦05′N, 72◦40′W. Alt. 740–760 m. la finca Santa Gertrudis, 2◦28′20′′N, 72◦41′30′′W, Alt. 280 m.
21 August 1992. P. Palacios 2417 and P. Franco, O. Rangel, and J. 21 January 1996. R. Lopez and O.J. Rodriguez 976 (COAH!,
Betancur (COL!–sterile); Sierra de Chiribiquete. Campamento MO!); San José del Guaviare. Antiqua represa. Alt. 200–250 m.
Norte. Prox. del campamento. 1◦7′N, 72◦50′W. Matorrales 27 December 1993. C. Sastre and J.P. Robin 9194 (COL!); San
de sabana. 6 December 1990. J.M. Cardiel, S. Castroviejo, G. José del Guaviare. Ciudad Perdida o Ciudad de Piedra. Alt.
Galeano and F. Gonzalez 1010 (COL!); Sierra de Chiribiquete. 250–300 m. 28 December 1993. C. Sastre and J.P. Robin 9218
En la via del Campamento a la Cueva de Pinturas. 1◦05′N, (COL!). Meta. Serrania de La Macarena, margen izquierda del
72◦40′W. Alt. 600 m. 17 August 1992. P. Franco 3718 and O. Río Guayabero, a 10 km abajo de Caño Lozada. Alt. 500 m. 16
Frontiers in Ecology and Evolution 07 frontiersin.org
102
fevo-10-1058334 January 6, 2023 Time: 11:34 # 8
Baranow et al. 10.3389/fevo.2022.1058334
January 1959. P. Pinto E., H. Bischler, and R. Jaramillo M. 206 Graminosa arbustiva en altiplanicie (Cerro Paru), 4◦34′N,
(COL!, P!); Reserva Nacional de la Macarena, southernmost 65◦31′W. Alt. 590 m. February 1992. A. Chaviel 205 (MO!);
slope of Macarena Mts, immediate to the Río Guayabero. Alt. Dpto Atures, Serrania de la Coromoto, Sector “El Tobagin,” a
250–300 m. 25 January 1968. J. Thomas, J. Hernandez C., and 37 km al. S de Pto. Ayacucho. 5◦24′N, 67◦35′N. Alt. 80–200 m.
P. Pinto E. 1589 (P!). Vaupés. Río Macu-Parana, tributary of the 19 January 1989. N. Cuello 344 (MO!); Dpto Atabapo, Zona
Río Papuri. 8 August 1943. P.H. Allen 3047 (COL!); Yapoboda, de Lomerio con Sabana Arbustiva y Altiplanicie con Herbazal
10 December 1943. P.H. Allen 3224 (MO!); Bacuraba Cachoeira Subarbustivo Tepuyano. 3◦33′N, 64◦29′W. Alt. 1400 m.
(the first major cataract on the Vaupés East of Mitú). Alt. 200 m. November 1991. Y. Fernandez and M. Yanez 856 (MO!); Por
4 November 1944. P.H. Allen 3311 (MO!); Env. of Río Mitú, debajo del Salto Remo, 110–71 km por arriba del Guayapo.
dry arud slopes of the Cerro of Mitú, El Cerro de Guacamaya, 4◦34′N, 67◦18′W. Alt. 120 m. May 1989. E. Foldats and J.
30 October 1976. E.W. Davis 201 (COL!, U!); Río Pira Paraná Velazco 9462 (MO!); Dpto Atabapo, Alto Río Orinoco, 15 km al.
(tributary of Río Apaporis, between 0◦15′S, 70◦30′W and W de la Esmeralda, Cerro Baraco. 3◦8′N, 65◦41′W. Alt. 300 m.
0◦25′N, 70◦30′W, 6 September 1952. R.E. Schulters and I. 1 March 1990. G.G. Aymard and L. Delgado 8283 (MO!); 9 km
Cabrera 17232 (U!, US!, UGDA-DLSz!–copy); Yurupari, orilla northeast of San Carlos de Río Negro. 1◦57′N, 67◦3′W. Alt.
Vaupés, 350 km arriba de Mitú. Alt. 220 m. 24 September 120 m. 25 November 1977. R.L. Liesner 3582 (MO!); 10 km NE
1939. J. Cuatrecasas 6961 (COL!); Río Vaupés, cachivera de of San Carlos de Río Negro. 1◦54′N, 67◦00′W. Alt. 120 m. 28
Yurupari. Alt. 400 m. 24–26 October 1952. H. Garcia Barriga January 1980. R.L. Liesner 8830 (MO!); 2 km east of San Carlos
14935 (COL!); Río Vaupés, Mitú and vicinity. September– de Río Negro. 1◦55′N, 67◦5′W. Alt. 120 m. 13 November 1977.
October 1966. R. Schultes 24344 (COL!); Mitú and vicinity. Río R.L. Liesner 3421 (MO!); 10 km NE of San Carlos de Río Negro,
Parana-Pichuna, savanna at major rapids, 6 September 1976. (ca. 20 km S of confluence of Río Negro and Brazo Casiquiare),
J.L. Zarucchi 1957a (COL!); Desembocadura del Ariari con 1◦56′N, 67◦03′W. Alt. 120 m. 24 April 1979. R.L. Liesner 6947
el Río Guayabero. Cabana del Incora “Bocas del Ariari,” 21 (MO!); Atures, Río Coro-Coro, W of Serrania de Yataje, 6–8 km
February 1969. P. Pinto E. and C. Sastre 942 (COL!, P!); Vicnity N of settlement of Yutaje, 5◦41′00′′N, 66◦07′30′′W. Alt. 320 m.
of Mitú. Trail to Cerro Mitú. Caatinga forest. Alt. 200–250 m. 2 23 February 1987. R.L. Liesner and B. Holst 21326 (MO!);
October 1991. J. Kress, J. Betancur, C. Roesel, and R. Echeverry Dpto Atures, 1 to 2 km E of Río Coro-Coro, W of Serrania de
91-3336 (COL!); Río Vaupés, Cerro de Circasia, entre el Río Ti Yataje, 8 km N of settlement of Yutaje, 5◦41′30′′N, 66◦07′30′′W.
y Namu. Alt. 380–450 m. 30 October 1952. H. Garcia Barriga Alt. 600–650 m. 25 February 1987. R.L. Liesner and B. Holst
15028 (COL!); Río Kubiyu, Cerro de Canenda. Alt. 380–680 m. 21383 (MO!); “El Tobogan de la Selva,” 35 km south of Puerto
2–4 November 1952. H. Garcia Barriga 15074 (COL!); Caño Ayacucho. Alt. 85 m. 21 February 1979. T. Plowman 7702 (F!);
Cubiyú. Comunidad Indigena La Sabana. 1◦15′N, 70◦51′W. Caño Cupaven, Río Orinoco at mouth of Río Atabapo. Alt.
Alt. 200 m. 26 April 1993. S. Mandrinan, G. Ngan, and J. 150 m. 11 May 1954. J. Silverio Level 82 (F!, MO!); Camino San
Page 1175 (COL!, NY!); Riberas del Río Inírida (69◦45′W), Carlos de Río Negro-Solano, 10–22 February 1989. B. Stergios,
sitio Raudal Alto o Mariapiri, margen derecha. Alt. 180 m. 3 K. Kubitzki, G. Aymard, and E. Melguiero 13396 (MO!, US!);
February 1953. A. Fernandez 2121 (COL!); Cerro Mitú. Alt. Río Negro, Piedra Ignea, Cerro Aratityope, 2◦10′N, 65◦34′W,
400–450 m. 4 September 1959. B. Maguire, C.K. Maguire, and A. approx. 70 km al SSW de Ocamo, con richuelos afluente al
Fernandez 44097 (COL!); Río Kuduyari. Yapoboda, sandstone Río Manipitare. Alt. 990–1670 m. 24–28 February 1984. J.A.
savanna near headwaters. 5 October 1951. R. E. Schultes and Steyermark, P. Berry, and F. Delascio 130051 (U!); Río Negro,
I. Cabrera 14243 (COL!); The same loc. 18 November 1952. piedra ignea, Cerro Aratitiope, approx. 70 km al SWW de
R.E. Schultes and I. Cabrera 18497 (COL!); Serrania de Taraira. Ocamo, 2◦10′N, 65◦34′W. Alt. 990–1670 m. 24–28 February
10 km al NW del raudal de la Libertad. 0◦53′, 69◦45′W. Bosque 1984. J.A. Steyermark, P. Berry, and F. Delascio 130051 (MO!);
de caatinga. Alrededores del campamento. Alt. 250 m. 31 Dept. Atabapo, Cerro Duida. 3◦40′N, 65◦45′W. Alt. 1400 m.
August 1993. J. Rodriguez 183 (COL!); Serrania de Taraira. 10 February 1982. J.A. Steyermark, M. Guariglia, N. Holmgren,
10 km al NE del raudal de la Libertad. 0◦58′S, 69◦45′W. Alt. J.L. Luteyn, and S. Mori 126433 (MO!, K!); Atabapo, sabanas
250 m. 2 August 1993. R. Cortes and J. Rodriguez 764 (COL!); y bisques ubicados al pie nor-oriental y oriental del Cerro
Cerro de Chiribiquete, a un lado del Río Macaya, terreno muy Cucurito, ribera SE del medio Caño Yagua. 3◦36′N, 66◦34′W.
pedregoso. 17 January 1944. G. Guiterrez and R.E. Schultes 683 Alt. 120 m. 8 December 1978. O. Huber and S.S. Tillett 2941
(NY!). Vichada. Parque Nacional Natural, “El Tupparo,” on (K!, U!); Bolivar. Roscio, 3 km S of El Pauji. 4◦30′N, 61◦35′W.
granitic outcrops between the mouth of the Río Tupparo to Summit of mountain bordering N side of “El Abismo,” thick
Raudal Maipures alon the Río Orinoco, 5◦12′N, 67◦50′W. Alt. low rocky scrub. Alt. 1050 m. 19 October 1985. B.K. Holst
90–130 m. 1 March 1985. J.L. Zarucchi and C.E. Barbosa 3521 and R.L. Liesner 2355 (MO!, U!); Río Negro, Slope of Cerro
(MO!). Venezuela. Amazonas. Río Sipapo entre Isla Lencho y Aracamuni. Aracamuni. Quebrade Camp, in area of rapids
Boca del Cuao. Mpio Autana, 4◦54′–5◦3′N, 67◦34′–67◦46′W. flowing over laja (stone), 1◦24′N 65◦38′W. Alt. 600 m. 20
28 January 1997. A. Castillo 4474 (MO!); Dpto Atabapo, Sabana October 1987. R.L. Liesner and F. Delascio 22240 (MO!, U!);
Frontiers in Ecology and Evolution 08 frontiersin.org
103
fevo-10-1058334 January 6, 2023 Time: 11:34 # 9
Baranow et al. 10.3389/fevo.2022.1058334
Transecta entre conucos al. E de Santa Rosa de Ucata, passando 1979. P.J.M. Maas and L.Y.T. Westra 4208 (U!); Pakaramia
por bosque humedo, hast arbustal de arena blanca al. E de este Mts., Mt. Aynatoi (sandstone). 5◦55′N 61◦W. Dry sandstone
pobiado, 4◦24′N, 67◦46′W. Alt. 80–85 m. 23 October 1989. G.A. rocks near falls. 16 October 1981. P.J.M. Maas et al. 5781
Romero and E. Melguiero 2235 (K!, MO!); Cerro granitico al. (COL!, MO!, U!); Kaieteur Plateau, 12 May 1944. B. Maguire
E del Raudal Gavilan, caminando ca 2 horas desde la parcel. and D.B. Fanshawe 23419 (NY!, U!); Fleuve Oyopack, Savane
5◦37′N, 67◦22′W. Alt. 100 m. 1 February 1991. G.A. Romero, roche, Roche Canari zozo, rive gauche. 8 July 1969, R.A.A.
E. Melgueiro, and C. Gomez 2291 (MO!); Laja granitica al. E Oldeman 332 (U!); French Guiana. Region de la Haute Crique
del Raudal Gavilancito, vegetation en pequenas depresiones Armantabo, bas Oyapock, 21 February 1981. J.J. de Granville
y grietas en la piedra, 5◦37′N, 67◦22′W. Alt. 80–100 m. 9 165 (U!). Brazil. Amazonas. Rio Tuari (afluente de Rio Negro),
February 1992. G.A. Romero, E. Melgueiro, and C. Gomez 2365 Lago Uirauacu (=Passaro Grande em Lingua Geral), 0◦20′N,
(MO!); Esmeralda Ridge, between Esmeralda and base of Cerro 67◦20′W. 13 November 1987. M.L. Kawasaki 144 (U!, US!);
Duida. Alt. 150 m. 21 August 1944. J.A. Steyermark 57744 Rio Uapes, Panure, catinga. 15 November 1947. J.M. Pires
(F!); Atabapo, Boca de Mesaque. 3◦04′N, 67◦06′W. Alt. 80 m. 1026 (COL!, US!). Bahia. Santa Cruz Cabralia, Mata costeira. 5
November 1989. J. Velazco 953 (MO!). Bolivar. Along highway November 1966. R.P. Balem and R.S. Pinheiro 2841 (F!); Marau,
between Santa Elena and Icabaru 103 km SW of Santa Elena, resting. 18 January 1967. R.P. Balem and R.S. Pinheiro 3180 (F!);
16 km NE of Icabaru, near bridge. 4◦20′N, 62◦45′W. Alt. 750 m. Una-Ilheus. 39◦02′W, 15◦07′S, Alt. 70 m. 25 December 1975.
24 July 1982. T.B. Croat 54045 (MO!); By main road, ca 11 kms P. Bamps 5053 (U!); Ba. Lancois. Rio Mueugezinho, Proximo a
E of Kavanayén. Alt. 1200 m. 26 July 1983. R. Kral Wit and BR-242. Em direcao a Serra Brajao. Alt. 1000 m. 20 December
A.C. Gonzalez 70462 (MO!); Gran Sabana, ca 15 km WSW 1984. A. Furlan et al. 37123 (K!). Km 10, Ponta-Olivacea road,
of Karaurin Tepui, Quebrada Tanuan. 5◦19′N, 61◦04′W. Alt. Mpio Ilheus, 14◦50′S, 39◦2′W, Alt. 30–50 m. 10 February 1985.
950 m. 1 May 1988. R.L. Liesner 24119 (MO!); 17 km E of El A. Gentry and E. Zardini 50008 (MO!); Coastal Zone, 16 km S
Pauji by road and 64 km W of Santa Elena by road, 4 km N of Cumuruxatiba, 39◦15′W, 17◦13′S. Alt. 0–50 m. 18 January
of highway. Río Las Ahallas, 4◦30′N, 61◦30′W. Alt. 850 m. 29 1977. R.M. Harley 18095 (K!, U!); Mato Grosso, margem direita
October 1985. R.L. Liesner 19122 (MO!); 3 km S of El Puaji, de R. Juruena, morrinio da cochoeira de S. Joao da Barra. 10
Morichal, 4◦30′N, 61◦35′W. Alt. 900 m. 19 October 1985. R.L. June 1977. N.A. Rosa and M.R. Santos 2081 (MO!, U!); Mun.
Liesner and B.K. Holst 18811 (MO!); 17 km E of El Pauji by Lencois, BR-242, 3–8 km W del desvio a Lencois. 12◦28′S,
road and 64 km W of Santa Elena by road, 4 km N of highway. 41◦22′W, Alt. 880 m. 26 November 1992. R. Mello-Silva and J.
Río Las Ahallas, 4◦30′N, 61◦30′W. Alt. 850 m. 1 November Vicente 5800 (K!, F!, MO!); Mun. Itabuna, 10 km S de Pontal
1985. R.L. Liesner 19311 (MO!); Sabana de Arekuna, E margin (Ilheus), camino a Olivenca, local de extraccion de arena,
of lower Río Caroni. 6◦31′N, 62◦53′W. Alt. 520 m. 29 August 14◦54′S, 39◦02′W. Alt. 50 m. 4 December 1992. R. Mello-Silva
1983. G.T. Prance and O. Huber 28316 (MO!); N de Raudalito, and J. Vicente 5583 (K!); Mpio de Castro Alves, Serra da Jiboia,
Río Sipapo. Alt. 120 m. 10 October 1988. G.A. Romero and F. 12◦51′11′′S, 39◦29′19′′W. 8 July 1992. L.P. de Quieroz, S. Mayo,
Guanchez 1631 (MO!); Km 146 al. sur de El Dorado. Alt. 1280 m. M. Nadruz, T.S.N. Sena, and M.L.S. Guedes 2946 (K!); Mun. de
15–18 November 1978. J.A. Steyermark, J.L. Luteyn and M.L. Una, Estrada Ilheus-Una, ±30 km au Sul de Olivenca, 15◦12′S,
Lebron-Luteyn 117553 (MO!); Gran Sabana, between Mission 39◦03′W. Alt. 40 m. 2 December 1981. G.P. Lewis and A.M. de
of Santa Teresita de Kavanayén northwest to Río Karuai, on Carvalho 722 (K!); Moun. De Ilheus, Estrada Olivenca, Villa
large mes. Alt. 1220 m. 26 October 1944. J.A. Steyermark 59387 Brasil, a 7 km de Olivenca. Restinga. 13 January 1981. A.M.
(F!); Sororopan tepui, crest of cerro between east and west end. Carvalho and J. Gatti 485 (K!); BA-Estrada Macuge-Andarai. 17
Alt. 2255 m. 14 November 1944. J.A. Steyermark 60117 (F!). December 1984. A.M. Giulietti et al. 36893 (K!); Mun. Lencois,
Guyana. Upper Mazurani River Region. Karowtipu Mountain. Trilha Lencois-Capao, 12◦33′34′′S, 41◦24′66′′W. Alt. 650 m.
5◦45′N 60◦35′W. Alt. 1000 m. 21 April 1987. B.M. Boom 28 January 1997. B. Stanard, S. Atkins, E. Saar, L. Passos, and
and D. Gopaul 7567 (MO!); Holitipu, trail betw. camp and M.L. Guedes 4581 (K!); Mun. Lencois, Morro da Chapadinha,
airstrip and surrounding area. 05◦59′N 61◦03′W. Alt. 1100 m. Chapadinha, divisa com Brejoes, 12◦27′00′′S, 41◦25′00′′W. Alt.
Tepui savanna and gallery forest. 7 February 1996. D.H. Clarke 750 m. 24 November 1994. E. Melo et al. 1328 (K!); Olivenca
1037 (NY!, U!); Paruima, 5 km N, Auratoi Savanna. 05◦51′N km 21 para a Faz. Ipiranga ao Norte. 10 October 1972. T.S.
61◦05′W. Alt. 760 m. 21 July 1997. D.H. Clarke et al. 6137 Santos 2456 (P!); Mun. Lencois, Chapadinha, Lencois, proximo
(U!); Cuyuni-Mazaruni Region. Pakaraima Mts., 12 m waterfall, ao Rio Mucugezinho, 12◦27′44′S, 41◦25′12′′W. Alt. 810 m. 27
large Partang River tributary, 12.7 km NE Imbaimadai. Scrub September 1994. G. Stam, A.M. Giulietti, and H.P. Bautista
forest merging with riparian gallery forest. 5◦48′N 60◦14′W. 922 (K!); Mun. Lencois, Serra da Chapadinha, 12◦27′41′′S,
Alt. 700 m, 25 May 1992. B. Hoffman 1868 and C.L. Kelloff, 41◦25′16′′W. Alt. 900 m. 05 January 1996. A.M. de Carvalho
G. Gharbarran, and S. Sprague (NY!, US!); Kaieteur savanna. et al. 2178 (K!). Para: Maraba, Alro de Serra, arredores do N5. 12
1936. G. Hollister s.n. (NY!); Pakaraima Mts. Mt. Latipu, top May 1982. A. Mesquita, R.B. Gilberto, and L. Marinho 116 (F!,
(Mazaruni R.), 5◦57′N 60◦38′W. Alt. 900 m. 10 November K!, MO!); Sete Varas airstrip on Rio Curua, 0◦95′S, 54◦92′W. 6
Frontiers in Ecology and Evolution 09 frontiersin.org
104
fevo-10-1058334 January 6, 2023 Time: 11:34 # 10
Baranow et al. 10.3389/fevo.2022.1058334
August 1981. J.J. Strudwick, G.L. Sobel et al. 4343 (K!); Maraba, apical part anyway. Gynostemium ca 30–40 mm long, apical
Serra de Carajas. 12 May 1983. N.L. Meneses s.n. (K!); Mpio wings 10–15 mm long, distinctly exceeding the column apex,
de Ameirim, reserva florestal da SEMA, 0-1◦S, 52–53◦W. 10 linear, slightly falcate, acute (Figure 5).
October 1987. A.S. Tayares 117553 (MO!). Rio de Janeiro. sine Ecology: Terrestrial or litophytic in savannas, among rocks,
loc. V. Soares 435 (K!). Roraima. Estrada Manaus-Caracarai xerophytic forests, and disturbed forests next to the roads.
km 130, campina das Pedras. 25 May 1974. W. Rodrigues, A. Flowering throughout the year.
Loureiro, and D. Coelho 9308 (MO!). BOLIVIA. Santa Cruz. Distribution: Venezuela, Ecuador, Colombia, Brazil, French
Vallegrande Prov., Corosito, 2 km al. S de los Sitanos. 18◦52′5′′S, Guiana, Guyana, Peru. Alt. up to 900 m.
64◦57′0. Alt. 1400 m. 2 September 1989. I.G. Vargas 286 (F!, Conservation status: EOO—LC, AOO—EN.
MO!). Representative specimens (Supplementary Map 4)—
Along with Sobralia elisabethae and S. granitica, it creates Colombia. Amazonas. Corregimiento departamental de la
a group of unique species characterized by narrow leaves and Pedrera. Margen izquierda del Río Caquetá, Cerro Yupati.
white flowers with various ornamentation on the lip disk. 1◦17′49′′S, 69◦37′03′′W. Alt. 200–400 m. 6 August 1997.
S. liliastrum is similar to S. granitica, but has larger D. Cardenas, C. Marin, R. Lopez, and N. Rodriguez 8528 and
flowers (58–62 mm long flower segments vs. 40–45 mm in 8563 (COAH!); Santa Isabel, sitio sabanas de Solarte. 1◦05′S,
S. granitica), the color of the lip keels (orange vs. light yellow 71◦10′W. 4–6 December 1996. M.V. Arbelaez, U. Matapi,
in S. granitica), the raised keels (only the central keel notably and N. Matapi 681 (COAH!); Araracuara. 3 March 1986.
raised in S. liliastrum vs. with two, subparallel keels at base, the P. Palacios and B. Plazas 1164 (COAH!). Caquetá. Araracuara.
disk with 9 erose-denticulate thickened keels in S. granitica), Orilla del Río Caquetá, balcon del Diablo. 0◦36′S 72◦24′W.
and the presence of pseudopollen on the S. liliastrum lip. The 19 November 1993. D. Cardenas, G. Gangi, and J. Manaidego
flower segments of S. elisabethae and S. liliastrum are similar in 4135 (COAH!); Parque Nacional Natural Chiribiquete. Río
size, but they differ in lip details. In the former species, the lip Cunare, Raudal del Tubo. 0◦26′N 72◦30.5′W. 3 February 1992.
is adorned with thickenings along veins running from a pair of N. Hernandez and N.C. Penuela CHI69 (COAH!); Solano,
keel-like, crenulate basal calli nearly to the apex, sometimes the margen izquierda del Río Caquetá, Sitio Paujil (Area del Caño
thickenings are not visible in the center of the lip but distinct in Paujil), 10 km al. NO de Araracuara. 0◦45′–0◦48′S, 72◦20′–
its apical part anyway. 72◦25′W. Alt. 100–350 m. 10 November 1992. V. Arbelaez and
V. Hernandez 326 (COAH!); Cabaceras del Río Mesay. 1–6 Mar.
3.1.1.5. Sobralia elisabethae R.H. Schomb. 1980. M.C. Pabón 971 (COAH!). Guainía. Trocha Nabuquen.
Verh. Vereins. Beförd. Gartenbaues Königl. Preuss. Staaten 2◦51′127′′N, 65◦38′339′′W. Alt. 500 m. 25 February 1995. M.P.
15: 137. 1841. Type (designated by Romero-González, 2003: Etter, A. Munoz, L. Baptiste, and A. Repizzo 508 (COAH!);
129): Venezuela. Bolivar. Vicinity of Mount Roraima, 1836, R.H. Inrida. Resquardo indigena Almidon-Ceiba, a 4 km NE de
Schomburgk 1059 (Lectotype: BM!, Isolectotypes: BM!, K!, P, W! la comunidad La Ceiba, camino a Cn Vitina. En bosquocito
7463).–Baranow and Szlachetko. The taxonomic revision of the xerofitico transitional entre el bosque de altura y la sabana,
Sobralia Ruiz & Pay. (Orchidaceae) in the Guyanas (Guyana, sobre superfi. 3◦39′20.3′′N, 67◦23′40.3′′W. Alt. 80–90 m.
Suriname, French Guiana). Pl Syst Evol. 302: 338. 2016.– 20 October 1998. E. Cordillo-R. et al. 372 (MO!). Guaviare.
Szlachetko et al. Materials to the Orchid Flora of Colombia 3: Mpio. San José del Guaviare. Serrania La Lindosa. Bosque
253. 2020. intervenido a orillas de la carretera. Alt. 220–250 m. 5 March
Plants 50–90 cm high, caespitose, erect, slender. Leaves 1994. D. Cardenas and G. Trujillo 4348 (COAH!); Mpio. San
numerous, up to 26 cm long and 2.5 cm wide, narrowly José del Guaviare. Carretera de San José a Puerto Arturo, km 3,
lanceolate, long-acuminate, suberect. Inflorescence 6–10 cm alrededores de la finca Santa Gertrudis, zona de afloramientos
long, terminal, laxly 5–8-flowered, rachis fractiflex. Flowers rocosos, 02◦28′20′′N, 72◦41′30′′W. Alt. 280 m. 21 January
opening successively, white, with yellow lip throat and keels. 1996. R. Lopez, D. Giraldo C., and H. Salgado 952 (COAH!);
Floral bracts 8–40 mm long, ovate-lanceolate. Pedicel and ovary Mpio. San José del Guaviare. En immediaciones de Ciudad de
34 mm long, slender. Dorsal sepal 50–60 mm long, 10–13 mm Piedra, Serrania La Lindosa, carretera San José-El Caprichio,
wide, narrowly lanceolate, acute to acuminate. Lateral sepals 02◦28′28′′N, 72◦41′48′′W. Alt. 290 m. 19 November 1995.
55–65 mm long, 14–16 mm wide, lanceolate, subfalcate, acute. R. Lopez, D. Giraldo C., and H. Salgado 829 (COAH!). Meta.
Petals 50–60 mm long, 10–13 mm wide, narrowly lanceolate, Mpio. La Macarena. Serrania de La Macarena, Caño Canoas,
acute, subfalcate. Lip 60 mm long, 35–40 mm wide, oblong cercanias a los chorros, formaciones de roca desnuda del Escudo
ovate in general outline, more or less notched at the apex, Guayanes. 2◦28′–29′N, 70◦44′W. Alt. 255–280 m. 31 December
crenulate and undulate along margins, especially in the apical 2005. J. Betancur, J. Aguirre, J. Contreras, and M. Rodriguez
half, with thickenings along veins running from a pair of keel- 11993 (COL!). Vaupés. Mitú & vicinity, along Río Vaupés
like, crenulate basal calli nearly to the apex, sometimes the between Río Ti and Rapids of Mandi, 23 September 1976. J.L.
thickenings not visible in the center of lip but distinct in its Zarucchi 2115 (K!); Mpio Mitú. Camino entre la comunidad
Frontiers in Ecology and Evolution 10 frontiersin.org
105
fevo-10-1058334 January 6, 2023 Time: 11:34 # 11
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 5
Sobralia elisabethae R. H. Schomb. (A) apical part of stem with inflorescence, (B) dorsal sepal, (C) petal, (D) lateral sepal, (E) lip, (F) gynostemium,
side view, and (G) apical part of gynostemium, front view [drawn by P. Baranow, panel (A) from lectotype, (B–G) redrawn from
Romero-González (2003)].
Mitú Cachivera y el cerro Guacamaya, 1◦11′40′′N, 70◦14′24′′W. Alt. 200–400 m. 16 May 2006. D. Cardenas, R. Pena, and
Alt. 180–370 m. 24 September 2007. D. Cardenas, Z. Cordero, A. Rivera 18723 (COAH!); Corregimiento departamental
N. Salinas, and A. Zuluaga 21087 (COAH!); Mpio. Mitú. de Yavarate, comunidad de Bogotá-Cachivera, camina a
Comunidad de Monford, via Monford-Mitú km 4. Sabaneta Acaricuara. 0◦49′45.3′′N, 70◦03′50.6◦W. N. Castano, N. Salinas,
varillal a catinga de 8–10 m de altura, 0◦37′17′′N, 69◦44′56.4′′W. A. Zuluaga, and W. Estrada 2737 (COAH!). Venezuela.
Alt. 160–170 m. 30 September 2007. D. Cardenas, Z. Cordero, Amazonas. Atabapo, Cerro Huachamacari, E slope. 3◦49′N,
N. Salinas, and A. Zuluaga 21334 (COAH!); Mpio. Mitú. 65◦42′W. Alt. 600–700 m. 3 November 1988. R.L. Liesner 25736
Sector compredito entre el cerro Guacamaya y Caño Sangre. (U!); Santa Lucia, Pedra de Cucui. 28 October 1967. Farney
1◦12′N, 70◦ 14′W. Alt. 200–300 m. June 2008. N.R. Salinas et al. 1822 (K!); Base occidental del Cerro Yapacana, 3◦38′N
and L.F. Jaramillo 718 (COAH!); Serrania de Taraira, 6 km 66◦52′W. Alt. 100 m. 10 December 1978. O. Huber and Tillett
al. N-W del raudal de la Libertad, Coord. 0◦58′S, 69◦45′W. 3023 (K!); Rios Pacimoni–Yatua, Casiquiare, 26 September
Alt. 250 m. 27 July 1993. R. and J. Rodrigues 609 (COAH!); 1957. B. Maguire et al. 41583 (K!); Bolivar. Vicinity of Mount
Mpio. Mitú. Cabeceras de Caño Cuduyari, comunidad de Roraima, 1836. R.H. Schomburgk 1059 (BM!, K!, P!, W-R!,
Wacuraba, margen derecha del cano. Camino que conduce de W-R!–drawing); Atabapo. Falda del extremo norte del Cerro
la comunidad a la sabana de Yapoboda. 1◦22′23′′N 70◦54′30′′W. Duida. 3◦40′N 65◦45′W. Alt. 800–900 m. 6 February 1982.
Frontiers in Ecology and Evolution 11 frontiersin.org
106
fevo-10-1058334 January 6, 2023 Time: 11:34 # 12
Baranow et al. 10.3389/fevo.2022.1058334
J.A. Steyermark et al. 126106 (BM!, K). Guyana. Utshi R. trail to linear-oblanceolate, acute, with short apiculus, somewhat
to Santa Elena, Venezuela, 05◦39′N 61◦09′W. Alt. 980 m. 31 oblique. Petals up to 40 mm long and 13 mm wide, obovate-
January 1996. D.H. Clarke 942 (NY!, U!); Cuyuni-Mazaruni lanceolate, acute, oblique, margins of apical half undulate. Lip up
Mts. Karowrieng River, 0.5–1 km SE Maipuri Falls, trail to rock to 45 mm long and 33 mm wide, oblong obovate to pyriform in
drawings, 5◦40′N, 60◦13′W, Alt. 625–650 m, 15 October 1992. outline, apically rounded, emarginate, margins above basal third
B. Hoffman 3021 with T. Henkel and H. Kennedy (NY!); 3 km undulate-crispate, the base with two, 5 mm long, subparallel
SE of S end of Haiamatipu, above Kobadoi Savanna, 5◦27′N keels, basally in close proximity, forming a small cavity beneath,
60◦39′W, 549–610 m. 16 June 1991. T. McDowell et al. 4619 the disk with 9 erose-denticulate thickened keels, dilated at the
(NY!, U!). French Guiana. Cochoeira das Arraras, esatingas apex, the central five subtriangular. Gynostemium up to 30 mm
entre rio Vaupes e Arary. 3 November 1945. R. Lemos Froea long, semiterete, slender, somewhat clavate, with a pair of lateral,
21310 (K!, US!); Brazil. Amapa. Rio Araguari, downriver from falcate, acute stelidia at the apex, much exceeding the anther
Porto Platon. 21 September 1961. J.M. Pires, Wm. Rodrigues, apex, up to 6 mm long, anther white, pollinia white yellow
and G.C. Irvine 51146 (U!). Ad flumina Casiquari, Kasiva et (Figure 6).
Pacimoni. 1853-4. R. Spruce 3014 (BM!). Amazonas: Amza Ecology: Litophytic or terrestrial on granitic outcrops and
camp N5, 6◦4′S, 50◦08′W, Alt. 700–750 m. 12 May 1982. C.R. edges of white-sand shrubland. Flowering in February, March,
Sperling, R.S. Secco, M. Condon, A.L. Mesquita, B.G.S. Ribeiro, November, and December.
and L.R. Marinho 5609 (K!, MO!); Maraba, Alto de Serra, Distribution: Colombia, Venezuela. Alt. 90–350 m.
arredores do N5, solo de canga (ferro). 12 May 1982. R. S. Secco Conservation status: EOO—LC, AOO—EN.
et al. 116 (MO!); Marraba, Serra dos Canajas. 2 April 1977. M.G. Representative specimens (Supplementary Map 5)—
Silva and R. Bahia 2991 (K!); Rio Negro, near mouth of Rio Colombia. Guainía. Mpio. Pto Inírida. Comunidad El Remanso.
Xie, Vista Alegre, opposite Sao Marcelino, 0◦55′N, 67◦13′W. Cerro de Mavicure. Formaciones vegetales sobre roca gramitica,
21 October 1987. P.J.M. Maas, D.W. Stevenson, C. Farney, J.F. 3◦27′N 67◦58′W. Alt. 300 m. 25 March 1998. A. Rudas, A.
Ramos, and R.P. Lima 6832 (U!). Prieto, D. Angel, C. Cardenas, and M. Celis 7336 (COAH!,
The species is very similar to Sobralia granitica in flower MO!). Guaviare. PNN Nukak, San José del Guaviare, Inspec. del
structure. They can be distinguished by the size of the flowers—
Tomachipan, Río Inrida, Caño Cocui, Cerro Cocui, Sabaneta
flowers of the latter species are smaller (40–45 vs. 50–65 mm
on Roca, 2◦08′11.8′′N, 71◦09′41.2′′W. Alt. 350 m. 11 February
long in S. elisabethae). Additional differences can be observed in
1996. M.P. Cordoba, A. Etter, and H. Mendoza 2191 (COAH!);
flower color—S. elisabethae has a white lip with a yellow throat
Mpio. San José del Guaviare. Vereda la Pizarra, Camino la
and reddish orange keels, while S. granitica has a hyaline white
Lindosa-La Recebera. December 2005. V. Pinoz and D. Cardona
lip with pale yellow, elevated keels.
438 (COAH. Venezuela. Amazonas. Mpio. Atures. Bosque-laja
3.1.1.6. Sobralia granitica G.A. Romero & Carnevali en Cerro “Uchonhua” (lengua Piaroa), a unos 5 km al N del
Harvard Pap. Bot. 5 (1): 184. 2000. Type: Venezuela. caserio San Pedro de Catanipo, a unos 60 km al SE de Puerto
◦ ′ ◦ ′
Amazonas. Municipio Atabapo, Caño Ucata, Cerro Lombriz, 9 Ayacucho. 5 41 N, 67 11 W. Alt. 120–150 m. 9 November
December 1994, G.A. Romero and S. Llamozas 3016 (Holotype: 1980. F. Guanchez 366 (TFAV, VEN); Cerro de afloramiento
VEN; Isotypes: AMES!, K!, SEL).—Szlachetko et al. Materials to granitico a 3 km al N del Cesario Piaroa “Bablilla de Pintado,” al
◦ ′ ◦ ′
the Orchid Flora of Colombia 3: 254. 2020. S de Puerto Ayacucho. 5 32 N, 67 31 . Alt. 90–110 m. 26 March
Stems caespitose, cane-like, up to 130 cm high, terete, 1981. F. Gunachez 953 (TFAV, VEN); Cerro granitico al El del
◦ ′ ◦ ′
erect, basal internodes up to 15 cm long, leafless, apical Raudal Gavilan. 5 37 N, 67 22 W. Alt. 100 m. 1 February 1991.
internodes up to 2 cm long, leafy. Leaves 12 cm long, 1.5 cm G.A. Romero, C. Gomez, and E. Melgueiro 2291 (AMES!, TFAV,
wide, narrowly lanceolate, long-acuminate, rigidly coriaceous, VEN); Mpio. Atabapo. Caño Ucato, Cerro Lombiz. 9 December
articulate with their sheaths, the sheaths 3 cm long, tightly 1994. G.A. Romero and S. Llamozas 3016 (VEN, AMES!, K,
clasping the stem. Inflorescence terminal, sessile, elongating SEL). Bolivar. Cerro San Boja. Alt. 100–300 m. 12 December
with age, fractiflex, successively single-flowered, subtended by 1955. J.J. Wurdack and J.V. Monachino 39809 (AMES!, NY,
a foliaceous, articulate bract, up to 5 cm long, not including VEN).
the sheath. Flowers showy, with submembranaceous, widely Sobralia granitica is similar to S. liliastrum, but it differs
spreading perianth segments, lasting only 1 day, sepals white, in the smaller size of the flowers, (floral segments length of
the tips greenish–yellow petals and lip hyaline white, disk of S. granitica is 40–45 mm while in S. liliastrum 58–62 mm),
lip light yellow. Floral bracts non-articulate, up to 17 mm long, the color of the keels (light yellow in S. granitica vs. orange
subimbricating lanceolate, long-acuminate. Pedicellate ovary up in S. liliastrum), the raised keels (vs. only the central keel
to 18 mm long. Dorsal sepal up to 40 mm long and 8 mm wide, notably raised in S. liliastrum), and the absence of pseudopollen
narrowly elliptic to linear-elliptic, acute, with a short apiculus. on the lip (vs. present in S. liliastrum). The plants are easily
Lateral sepals 42 mm long, 9 mm wide, narrowly elliptic distinguishable in the field, but only with careful examination
Frontiers in Ecology and Evolution 12 frontiersin.org
107
fevo-10-1058334 January 6, 2023 Time: 11:34 # 13
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 6
Sobralia granitica G. A. Romero & Carnevali (A) apical part of stem with inflorescence, (B) dorsal sepal, (C) petal, (D) lateral sepal, (E) lip, (F)
gynostemium, side view, and (G) apical part of gynostemium, front view [drawn by P. Baranow, (A) from Rudas et al. 7336, (B–G) redrawn from
Romero-González (2003)].
of the keels, they can be told apart in herbarium material or inconspicuously dentate-erose, undulate. Gynostemium 26–
(Romero-González, 2003). 30 mm long, slender at the base, expanded on each side toward
the apex to form falcate horn-like apical stelidia (Figure 7).
Ecology: Epiphytic or terrestrial on road cuts
3.1.1.7. Sobralia luerorum Dodson and embankments. Flowering in January–April,
Orquideología 21 (1): 33. 1998. Type: ECUADOR. Azuay. August, and November.
Cuenca to Guarumales, between dam and casa de Maquinas, Distribution: Colombia, Ecuador. Alt. 1500–2200 m.
Alt. 1500 m. 9 March 1985. C.H. Dodson, P. Dodson, C., and J. Conservation status: EOO—LC, AOO—EN.
Luer 15872 (Holotype: RPSC!; Isotypes: AMES, QCA, QCNE– Representative specimens (Supplementary Map 6)—
illustration of type).—Szlachetko et al. Materials to the Orchid Ecuador. Azuay. Cola de San Pablo, Noreste de Paute en el Río
Flora of Colombia 3: 255. 2020. Paute, Entre Guarumales y el tunel. Alt. 1500 m. 9 March 1985.
Plants up to 350 cm tall, robust, caespitose, stem cane-like, C. and P. Dodson, C. and J. Luer and A. Hirtz 15782 (AMES!,
surrounded for the basal portion with clasping sheaths. Leaves RPSC!); Quebrada Chorro Blanco, Río Paute Valley, 8 km SE
◦ ′ ◦ ′
up to 35 cm long, 10 cm wide, elliptic to elliptic-lanceolate, of the Paute Dam at Amaluza, 78 33 W. 2 38 S. Alt. 1700 m.
coriaceous, acuminate, distichous, plicate, heavily veined on the 3 February 1988. U. Molau, B. Eriksen, and M. Fredrikson
2882 (MO!). Napo. Km 117–134, Quito-Tena, beyond Cosanga
underside, clasping the stem at the base, articulated to leaf-
at Cordillera de Guacamayo. Alt. 1900–2100 m. 17 January
sheath surrounding the stem. Inflorescence up to 20 cm long,
1990. C.H. Dodson and T. Neudecker 19193A (MO ex RPSC!).
fractiflex, with a large, spathe-like bract at each node, the flowers Tungurahua. Baños–Puyo road near Río Negro, border with
produced singly in succession over prolonged periods with Santiago-Zamora. Alt. 1200 m. 24 April 1980. A. Gentry and
flowering concurrent throughout the population. Sepals and C. Bonifaz 28740 (MO!). Colombia. Antioquia. Mpio Briceno.
petals white, the lip white heavily splashed with red–purple on Vereda San Fermin, 2–3 km sobre la via Ventanas (Mpio
the lamina and in the throat, veins in the lip yellow. Dorsal sepal Yaruma) Briceno, 7◦10′N, 75◦30′W. Alt. 1700–1900 m. 3
up to 75 mm long and 20 mm wide, narrowly oblong, acute. November 1990. R. Callejas and M.V. Arbelaez 9603 (AMES!,
Lateral sepals up to 70 mm long and 22 mm wide, narrowly NY!). Putumayo. Valle de Sibundoy, 1 km S Balsayaco. Alt.
elliptic, oblique, and acute. Petals up to 70 mm long and 22 mm 2200 m. 20 August 1963. M.L. Bristol 1319 (AMES!).
wide, oblong-obovate, obtuse, apical margins more or less erose. Sobralia luerorum is similar to S. gloriosa, but can
Lip up to 70 mm long and 40 mm wide, elliptic, retuse at the be distinguished by the larger flowers of thinner texture,
apex, with a pair of shallow lamellae in the throat, margins entire white sepals and petals, the lip with red–purple splashing
Frontiers in Ecology and Evolution 13 frontiersin.org
108
fevo-10-1058334 January 6, 2023 Time: 11:34 # 14
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 7
Sobralia luerorum Dodson (A) apical part of stem with inflorescence, (B) dorsal sepal, (C) petal, (D) lateral sepal, (E) lip, and (F) gynostemium,
side view (drawn by P. Baranow from Bristol 1319).
on the lamina and throat (in S. gloriosa sepals are 123. 1978.—Szlachetko et al. Materials to the Orchid Flora of
yellow to brown, the petals are white to yellow and Colombia 3: 256. 2020.
lip is white with purple striation), the elliptic-retuse Plants over 200 cm tall. Leaves up to 37 cm long and 12 cm
lip (vs. broadly elliptic and bilobed at the apex lip in wide, ovate to ovate-elliptic, long-acuminate. Inflorescence up
S. gloriosa), and a pair of calli in the throat of the lip to 30 cm long, rachis fractiflex, loosely many-flowered. Floral
inconspicuous (rather than large and conspicuous, as in bracts up to 80 mm long, cymbiform, the lowermost ovate-
S. gloriosa). lanceolate, acute, the upper ones obtuse, longer than the ovary.
According to the protologue, the specimens of S. luerorum Pedicellate ovary up to 22 mm long, cylindrical, glabrous.
reach up to 200 cm in height. However, some of the examined Flowers produced in succession, rather fleshy, creamy white
herbarium collections (e.g., C. and P. Dodson, C. and J. Luer and with purple striation on lip. Dorsal sepal up to 60 mm long
A. Hirtz 15782) and the plants cultivated in our living collection and 15 mm wide, ovate-lanceolate to oblong lanceolate, acute,
allow us to verify the information and state, that the stems can dorsally mucronate, the margins involute and undulate. Lateral
reach up to 350 cm. sepals up to 50 mm long and 18 mm wide, connate for 5 mm
basally, obliquely ovate-lanceolate to oblong lanceolate, acute,
3.1.1.8. Sobralia gloriosa Rchb.f. mucronate dorsally, with involute and undulate margins. Petals
Xenia Orchid. 2: 178. 1873. Type (designated by Garay, 1978: up to 60 mm long and 20 mm wide, oblong obovate to narrowly
122): Ecuador. Pichincha. From the forest of the Western side elliptic, obtuse, with crenulate and undulate margins above. Lip
of Pichnincha, Alt. 2300 m. Sep. W. Jameson 32 (Lectotype: W! up to 50 mm long and 40 mm wide, rhombic-elliptic in general
21547).–Garay in Harling and Sparre. Fl. Ecuador. Orchid. 9: outline, obscurely 3-lobed, lateral lobes erect, enfolding the
Frontiers in Ecology and Evolution 14 frontiersin.org
109
fevo-10-1058334 January 6, 2023 Time: 11:34 # 15
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 8
Sobralia gloriosa Rchb. f. (A) apical part of stem, (B) fragments of the inflorescence, (C) lip, (D) dorsal sepal, (E) petal, and (F) lateral sepal [drawn
by P. Baranow (A,B) and A. Król (C–F) from the lectotype].
column, rounded at apex; middle lobe suborbicular in outline, and hiked up ridgeline to 1550 m alt. 19 January 1987. C.H.
bilobed apically with wavy-crispate margin; disk obliquely Dodson, A. Hirtz, D. Benzing C., and J. Luer 16886 (RPSC!).
bicallose at the base and with 5–7 parallel verrucose thickened Pastaza. On roadside at km 70 Baños to Puyo. Alt. 1900 m.
veins from the base to the apex. Gynostemium up to 40 mm 18 February 1963. L.B. Thien 2270 (F!). Pichincha. km 88–92,
long, clavate, stelidia do not exceed the anther apex (Figure 8). Quito-Sto Domingo. Alt. 1200 m. 4 July 1979. C.H. Dodson, M.
Ecology: Terrestrial on steep embankments in wet montane Fallen and P. Morgan 7776 (RPSC!); Road from Quito to Santo
cloud forest. Flowering throughout the year. Domingo via Chiriboga, 8 December 1986. C.H. Dodson and
Distribution: Ecuador, Colombia, Peru (Garay, 1978). Alt. E. Hagsater 16702 (RPSC!); Reserva Floristica-Ecologica Río
1800–2300 m. Guajalito, km 59 de la carretera antiqua Quito-Sto Domingo
Conservation status: EOO—VU, AOO—EN. de Los Colorados, a 3.5 km al NE de la carretera, estribaciones
Representative specimens (Supplementary Map 7)— occidental del Volcan Pichincha. 0◦13′53′′S, 78◦48′10′′W. Alt.
Ecuador. Carchi. Trail from Rafael Quindis mountain finca to 1800–2200 m. 28 December 1985. J. Jaramillo 8312 (MO!);
Río Verde and short distance up Río Verde, 0◦52′N, 78◦8′W. Chiriboga road, old Santo Domingo-Quito road, 31 km
Alt. 1890 m. 28 November 1987. W.S. Hoover and S. Wormley northeast of Alluriquin, Alt. 6000 ft. 5 August 1980. R.P.
1873 (MO!); Ridge to NE of Rafael Quindis mountain finca, Sauleda et al. 4000 (AMES!); along road Nanegal-Nanegalito.
0◦52′N, 78◦8′W. Alt. 2000 m. 29 November 1987. W.S. Hoover Alt. 1200–1500 m. 9 July 1991. H. van der Werff, B. Gray,
2024 (MO!); Trail from Rafael Quindis mtn finca to Río Verde and G. Tipas 12264 (MO!). Colombia. Valle del Cauca. Along
and short distance up Río Verde, 0◦52′N, 78◦8′W. Alt. 1890 m. road between San José del Palmar and Ansermanuevo. 4◦49′N,
28 November 1987. W.S. Hoover and S. Wormley 1872 (MO!). 76◦09′W. Alt. 1960 m. 12 May 1983. T.B. Croat 56717 (COL!,
Imbabura. 8 km east of Lita on road to Ibarra and 8 km up MO!, NY!); Mpio. El Cairo. Vereda El Pacifico, 10 km desde
road from Cachaco to Santa Rosa de Cachaco to an elev 1150 m el desvio a San José del Palmar de la carretera Albán-Cartago.
Frontiers in Ecology and Evolution 15 frontiersin.org
110
fevo-10-1058334 January 6, 2023 Time: 11:34 # 16
Baranow et al. 10.3389/fevo.2022.1058334
4◦48′42′′N, 76◦10′16′′W. Alt. 1867 m. 29 December 2007. transverse thickenings that prevent the spreading of the base
R. Arevalo, J. Betancur, N. Salinas, L. Clavijo, and A. Zaluaga of the lip, the axial ridge lightly grooved, rather soft, rugulose,
804 (COL!); Hacienda Tokio, behind microwave tower, ca 10 km in apex finely bullate, white. Gynostemium ca 35 mm long.
S of Queremal, 3◦30′N, 76◦42′W. Alt. 2000 m. 26 February Stelidia linear, acute, slightly exceeding the gynostemium apex
1983. A. Gentry, A. Juncosa, and F. Gomez 40817 (COL!, MO!). (Figure 9).
Valle/Choco: Mpio El Cairo, Correg. Boqueron, Vereda Las Ecology: Terrestrial at the edges of forests and clearings.
Amarillas, Serrania de Los Paraguas, along road to and beyond Flowering from February till April and in September.
Cerro del Ingles, 17–23 km W of El Cairo, 4◦45′N, 76◦20′W, Alt. Distribution: Colombia, Venezuela. Alt. 1200–2400 m.
1750–2050 m. 13 May 1988. J. L. Luteyn, P. Silverstone-Sopkin, Conservation status: EOO—LC, AOO—EN.Representative
M. Dolores Hereida, and N. Paz 12274 (AMES!). Valle: Alt. specimens (Supplementary Map 8)—Colombia. Antioquia.
2000 m. May 1939. E. Dryander 2359 (US!). Region de Murri, road between Nutibara and La Blanquita,
Along with Sobralia luerorum, S. gloriosa has strongly 14.3–17.5 km from centro of Nutibara, 6◦45′N, 76◦23′N, 1620–
fractiflex rachis which allows to separate the two species from 1860, 10 February 1989. J.M. Mc Dougal, D. Restrepo, and D.S.
the other representatives of the section with broad leaves (over Sylva 3853 (MO!); Mpio. Frontino. Corregimiento Nutibara,
5 cm width). The taxa can be distinguished by flower color and Cuenca alta del Río Cuevas. Alt. 1640 m. 11 April 1987.
lip protuberances. The differences between them are indicated D. Sanchez et al. 1048 (MO!); Mpio. Frontino. Corregimiento
in the notes concerning S. luerorum. Nutibara, Cuenca alta del Río Cuevas. Sobre tulud, 2 m de
alto. Alt. 1640 m. 11 April 1987. F.J. Roldan, J. Betancur et al.
3.1.1.9. Sobralia ruckeri Linden & Rchb.f. 1048 (COL!, NY!); Mpio. Frontino. Corregimiento Nutibara,
Bonplandia (Corrientes) 2: 278. 1854. Type: Colombia. Cuenca alta del Río Cuevas. Alt. 1750 m. 14 April 1987.
Sine prec. loc. L.J. Schlim 1203 (Holotype: W!, UGDA- D. Sanchez et al. 1139 (COL!, MO!, NY!). Boyacá. Mpio.
DLSz!–drawing).—Garay & Dunsterville. Venezuelan Orchids Duitama. Trayecto entre la vereda El Carmen y Virolin. 21
Illustrated 406. 1959.—Szlachetko et al. Materials to the Orchid September 1994. J.L. Fernandez-Alonso, C. Ariza, A. Baena,
Flora of Colombia, 3: 257. 2020. J. Gomez, A. Espinoza, A. Pico, D. Riano, and D. Sarmiento
= Sobralia charlesworthii hort., Gard. Chron. 353. 1910. 12070 (COL!); Carretera Duitama. Charala, 65 km de Duitama.
Type: cult. ex Charlesworth (Holotype: K! 000364502). Adelante de Virolin. 9 June 1972. G. Lozano C. 2228 (COL!).
Plants up to 300 cm tall, robust, stem up to 1.5 cm Norte de Santander. Ocaña. Alt. 1700–2000 m. L.J. Schlim
in diameter, erect, leafy, growing in dense clumps, slightly 1203 (W!). Putumayo. Entre el Pepino y Mocoa. Cerca al Río
compressed or subterete. Leaves up to 35 cm long and 12 cm Putumayo. Alt. 1200 m. 11 January 1963. A. Fernandez P.
wide, lanceolate to ovate-lanceolate, attenuate, the uppermost 6015 (COL!). Santander. Gambita. Alt. 2400 m. 12 February
leaves tend to be somewhat cymbiform in the basal part, 2010. M. Ospina H. 1611 (COL!). Valle del Cauca. Km 18 y
sheaths spacious, ribbed. Inflorescence up to 6-flowered, several km 20 de la carretera de Cali a Buenaventura entrado por la
of which can be out simultaneously; strongly sinuous and finca Zingara. Cumbre de la Cordillera occidental. Alt. 1500–
stout, subterete. Floral bracts basally cymbiform and in their 2000 m. 28 February 1988. I. Cabrera R. and H. van der
apical portion almost identical to the leaves but smaller—up Werff 15766 (MO!). Venezuela. Zulia. Sierra de Perijá, Loma
to 200 mm long, getting progressively smaller toward the apex arriba de la quebrada del Río Omira-kuna (Tumuriasa), cerca
of raceme. Pedicellate ovary varies in length from 30 mm de la frontera Colombo-Venezolana suroeste de Pishikakao
in apical flowers up to 100 mm in the basal ones. Sepals e Iria hacia la Mision de Sucurpo. Alt. 1980 m. 27 March
very dark magenta–purple, sepals fairly intense rose–purple 1972. J.A. Steyermark, G.C.K., and E. Dunsterville 105664
with a pale mid-vein, petals rose–purple with a pale mid- (AMES!).
vein, lip dorsally rose–purple grading to a very dark wine– The characteristic leaves which are gradually getting smaller
purple at the apex, ventrally with a large patch of light rose– toward the apical part of the stem and fluently transform into
purple at the base, changing abruptly to very dark wine– leaf-like floral bracts are unique among the whole genus and
purple for the remainder. Dorsal sepal up to 85 mm long and allow us to distinguish the species at the first glance. The
25 mm wide, ligulate-lanceolate to linear-oblanceolate, acute, features, along with the shape and color of the floral segments,
mucronate, fleshy, basally connate with lateral sepals. Lateral especially the lip, prompted the decision to synonymize
sepals up to 85 mm long and 20 mm wide, oblanceolate, S. charlesworthii under the name S. ruckeri. Such a concept was
acute, mucronate, more or less falcate. Petals up to 85 mm mentioned in the description of S. charlesworthii—it suggests
long and 33 mm wide, widely oblanceolate or oblong obovate, that S. charlesworthii may be just a form of S. ruckeri.
oblique, firm but much thinner than the sepals, the thin The only species that could be misidentified with Sobralia
margin of the apical third variably undulate. Lip 80 mm long, ruckeri is S. splendida, but the latter taxon differs in the
60 mm wide, ovate to elliptic, axis strongly thickened ventrally protuberances present on the lip surface. The details are listed
into a yellow ridge that starts from the thick, finely sulcate, in the notes concerning S. splendida.
Frontiers in Ecology and Evolution 16 frontiersin.org
111
fevo-10-1058334 January 6, 2023 Time: 11:34 # 17
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 9
Sobralia ruckeri Linden & Rchb. f. (A) apical part of stem with inflorescence, (B) dorsal sepal, (C) petal, (D) lateral sepal, (E) lip, and (F)
gynostemium, side view [drawn by P. Baranow (A) and A. Król (B–F) from the holotype].
3.1.1.10. Sobralia gambitana Baranow, Szlach. & long, becoming smaller toward the rachis apex, lanceolate-
Kindlmann, sp. nov. cymbiform, acuminate. Ovary 10–11 mm long, cylindrical.
Type: COLOMBIA. Santander. Mnio Gámbita, vereda El Dorsal sepal 66 mm long, 22 mm wide, lanceolate or elliptic-
Palmar. Alt. 2500 m. 12 May 1982. A. Becerra and M. Constanza lanceolate, shortly acuminate. Lateral sepals 60 mm long,
23 (Holotype: COL! 256896, UGDA-DLSz!–drawing). 20 mm wide, oblong elliptic, shortly acuminate, inconspicuously
Similar to S. tamboana in habit and flower structure. oblique. Petals 62 mm long, 36 mm wide, obovate, linear and
However, it can be separated by the pair of thickenings at the falcate basally, acute, margins slightly crenate. Lip 60 mm long,
base of lip—they are fused together in the new species while in 50 mm wide, rhombic-elliptic in outline, deeply concave basally,
S. tamboana, they are separated. The two species differ also in the margins of apical half undulate, base with two united keels
color of flowers—they are lilac with purple lip edges and yellow running up to one-third of the lip. Gynostemium 37 mm long,
at the center in the new entity. In S. tamboana, flowers are pale slightly falcate, club-like, apical stelidia triangular, falcate, acute,
yellow with a red-brown wash inside the throat of the lip and a not exceeding the anther (Figure 10).
Ecology: No data. Flowering in May.
red-brown spot on the lamina of the lip. Moreover, S. gambitana
Distribution: Colombia (Santander). Alt. 2500 m.
is two times as tall as S. tamboana (ca 250 cm vs. 120 cm) while
Conservation status: EOO—CR, AOO—CR.
its floral segments are distinctly smaller than those of S. tamboana Representative specimens (Supplementary Map 9)–
(60–66 vs. 78–92 mm). Colombia. Santander. Mnio Gámbita, vereda El Palmar. Alt.
Etymology: Named in allusion to Colombian Municipio 2500 m. 12 May 1982. A. Becerra and M. Constanza 23 (COL!
Gámbita, where the type material was collected. 256896, UGDA-DLSz!–drawing).
Plants ca 250 cm tall. Leaves 13–14 cm long, 1.8– The descriptions of the new taxon on the basis of a single
2.5 cm wide, oblong elliptic linear-lanceolate, acuminate, basally collection may be doubtful, but in this case, we have the
cuneate, strongly plicate. Inflorescence ca. 4 cm long, rachis combination of the morphological features and flower color
flexouose, glabrous. Flowers lilac, lip with purple edges and that convince us that the collection deserves the status of a
a yellow line in the center. Floral bracts up to 20 mm separate species.
Frontiers in Ecology and Evolution 17 frontiersin.org
112
fevo-10-1058334 January 6, 2023 Time: 11:34 # 18
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 10
Sobralia gambitana Baranow, Szlach. & Kindlmann (A) apical part of stem with inflorescence, (B) dorsal sepal, (C) petal, (D) lateral sepal, (E) lip,
and (F) gynostemium, side view (drawn by P. Baranow from the holotype).
Sobralia gambitana is similar to S. tamboana in habit and than those of S. tamboana (60–66 mm vs. 78–92 mm). The
flower structure. However, it can be separated by the position of two species differ also in the size of leaves and rachis of the
the lip basal thickenings. In the new species, the two basal ridges inflorescence. The detailed comparison is presented in Table 1.
are fused together while in S. tamboana they are separated. The It looks like S. tamboana, but has relatively larger
two species differ also in flower color—the new species are lilac floral bracts and more slightly fractiflex inflorescence than
with purple lip edges and yellow at the center. In S. tamboana S. gambitana.
flowers are pale yellow with a red-brown wash inside the throat
of the lip and a red-brown spot on the lamina of the lip. 3.1.1.11. Sobralia tamboana Dodson
Sobralia gambitana is two times as tall as S. tamboana (ca Orquideología 21 (1): 44. 1998. Type: ECUADOR.
250 cm vs. 120 cm) while its floral segments are distinctly smaller Esmeraldas. Lita to San Lorenzo, Km 6, Alt. 650 m. 29
Frontiers in Ecology and Evolution 18 frontiersin.org
113
fevo-10-1058334 January 6, 2023 Time: 11:34 # 19
Baranow et al. 10.3389/fevo.2022.1058334
TABLE 1 Comparison of Sobralia gambitana Baranow, Szlach. and Conservation status: EOO—CR, AOO—CR.
Kindlmann and S. tamboana Dodson. Representative specimens (Supplementary Map 10)—
Characters Sobralia Sobralia Ecuador. Lita to San Lorenzo, Km 6, Alt. 650 m, 29 December
gambitana tamboana 1990. C.H. Dodson and T., and P.M. Dodson 19096 (RPSC!); Km
Plant height 250 cm 120 cm 5, Lita to El Cristal, Alt. 250 m. 26 March 1993. C.H. Dodson and
Leaves length 13–14 cm 26 cm G. Carnevali 19243 (RPSC!).
Similar to Sobralia rosea but distinguished by flexuose
Leaves width 1.8–2.5 cm 8 cm
inflorescence, the pale yellow flowers with a diffuse red-brown
Inflorescence length 4 cm 12 cm spot and the lack of low, parallel lamellae on the lip lamina.
Ovary 10–11 mm 40 mm
Dorsal sepal size 66 mm× 22 mm 92 mm× 28 mm 3.1.1.12. Sobralia splendida Schltr.
Dorsal sepal shape Lanceolate or Narrowly oblong-elliptic Repert. Spec. Nov. Regni Veg., Beih. 7: 44. 1920. Type:
elliptic-lanceolate Colombia. Sine prec. loc. M. Madero (B†).—Szlachetko et al.
Dorsal sepal apex Shortly acuminate Acute Materials to the Orchid Flora of Colombia 3: 258. 2020.
Plants up to 300 cm tall, stem erect, leafy, growing in
Lateral sepals size 60 mm× 20 mm 80 mm× 30 mm
dense clumps, slightly compressed or subterete. Leaves 40–
Lateral sepals shape Oblong elliptic Obliquely oblong ovate 45 cm long, 9–10 cm wide, elliptic, acuminate, basally cuneate.
Lateral sepals apex Shortly acuminate Acute Inflorescence 10–12 cm long. Rachis flexouose, glabrous. Flower
Petals size 62 mm× 36 mm 78 mm× 30 mm color unknown. Floral bracts up to 90 mm long, lanceolate-
Petals shape Obovate Obliquely oblong-elliptic cymbiform, acuminate. Ovary 40 mm long, cylindrical. Sepals
basally connate together for one/fourth of their length. Dorsal
Petals apex Acute Obtuse
sepal 85–90 mm long, 20 mm wide, oblong-ligulate to
Lip size 60 mm× 50 mm 80 mm× 40 mm oblanceolate, acute. Lateral sepals 85–90 mm long, 20 mm wide,
Lip shape Rhombic-elliptic Oblong-elliptic oblong-ligulate, acute, oblique. Petals 85–90 mm long, ligulate-
Lip basal keels arrangement United Separated oblanceolate, acute, slightly wider, and thinner in texture than
the sepals, subfalcate. Lip 90 mm long, 40 mm wide, oblong
ovate in outline above cuneate base, margins of apical half
December 1990. C. H. Dodson and T. and P. M. Dodson 19096 undulate, base with two parallel keels running up to its middle,
(Holotype: RPSC!; illustration of type). the central vein in the central part of the lip ornamented with
Plants up to 120 cm tall, caespitose, rhizome short, stems lamella, each of the protuberances with parallel rows of papillae
cane-like, surrounded in the basal portion with clasping sheaths. on both sides. Gynostemium 57 mm long, slightly curved, apical
Leaves up to 26 cm long and 8 cm wide, elliptic, chartaceous, stelidia oblong-falcate, not exceeding the anther (Figure 12).
acuminate at the apex, distichous, plicate, and heavily veined on Ecology: Terrestrial.
the underside. Inflorescence ca 12 cm long, lightly flexuose with Distribution: Colombia. Alt. 500 m.
large, spathe-like bract at each node, the flowers produced singly Representtive specimens—Colombia. Cauca. Alt. ca. 500 m.
in succession over prolonged periods with flowering concurrent M. Madero (B†).
throughout the population. Ovary ca 40 mm long. Flowers pale According to Schlechter (1920), this species is similar to
yellow with a red-brown wash inside the throat of the lip and Sobralia ruckeri, from which it differs by the lip structure, i.e.,
a red-brown spot on the lamina of the lip. Sepals free to the by the presence of the prominent, high papillae arranged in
base. Dorsal sepal 92 mm long, 28 mm wide, narrowly oblong- the rows running on both sides of each of the ridge of the lip.
elliptic, acute. Lateral sepals to 80 mm long and 30 mm wide, Our study supports his observations. The other species similar
obliquely oblong ovate, acute. Petals to 78 mm long and 30 mm in lip form and gynostemium morphology to S. splendida is
wide, obliquely oblong-elliptic, obtuse, lightly reflexed at the S. hoppii. In the former species, the lip base is ornamented with
apex, margins slightly crenate in the apical half. Lip up to 80 mm two lamellae running to its middle and the central vein in the
long and 40 mm wide, oblong-elliptic in general outline, upper central part is ornamented with lamella as well. In the latter, the
half more or less deltoid, flared, retuse at the apex, concave, lip has two basal keels, and the median vein is thickened, with
throat with a pair of shallow lamellae. Gynostemium 40 mm two additional thickenings near the middle.
long, slender at the base, flattened on the underside, expanded
on each side toward the apex to form falcate, horn-like stelidia 3.1.1.13. Sobralia hoppii Schltr.
(Figure 11). Repert. Spec. Nov. Regni Veg., Beih. 27: 13. 1924.
Ecology: Epiphytic or terrestrial on road cuts and Type (designated here): Colombia. Caqueta. Ostkordillere,
embankments. Flowering in March and December. Putumayo-Gebiet, Alt. 3000 m. September 1922, W. Hopp 164
Distribution: Ecuador. Alt. 250–650 m. (B†); Von Buenaventura bis Juntas. Alt. to 300 m. 21 July
Frontiers in Ecology and Evolution 19 frontiersin.org
114
fevo-10-1058334 January 6, 2023 Time: 11:34 # 20
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 11
Sobralia tamboana Dodson (A) apical part of stem with inflorescence, (B) dorsal sepal, (C) petal, (D) lateral sepal, (E) lip, and (F) gynostemium,
front view [drawn by P. Baranow from Dodson (1998)].
1881. W. Hopp 753 (Neotype: W!, UGDA-DLSz!–drawing).— 10–14 mm wide, oblong-ligulate, acuminate. Lateral sepals
Szlachetko et al. Materials to the Orchid Flora of Colombia 3: 60–83 mm long, 10–14 mm wide, obliquely oblong-ligulate,
259. 2020. acuminate. Petals 52–83 mm long, 22 mm wide, obliquely
Plants probably 150 cm tall, erect, robust, glabrous. Leaves oblong, obtuse to subobtuse, with more or less undulate
23–30 cm long, 7.5–11 cm wide, elliptic, acuminate, many- margins. Lip 52–85 mm long in total, 30–40 mm wide
veined, coriaceous, stiff. Raceme up to 35 cm long, 5–12- when expanded, unguiculate, ovate to oblong ovate in the
flowered, rachis flexuose, glabrous, or sparsely furfuraceous. general outline above, more or less pandurate toward apical
Flowers rather large, pure white or yellowish-white. Floral quarter, emarginate, undulate in front, with 2 basal keels,
bracts up to 130 mm long, ovate, long-acuminate. Ovary median vein thickened, with two additional thickenings
32 mm long, glabrous. Dorsal sepal 60–83 mm long, near the middle. Gynostemium 37–67 mm long, stelidia
Frontiers in Ecology and Evolution 20 frontiersin.org
115
fevo-10-1058334 January 6, 2023 Time: 11:34 # 21
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 12
Sobralia splendida Schltr. (A) dorsal sepal, (B) lateral sepal, (C) petal, (D) lip, and (E) gynostemium, side view [redrawn by A. Król from Schlechter
(1929)].
relatively obscure, obliquely triangular, shorter than anther lip and its callosities, the species is easily distinguishable from all
(Figure 13). other taxa of this group.
Ecology: Terrestrial. Flowering in May, July, and in As the original collection is not available—we assume it
September. could have been destroyed during World War II—we decided to
Distribution: Colombia. Alt. designate the neotype for the species. We have chosen the only
300_metricconverterProductID3000 m–3000 m. existing collection of the species gathered by Hopp (no. 753),
Conservation status: EOO—CR, AOO—CR. who was also the collector of the original type material. Besides,
Representative specimens (Supplementary Map 11)— the selected collection is well documented by the drawings left
Colombia. Caquetá. Putumayo-Gebiet, Ostkordillere. Alt. in Vienna and UGDA herbaria.
3000 m. September 1922. W. Hopp 164 (Schlechter, 1924).
Chocó. Mpio. Carmen del Atrato. Carretera Quibdó-Carmen 3.1.1.14. Sobralia pulcherrima Garay
del Atrato. 5◦43.6′-43.5′N, 76◦36.2′-18.4′. Alt. 80–510 m. 11 In Harling & Sparre, Fl. Ecuador 9: 128. 1978. Type:
May 2007. R. Arevalo, J. Betancur, S. Hoyos, and E. Renteria 740 Ecuador. Pichincha: road Nanegal–Nanegalito, Alt. 1200–
(COL!). Valle del cauca. Von Buenaventura bis Juntas. Alt. to 1550 m. G. Harling and L. Andersson 11571 (Holotype:
300 m. 21 July 1881. W. Hopp 753 (W!, UGDA-DLSz!–drawing). GB; Isotype: AMES 00104326; K–drawing!).—Szlachetko et al.
According to Schlechter (1924), this species resembles Materials to the Orchid Flora of Colombia 3: 260. 2020.
Sobralia rosea and can be easily misidentified with it, but has = Sobralia lindenii Grignan, Lindenia 13: t. 5855. 1895, not
smaller, pure white or yellowish-white flowers. In the form of the hort. Type: no data.
Frontiers in Ecology and Evolution 21 frontiersin.org
116
fevo-10-1058334 January 6, 2023 Time: 11:34 # 22
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 14
Sobralia pulcherrima Garay (A) apical part of stem with
inflorescence, (B) lateral sepal, (C) dorsal sepal, (D) petal, (E) lip,
FIGURE 13 and (F) gynostemium, side view (drawn by P. Baranow from
Sobralia hoppii Schltr. (A) apical part of stem with the Asplund 16728).
inflorescence, (B) lip, and (C) apical part of the gynostemium,
side view (drawn by P. Baranow from the neotype).
115 mm long and 65 mm wide, ovate-elliptic in general
Plants up to 400 cm tall, caespitose. Stem erect, rather outline, with a tubular base, then flabellate spreading in
robust, lower half leafless, completely enclosed by remnants front, very undulate-crispate, bilobed in front with erose
of leaf sheaths, leafy above. Leaves up to 33 cm long denticulate margin, disk with 3 lamellae running from the
and 9 cm wide, ovate-lanceolate or elliptic-lanceolate, long- base to the middle, the median lamella erect, high-carinate, the
acuminate, gradually tapering to a more or less rounded lateral ones appressed to the disk, on both sides of lamellae
base, sessile and articulated with glabrous sheaths, plicate. veins thickened and barbate. Gynostemium up to 65 mm
Inflorescence up to 30 cm long, sessile, elongating with long, clavate, arcuate, bifalcate, stelidia as long as anther
age, flexuous, loosely few-flowered. Flowers 1 or 2 at a (Figure 14).
time produced in succession, large, showy, white with dark Ecology: Terrestrial in lowland and premontane forest
purple veins on the lip disk. Floral bracts up to 100 mm edges. Flowering throughout the year.
long, ovate-lanceolate, cymbiform, erectly spreading with Distribution: Ecuador, Colombia. Alt. up to 2000 m.
arcuate tips. Pedicellate ovary up to 30 mm long. Dorsal Conservation status: EOO—LC, AOO—EN.
sepal up to 105 mm long and 30 mm wide, oblanceolate- Representative specimens (Supplementary Map 12)—
oblong, subfleshy, acute or abruptly acuminate, somewhat Ecuador. Carchi. Approx. 3 km above Maldonado. Alt.
tapering toward the base, more or less undulate, connate 1550 m. B. Boyle and J. Bradford 1854 (MO!); Maldonado
with lateral sepals for up to 15 mm. Lateral sepals up to to Chical, km 3. Alt. 1410 m. 30 April 1993. C.H. Dodson
105 mm long and 30 mm wide, lanceolate to oblanceolate- 19084 (RPSC!). Esmeraldas. Lito to San Lorenzo, km 4.
oblong, subfleshy, acute or abruptly acuminate, somewhat Alt. 230 m. 26 March 1994. C.H. Dodson and G. Carnevali
tapering toward the base, margins undulate. Petals up to 19235 (RPSC!). Pastaza. Puyo-Napo road. 11–18 October
105 mm long and 30 mm wide, oblanceolate-obovate, acute, 1975. P.M. Synge 9 (K!). Pichincha. near the bridge over
subfalcate, with more or less undulate margins. Lip up to the Río Pilaton between Chiriboga and Santo Domingo de
Frontiers in Ecology and Evolution 22 frontiersin.org
117
fevo-10-1058334 January 6, 2023 Time: 11:34 # 23
Baranow et al. 10.3389/fevo.2022.1058334
los Colorados. Alt. 1100 m. 1 July 1955, E. Asplund 16728 foothills of the Andes. Both species while pressed and
(AMES!); About 65 miles SW of Quito. 12 November 1969. dried out can be separated by the lip details. The
P. Clark s.n. (F!); Road Nanegal to Nanegalito. Alt. 1200– lip disk of S. pulcherrima has three lamellae running
1550 m. G. Harling and L. Andersson 11571 (GB; AMES!, from the base to the middle, and the median one
K–drawing!); Santo Domingo-Quito Road, 8 km northeast of is high-carinate. On the contrary, the lip disk of
Santo Domingo. Alt. 76 m. 29 July 1980. R.P. Sauleda, M. S. rosea from the base to the center is transversed
Ragan, H. Luther, R. Wunderlin, B. Hansen, L. Davenport,and by 5–7 low, parallel ridges, with fine, radiating, white
J. Wiersema 3799 (AMES!, MO!, U!); Route Tandayapa- veins in the center.
Nenegalito, Fundacion Maquipucuna, 00◦00′S, 78◦40′W. Alt.
1400 m. 24 January 1996. F. Billet and B. Jadin 6700 3.1.1.15. Sobralia rosea Poepp. & Endl.
(MO!). Zamora-Chinchipe. Cordillera del Cóndor, vertiente Nov. Gen. Sp. Pl. 1: 54, t. 93. 1836. Type (designated
occidental. Cuenca del Río Tundayme. Carretera hacia el by Szlachetko et al.:261. 2020): Peru. Sine loc. E.F. Poeppig
destacamento militar Condor Mirador. Formacion rocosa 1076 (Lectotype: W! 47809, Isolectotype: W! 47808).–
arenisca, suelo arenoso. 3◦37′48′′S, 78◦26′50′′W. Alt. 1690– Schweinfurth. Orchids of Peru 74. 1958.–Garay in Harling
2000 m. 21 March 2006. W. Quizhpe and F. Luisier 2034 & Sparre. Fl. Ecuador. Orchid. 9: 133. 1978.—Szlachetko
(MO!). et al. Materials to the Orchid Flora of Colombia 3: 261.
Colombia. Chocó. Road between Medellín and Quibdó 2020.
at km 134.5. 5◦46′N, 76◦20′W. Alt. 1070 m. 13 April 1983. =Sobralia lindenii Hort., Gard. Chron. 18: 424. 1895. Type:
T.B. Croat 55918 (MO!); Carretera Tutunendo-El Carmen. Introduced from tropical America, flowered by T. Lawrence in
Entre km 135 y 120. Alto Río Atrato. Alt. 800–1200 m. 29 1894 and C. J. Lucas in 1895 (Holotype: K!).
April 1979. E. Forero, R. Jaramill M., H.Y. Bernal, H. Leon, Plants up to 150 cm tall. Stem erect, robust, cane-
and M.M. Pulido 6091 (COL!, P!); Hoya del Río San Juan. like, leafless below, many-leaved above. Leaves up to 35 cm
Arriba de Palestina, entre Quebrada La Sierpe (Palestina) long and 8 cm wide, lanceolate, long-acuminate, sessile on
y Quebrada El Quicharo. 4◦10′N, 77◦10′W. 27 Mar. 1979. glabrous sheaths. Inflorescence up to 15 cm long, sessile, few-
E. Forero, R. Jaramillo M., L.E. Forero P., and Hernandez N. flowered, flexuosus. Flowers 1 or 2, produced in succession,
4103 (COL!, MO!); Río Yuto between Lloró and La Vuelta. rather thin in texture, pale rose color, the main disk of
Alt. 100 m. 18 January 1979. A. Gentry and E. Renteria lip dark purple–magenta with narrow white margin and
A. 17426 (COL!); Ca 15 km W of Siete. 6 January 1979. transversed by white veins. Floral bracts up to 90 mm,
A. Gentry and E. Renteria A. 23718 (COL!, MO!); Río Yuto cymbiform, ovate, acute to acuminate. Pedicellate ovary up
between Lloró and La Vuelta. Alt. 100 m. 18 January 1979. to 25 mm long, cylindric, glabrous. Dorsal sepal up to
A. Gentry and E. Renteria A. 24348 (COL!, P!); Hwy. Bolivar- 100 mm long and 20 mm wide, narrowly oblanceolate,
Quibdó, near km 135, 5◦50′N, 76◦20′W. Alt. 975 m. 28 acute, dorsally fleshy, subulate, basally connate with lateral
October 1983. A. Juncosa 1122 (MO!, NY!). Valle del Cauca. sepals for up to 10 mm. Lateral sepals up to 100 mm
Mpio Dagua. Corregimientoo El Danubio, Alto Anchicaya. long and 20 mm wide, narrowly oblanceolate, acute, dorsally
Alt. 200 m. 19 June 1984. W. Devia A. 568 (MO!); Río fleshy, subulate. Petals up to 100 mm long and 25 mm
Anchicaya near CVC hydroelectric plant, 3◦40′N, 76◦50′W. wide, oblanceolate-elliptic, obtuse to acute, with somewhat
Alt. 400–500 m. A. Gentry 35656 (COL!, MO!); Carretera undulate margin. Lip up to 110 mm long and 55 mm
from Buenaventura to Cali, km 20. 4 June 1982. H. Murphy wide, oblong ovate to ovate-elliptic in outline, tubular
573 (COL!, MO!); Along the road El Queremal-La Elsa. On in a natural position, then expanding in a suborbicular,
steep slopes. 15 February 2011. D. Szlachetko, A. Niessen and bilobed, frontal blade, when expanded, from a cuneate
M. Moreno s.n. (UGDA-DLSz–spirit!); Between Buenaventura base obovate, undulate, crispate in front, retuse to deeply
and Cali on old highway, 5 km S of Río Sabaletas along bilobed at apex, disk transversed in center by 5–7 low,
steep soggy bank along road, 3◦44′N 76◦57′W. Alt. 145 m. parallel ridges which are confluent at base, where on
10 February 1990. T.B. Croat and J. Watt 70413 (CUVC!, both sides subpubescent or papillose. Gynostemium up to
MO). 55 mm long, clavate, arcuate, stelidia shorter than anther
This species resembles Sobralia rosea but can be (Figure 15).
distinguished by the flower color—S. pulcherrima always Ecology: Terrestrial in forest edges and steepy river sides.
has a white lip with broad white margins and the disk Flowering throughout the year.
is prominently purple-veined. S. rosea is always with a Distribution: Ecuador, Colombia, Peru, Brazil. Alt. from sea
narrow, white margin while the whole disk is crimson- level up to 3300 m.
purple with white radiating veins. S. pulcherrima is Conservation status: EOO—LC, AOO—EN.Representative
limited in distribution to the western foothills of the specimens (SupplementaryMap 13)—Ecuador. Azuay. Cola de
Andes, while S. rosea can be found on the eastern San Pablo, Norriente de Paute en el Río Paute. Alt. 1300 m. 9
Frontiers in Ecology and Evolution 23 frontiersin.org
118
fevo-10-1058334 January 6, 2023 Time: 11:34 # 24
Baranow et al. 10.3389/fevo.2022.1058334
FIGURE 15
Sobralia rosea Poepp. and Endl. (A) apical part of stem with inflorescence, (B) lip, (C) dorsal sepal, (D) petal, (E) lateral sepal, and (F)
gynostemium, side view (drawn by P. Baranow from lectotype).
March 1985, C. and P. Dodson, C. and J. Luer, and A. Hirtz Rios. Quevedo-Latacunga road, km 46 from Quevedo, 79◦11′W.
15779 (RPSC!). Esmeraldas. Along Río Lita in the vicinity of 0◦55′S. Alt. 600 m. 4 April 1973, L. Holm-Nielsen, S. Jeppesen,
the village of Lita. Alt. 600–650 m. 8 September 1976, T.B. B. Lojtnant, and B. Ollgaard 2896 (AMES!, K!, MO!). Carchi.
Croat 38937 (MO!); Km 11 Lita to San Lorenzo. Alt. 760 m. Road Tulcán to Maldonado via Paramo El Angel, km 74. Alt.
12 May 1990, C.H. Dodson, A. Gentry, B. Boyle, and D. Rubio 1750 m. 1 August 1985, C.H. Dodson and A. Embree 16192
18241 (RPSC!); Along road under construction from Lita to (RPSC!); Between Chical and Peñas Blancas trailside and forest
Alto Tambo (21 km). Alt. 750–820 m. 19 May 1987, C.H. edge, valley of San Juan on Colombia border. Alt. 1100–1250 m.
Dodson, H. van der Werff, and W. Palacios 17130 (RPSC!). Los 25 September 1979, A. Gentry and G. Shupp 26476 (MO!).
Frontiers in Ecology and Evolution 24 frontiersin.org
119
fevo-10-1058334 January 6, 2023 Time: 11:34 # 25
Baranow et al. 10.3389/fevo.2022.1058334
Cotopaxi. Tenefuerte, km 52 Quevedo-Latacunga. Alt. 800– side of Río Pastaza, 3 March 1969, H. Lugo S. 621 (AMES!,
900 m. 9 April 1984, C.H. Dodson and W. and M. Thurston MO!); Along Pastaza River below Machay. Alt. 1350 m. 18
14216 (RPSC!); Tenefuerste. Río Pilalo, km 52–53, Quevado, March 1939, C. W. Penland and R.H. Summers 113 (AMES!);
Latacunga. Alt. 750–1300 m. 21 February 1982, C.H. Dodson Río Verde. 21 April 1971, H. Lugo S. 1770 (AMES!, MO!);
and A.H. Gentry 12726 (RPSC!). Morona-Santiago. Indanza- Along road from Baños to Puyo from Río Blanco to Puyo. Alt.
Limón (General Plaza). Alt. 1300–1600 m. 23 March 1974, 700–1800 m. 23 February 1963, L.B. Thien 2302 (F!). Zamora-
G. Harling and L. Andersson 12753 (AMES!). Napo. Reserva Chinchipe. Road Loja-Zamora, km 54, 78◦59′W. 4◦02′S. Alt.
Biologica Jatun Sacha. Río Napo, 8 km al. E de Misahuallí. 1300 m. 18 April 1973, L. Holm-Nielsen, S. Jeppesen, B. Lojtnant,
1◦04′S, 77◦36′W. Alt. 450 m. 24 April–5 May 1987, C.E. Ceron and B. Ollgaard 3774 (AMES!, K!, MO!); Río Negro, Rd. Baños-
M. 1302 (MO!); Laguna Anangu, N side, 00◦31′S, 76◦24′W. Puyo. Alt. 1500 m. 15 October 1984, C.H. and P.M. Dodson,
Alt. 250 m. 25 January 1985, B. Ollgaard 57158 (MO!). Napo- and A. Hirtz 15370 (RPSC!); Road Loja to Zamora, km 48.
Pastaza. Valley of Río Pastaza and adjacent uplands. Alt. 1060– Alt. 1400 m. 17 May 1867, B. Sparre 16344 (US!). Zamora-
1500 m. 17 April 1945, W.H. Camp E-2382 (AMES!); Mera. Chinchipe. Road Loja-Zamora, El Retorno-Zumbi. Alt. 1000 m.
1 March 1940, H. Lugo M. 7 (B!, MO!). Pastaza. Along the May 1985, D. Dalessandro 460 (RPSC!). Colombia. Antioquia.
highway between Shell and Mera. Alt. 1000 m. 18 March 1988, Hillsides near Puente Linda, 5 km above Río Samana. Alt.
B. Boom and D. Beardsley 8442 (US!); Pastaza Canton, Estacion 1000 m. 26 July 1960, F.A. Barkley and G. Gutierrez V. 35345
experimental Pastaza, via Puyo–Macas, Trama km 31.5–33 Puyo (AMES!); Río Grande. April 1947, Bro Daniel 4000 (US!). Cauca.
Macas, borde del carretero. 1◦30′S, 77◦56′W. Alt. 1040 m. 16 El Tambo, Parque Nacional Natural Munchique, vereda La
February 2002, J. Caranqui, M. Melampy, and J. Lara 399 (MO!); Romelia, la Gallera. Alt. 2835 m. 26 July 1993, C. Barbosa
Cantón Arajuno, bosque protector Pablo Lopez del Oglan Alto et al. 8588 (COL!, MA!). Chocó. Río San Juan, cercenias de
y Estacion Cientifica de la Universidad Central de Ecuador, Palestina. Alt. 5–50 m. 12–14 March 1944, J. Cuatrecasas 16942
1◦19.25′S, 77◦41.19′W. Alt. 600 m. 5 March 2006, C.E. Ceron, (AMES!; F!); Km 55 de la carretera Ansermanueve-San José del
C.I. Reyes, and L. Marcelo Vargas 56657 (MO!); Along the road Palmar. Alt. 1700–1950 m. 19 March 1980, G.C. Lozano and J.
between Puyo and Baños, 2.7 km W of Mera, 4.6 km W of Diaz 3229 (COL!, F!). Nariño. Mpio Tumaco. La Guayacana.
Shell, 1◦27′S, 78◦50′W. Alt. 1110 m. 5 May 1984, T.B. Croat 27 June 1951, R. Romero Castañeda 2909 (COL!, MO!). Mpio.
59084 (MO!); Puyo-Puerto Napo road. 25 December 1972, R.H. Barbacoas. Chucunes via La Planada a 1 km antes de llegar
Williamse 16 (U!); Hacienda San Antonio Baron von Humboldt, a la reserve. Alt. 1800 m. 10 March 1995, G. Lozano, J.L.
2.5 km Norte de Mera en carretera a Baños-Puyo. Alt. 1050– Fernadez Alosno, and E. Morales 6878 (COL!); Mpio. Barbacoas.
1300 m. 23 March 1985, C.H. Dodson and L.M. Bermeo 15604 Correg. Junin. Via Junin-Barbacoas. Alt. 960–1100 m. 14 March
(AMES!, K!, MO!); Mara, road cut near Mangayacu. Alt. 1100 m. 1995, G. Lozano, J.L. Fernadez Alosno, and E. Morales 6977
28 January 1956, E. Asplund 19085 (AMES!, B!, K!); Hacienda (COL!); Barbacoas. Corregimiento Santander (Buenavista) a
San Antonio Baron von Humboldt, 2 km al. Norte de Mera, Barbacoas (Vertiente del Río Telembi). Alt. 840 m. 3–5 August
1◦27′S, 78◦06′W. Alt. 1100 m. 20 February 1985, W. Palacios, 1948, H. Garcia Barriga 13188 (COL!); Km 68 del Ferrocarril
M. Baker and J. Zaruma 62 (RPSC!). Pichincha. El Chaupi, Tumaco-El Diviso. 28 July 1952, R. Romero Castañeda 3334
along the road to Iliniza. Alt. 3300 m. 19 April 1967, B. Sparre (COL!); Frontera Colombo-Ecuatoriana. Selva higrofila del Río
15645 (US!); Los Rios. Km 90, Camino Viejo via Chiriboga, San Miguel. Margenes del Río entre los afluentes Churruyaco
Quito-Santo Domingo. Alt. 1100 m. 7 April 1984, C.H. Dodson y Bermejal. Alt. 350–400 m. 12 December 1940, J. Cuatrecasas
and W. and M. Thurston 14173 (RPSC!); along the river just 11015 (COL!). Putumayo. Valle de Sibundoy. Alt. 2500–3000 m.
outside the town of Mindo on the new road to Liloa. Alt. 1963, C. Krauss 51 (COL!); Río Pepino, carretera a 10 km de
1300 m. C.H. Dodson, E. Hagsater, and A. Hirtz 16669 (RPSC!); Mocoa. Bosque alto. Alt. 850 m. 6 January 1957, M. Ospina H.
Km 40–51 on road Santo Domingo de los Colorados-Quito, 117a (COL!); Margenes del Río Guamues entre San Antonio
forested slopes along Río Pilaton, 0◦55′S, 78◦55′W. Alt. 1100– y la desembocadura, 20 December 1940, J. Cuatrecasas 11220
1400 m. 14 June 1973, L. Holm-Nielsen, S. Jeppensen, B. Lojtnant, (COL!); Vertiente oriental de la cordillera, entre Sachamates y
and B. Ollgaard 7154 (AMES!). Sucumbíos. Río San Miguel San Francisco de Sibundoy, Planada de Minchoy. Alt. 2100 m.
o Sucumbios, Santa Rosa y los alredadores. Alt. 380 m. 7– 30 December 1940, J. Cuatrecasas 11439 (F!, US!); Entre
8 April 1942, R.E. Schultes 3559 (COL!); Tungurahua. Río San Francisco y El Pepino. Alt. 1900–2400 m. 2 August 1961,
Verde Grande. Alt. 1500 m. 30 March 1956, E. Asplund 20049 A. Fernandez-Perez 5853 (COL!); Mpio Villa Garzón. Carretera
(AMES!); Baños-Puyo, km 35. 1◦24′S, 78◦12′W. Alt. 1170 m. a Puerto Asis. 1◦10′N, 76◦34′W. Alt. 1350 m. 3–4 May 1994,
11 February 1978, P. Bamps 6232 (MO!); Between Baños and J.L. Fernandez A., A. Camero, and E. Mesa 11467 (COL!,
Río Verde. Alt. 1680 m. 29 April 1951, P. R. Bell 812 (BM!); MO!); Río Pepino, carretera a 10 kms de Mocoa. Alt. 850 m.
Valley of Pastaza River, between Baños and Cashurco, 8 h east 6 January 1957, M. Ospina H. and J.M. Idrobo 117 (AMES!);
of Baños. Alt. 1300–1800 m. 25 September 1923, A.S. Hitchcock Mpio. Mistrató. Hacia San Antonio del Chami. Quebrada Sutu
21754 (AMES!; US!); Río Estancias, near Río Negro, southern y Empalados. Alt. 1700–1800 m. 26 April 1992, G.C. Lozano
Frontiers in Ecology and Evolution 25 frontiersin.org
120
fevo-10-1058334 January 6, 2023 Time: 11:34 # 26
Baranow et al. 10.3389/fevo.2022.1058334
and Estudiantes Introduccion Systematica 6382 (COL!); Cerro de March 1971. J. Schunke V. 4753 (B!, F!, US!); Mariscal Cáceres,
Portachuelo, entre Mocoa y Sacchamates. Alt. 1600–2000 m. 9 Tocache Nuevo. Camino al. Caserio de Santa Rosa de Mishollo,
December 1942, R.E. Schultes and C.E. Smith 3049 (COL!, K!, 4 km de Puerto Pizana, 20 May 1971, J. Schunke V. 4916 (F!); San
NY!, US!). Valle del Cauca. Costa del Pacifico, Río Cajambre, Roque. Alt. 1350–1500 m. January–February 1930, L. Williams
Barco. Alt. 5–80 m. 21–30 April 1944, J. Cuatrecasas 17250 7795 (F!).
(AMES!, COL!, F!); Chichito, Western Cordillera. Alt. 1600 m. When Sobralia lindeni was described in 1895 from
November 1937, E. Dryander 1994 (US!); Wooded cliffs of Río cultivated material upon which the description was
Dagua. Alt. 80–100 m. 6–8 May 1922, E.P. Killip 5057 (AMES!, based represented undoubtedly S. rosea. As a matter
NY!); Boca del Lobo, Buenaventura Bay. 9 June 1944, E.P. Killip of fact, because of the great similarity in the general
and J. Cuatrecasas 38985 (F!, US!); Km 80 Cali-Buenaventura. appearance of this species and S. pulcherrima the two
Alt. 350 m. 1 July 1965, C.H. Dodson and H. Hills 3215 (F!); have been combined in Lindenia in 1897 under S. lindeni
Cerca a la Elsa. Alt. 1250 m. 5 August 1966, S. Espinal T. as representing two distinct forms, those with white
1903 (AMES!, CUVC!); Queremal, Crece el Saludos. 20 January flowers and those with pale lilac or rose-colored flowers.
1980, I. Guarin O. 63 (COL!); New road Cali-Buenaventura, The white-flowered form is the true S. pulcherrima
La Pesuòa, 14 February 2011, D. Szlachetko, C. Uribe, and (Garay, 1978).
M. Moreno 9036 (UGDA-DLSz–spirit!). Venezuela. Táchira.
Between la Providencia and San Vicente de la Revancha,
southwest of Santa Ana. Alt. 1650 m. 8 January 1968, J.A. 3.2. Incertae sedis
Steyermark and G.C.K. and E. Dunsterville 100533 (AMES!).
Peru. Amazonas. Bagua Prov. Yamayakat bosque de Rivera. 3.2.1. Sobralia augusta Hoehne
4◦55′S, 78◦19′W. Alt. 320 m. 31 January 1996, N. Jaramillo, M. Arq. Bot. Estado São Paulo 1: 128. 1944. Type: Brazil.
Jaramillo, and D. Chamit 1024 (MO!); Bagua Distr. Aramango, Mato Grosso, Rio Juruena, Salto augusta, February 1912, F.C.
Soldado Oliva, 5◦18′S, 78◦20′W. Alt. 600 m. 6 February 1999, Hoehne 5349 (SP).
R. Vasquez, C. Vargas C., J. Yactayo, and E. Palomino 26046 The only material devoted to S. augusta that
(MO!); Central Cordilleras of the Andes. Alt. 2700–3300 m. 30 we could study is the drawing publisher in Flora
March 1938, L. Williams 7603 (AMES!, F!). Cusco. Marcapata. Brasilica (Vol. XII, Table 51). Based on the illustration,
Alt. 2000 m. 24 July 1957, C. Vargas C. 1168 (CUZ, F!); we can suspect, that taxon is a synonym of
Cardena. Alt. 1020 m. 29–30 July 1946, C. Vargas C. 6194 S. liliastrum. However, until the type material will
(F!); Maniri. Alt. 1200–1900 m. 8 December 1962, C. Vargas be available for analysis, we decide not to change its
C. 14064 (CUZ, F!). Huánuco. Cuchero, Sine loc. 1829, E.F. taxonomic status.
Poeppig s.n. (W! 47810); Bajando de Carpish a Tingo María.
Alt. 2700–2900 m. 5 March 1947, R. Ferreyra 1817 (AMES!).
Huánuco. Pampayacu. Hacienda at mouth of Chinchad Rio, Data availability statement
Alt 3500′, 19–25 July 1923, J.F. Macbride 5017 (F!). Junin.
Colonia Perené. Alt. 680 m. 30 March 1938, E.P. Killip and The raw data supporting the conclusions of this article will
A.C. Smith 24948 (AMES!; F!, US!); Satipo Prov. Gran Pajonal, be made available by the authors, without undue reservation.
Chequitavo, 10◦45′S, 74◦23′W. Alt. 1200 m. 27 March 1984,
D.N. Smith 6544 (MO!); Prov. Huánuco, Highway La Oroya–
Tingo María, km 66 east of Huánuco. Alt. 1620 m. 8 March 1977, Author contributions
J.D. Boeke 1167 (MO!). Oxapampa. Cueva Grande, Estacion
near Pozuzo. Alt. 3500′, 23 June 1923, J.F. Macbride 4804 PB: herbarium material revision, data gathering, analysis
(F!). San Martín. Boqueron Pass, 92 km from Tingo María on of distribution, manuscript writing, figures, and maps. DS:
highway to Pucallpa. Alt. 400 m. 16 December 1949–5 January herbarium material revision, data analysis, and manuscript
1950, H.A. Allard 2755 (US!); Tingo María. Alt. 625–1100 m. writing. PK: analysis of the results and manuscript writing and
30 October 1949–19 February 1950, H.A. Allard 22567 (US!); editing. All authors contributed to the article and approved the
Across Río Tocache from Tocache Nuevo, road to Juanjuí. Alt. submitted version.
500 m. 16 July 1982, A. Gentry, D. Smith and R. Tredwell
37640 (MO!); near Mayobamba. Alt. 1200–1600 m. March 1934,
G. Klug 3602 (AMES!, F!, K!, MO!); Prov. Rioja, Rioja, Salida a Funding
Mashoyacu-Shucaqai. Bosque protection Amto Mayo, Toma de
Agua, Quebrada Cuchachi, Alt. 1000 m. 15 July 1995, I. Sanchez This study presents part of the results of the project
Vega 8052 (F!); Prov. Mariscal Caceras Dtto. Tocacho Nuevo supported by a grant from the National Science Centre, Poland
(Muyuna de Huayrurillo) (margen derecha del Río Huallaga), 10 (2018/02/X/NZ8/00282).
Frontiers in Ecology and Evolution 26 frontiersin.org
121
fevo-10-1058334 January 6, 2023 Time: 11:34 # 27
Baranow et al. 10.3389/fevo.2022.1058334
Conflict of interest SUPPLEMENTARY MAP 1
Distribution map of Sobralia paradisiaca Rchb.f.
The authors declare that the research was conducted SUPPLEMENTARY MAP 2
Distribution map of Sobralia chrysantha Lindl.
in the absence of any commercial or financial relationships
SUPPLEMENTARY MAP 3
that could be construed as a potential conflict of Distribution map of Sobralia liliastrum Lindl.
interest.
SUPPLEMENTARY MAP 4
Distribution map of Sobralia elisabethae R. H. Schomb.
SUPPLEMENTARY MAP 5
Publisher’s note Distribution map of Sobralia granitica G.A. Romero & Carnevali.
SUPPLEMENTARY MAP 6
Distribution map of Sobralia gambitana Baranow, Szlach. & Kindlmann.
All claims expressed in this article are solely those of the
authors and do not necessarily represent those of their affiliated SUPPLEMENTARY MAP 7
Distribution map of Sobralia luerorum Dodson.
organizations, or those of the publisher, the editors and the
SUPPLEMENTARY MAP 8
reviewers. Any product that may be evaluated in this article, or Distribution map of Sobralia gloriosa Rchb. f.
claim that may be made by its manufacturer, is not guaranteed SUPPLEMENTARY MAP 9
or endorsed by the publisher. Distribution map of Sobralia ruckeri Linden & Rchb.f.
SUPPLEMENTARY MAP 10
Distribution map of Sobralia tamboana Dodson.
Supplementary material SUPPLEMENTARY MAP 11
Distribution map of Sobralia hoppii Schltr.
SUPPLEMENTARY MAP 12
The Supplementary Material for this article can be Distribution map of Sobralia pulcherrima Garay.
found online at: https://www.frontiersin.org/articles/10.3389/ SUPPLEMENTARY MAP 13
fevo.2022.1058334/full#supplementary-material Distribution map of Sobralia rosea Poepp. & Endl.
References
Bachman, S., Moat, J., Hill, A., de la Torre, J., and Scott, B. (2011). IUCN (2022). The IUCN red list of threatened species. Version 2022-
Supporting red list threat assessments with GeoCAT: Geospatial conservation 1. Available online at: https://www.iucnredlist.org (accessed November 20,
assessment tool. ZooKeys 150, 117–126. doi: 10.3897/zookeys.150. 2022).
2109 Kolanowska, M., Tsiftsis, S., Dudek, M., Konowalik, K., and Baranow, P. (2022).
Baranow, P. (2015). Taxonomic notes on Sobralia section abbreviatae Niche conservatism and evolution of climatic tolerance in the neotropical orchid
(Orchidaceae) in colombia, with description of a new species. Plant Syst. Evol. 301, genera Sobralia and Brasolia (Orchidaceae). Sci. Rep. 12:13936. doi: 10.1038/
41–60. doi: 10.1007/s00606-014-1053-5 s41598-022-18218-4
Baranow, P., and Szlachetko, D. L. (2016). The taxonomic revision of Lindley, J. (1854). Folia orchidacea, sobralia. London: J. Matthews.
the Sobralia ruiz & pay. (Orchidaceae) in the guyanas (Guyana, Suriname,
French Guiana). Plant Syst. Evol. 302, 333–355. doi: 10.1007/s00606-015- Neubig, K. M., Whitten, W. M., Blanco, M. A., Endara, L., Williams, N. H., and
1266-2 Koehler, S. (2011). Preliminary molecular phylogenetics of Sobralia and relatives
(Orchidaceae: Sobralieae). Lankesteriana 11, 307–317. doi: 10.15517/lank.v11i3.
Baranow, P., Dudek, M., and Szlachetko, D. L. (2017). Brasolia, a new genus 18286
highlighted from Sobralia (Orchidaceae). Plant Syst. Evol. 303, 853–871. doi: 10.
1007/s00606-017-1413-z Pridgeon, A. M., Cribb, P., Chase, M. W., and Rasmussen, F. N. (2006). Genera
orchidacearum, epidendroideae, Vol. 4, Part 1. Oxford: Oxford University Press.
Baranow, P., Rojek, J., Dudek, M., Szlachetko, D., Bohdanowicz, J., Kapusta,
M., et al. (2022). Chromosome number and genome size evolution in brasolia Reichenbach, H. G. (1873). Xenia orchidacea, Vol. 2. Leipzig: F. A. Brockhaus.
and sobralia (Sobralieae, Orchidaceae). Int. J. Mol. J. Sci. 23:3948. doi: 10.3390/ Romero-González, G. A. (2003). The Sobralia liliastrum (Orchidaceae:
ijms23073948 Sobraliinae) complex in the venezuelan guyana. Harvard Pap. Bot. 8, 115–129.
Brieger, F. G. (1983). “Subtribus sobraliinae,” in Die orchideen I, Vol. 13, eds F. G. Schlechter, R. (1920). Die orchideenfloren der südamerikanischen
Brieger, R. Maatsch, and K. Senghas (Lieferung: Verlag Paul Parey), 780–800. kordillerenstaaten II. Colombia (II. Beschreibungen neuer arten). Repert.
Dodson, C. H. (1998). New orchid species and combinations from ecuador, Spec. Nov. Regni Veg. Beih. 7, 37–161.
fascicle 6. Orquideología 21, 3–60. Schlechter, R. (1924). Beiträge zur orchideenkunde von Colombia. I.
Dressler, R. (2002). The major sections or groups within Sobralia, with four new orchidaceae hoppianae. Repert. Spec. Nov. Regni Veg. Beih. 27, 5–123.
species from Panama and Costa Rica, S. crispissima, S. gloriana, S. mariannae and
S. nutans. Lankesteriana. 5, 9–15. doi: 10.15517/lank.v2i3.23088 Schlechter, R. (1929). Figuren-Atlas zu den orchideenfloren dersüdamerikanischen kordillerenstaaten. Repert. Spec. Nov. Regni Veg. Beih.
Dressler, R. L., Blanco, M. A., Pupulin, F., and Neubig, K. M. (2011). Proposal 57, 1–142.
to conserve the name Sobralia (Orchidaceae) with a conserved type. Taxon 60,
907–908. doi: 10.1002/tax.603030 Szlachetko, D. L., Kolanowska, M., Baranow, P., and Dudek, M. (2020).Materials to the orchid flora of colombia, orchidaceae, spiranthoideae –
Dunsterville, G. C. K., and Garay, L. A. (1959). Venezuelan orchids illustrated 1. Cranichideae, vanilloideae, Vol. 3. Glashütten: Koeltz Botanical Books.
London: Andre Deutsch. Thiers, B. (2022). Index herbariorum: A global directory of public herbaria
Garay, L. A. (1978). “Orchidaceae (cypripedioideae, orchidoideae, and associated staff. New York Botanical Garden’s Virtual Herbarium. Available
neottioideae),” in Flora of ecuador 9, eds G. Harling and B. Sparre (Göteborg: online at: http://sweetgum.nybg.org/science/ih/; www.naturalearthdata.com;
University of Göteborg and Swedish Museum of Natural History), 1–305. www.qgis.org/pl/
Frontiers in Ecology and Evolution 27 frontiersin.org
122
fevo-11-1135316 February 8, 2023 Time: 15:42 # 1
TYPE Original Research
PUBLISHED 14 February 2023
DOI 10.3389/fevo.2023.1135316
The effect of habitat
OPEN ACCESS transformation on a twig epiphytic
EDITED BY
Dennis Whigham, orchid: Evidence from population
Smithsonian Institution, United States
REVIEWED BY
Vladan Djordjević, dynamics
University of Belgrade, Serbia
Jonas Morales-Linares,
Benemérita Universidad Autónoma de Puebla, Nhora Helena Ospina-Calderón 1,2*, Raymond L. Tremblay 3,
Mexico Alba Marina Torres 2 and Nicola S. Flanagan 1
*CORRESPONDENCE
Nhora Helena Ospina-Calderón 1Departamento de Ciencias Naturales y Matemáticas, Pontificia Universidad Javeriana Seccional Cali, Cali,
nhora.ospina@correounivalle.edu.co Colombia, 2Posgrado en Ciencias-Biología, Universidad del Valle, Cali, Colombia, 3Department of Biology,
University of Puerto Rico–Humacao Campus, Humacao, PR, United States
SPECIALTY SECTION
This article was submitted to
Conservation and Restoration Ecology,
a section of the journal The tropical Andean landscape has been dramatically transformed over the
Frontiers in Ecology and Evolution last century with remaining native forest limited to small fragments within a
RECEIVED 31 December 2022 heterogeneous matrix of crops, cattle pastures, and urban environments. We aimed
ACCEPTED 27 January 2023
PUBLISHED 14 February 2023 to explore the impact of habitat transformation on the population dynamics in an
CITATION endemic twig epiphytic orchid located within the undisturbed forest and within
Ospina-Calderón NH, Tremblay RL, Torres AM modified matrix habitat in two regions with contrasting landscape structures: with
and Flanagan NS (2023) The effect of habitat
transformation on a twig epiphytic orchid: a dominant shade coffee matrix and a dominant grassland matrix. Over 2 years,
Evidence from population dynamics. we surveyed 4,650 individuals of the Colombian endemic orchid, Rodriguezia
Front. Ecol. Evol. 11:1135316. granadensis. We undertook four post-breeding censuses in three sites in each
doi: 10.3389/fevo.2023.1135316
region in both native forest and pasture sub-sites (12 sub-sites; 48 censuses in
COPYRIGHT
© 2023 Ospina-Calderón, Tremblay, Torres and total), and constructed demographic transition matrices (n = 36). The transition
Flanagan. This is an open-access article probabilities were calculated using a Bayesian approach and population grow rates
distributed under the terms of the Creative
Commons Attribution License (CC BY). The use, were evaluated using asymptotic models and elasticities using transient dynamics.
distribution or reproduction in other forums is Between regions, higher population growth rate and inertia (defined as the largest
permitted, provided the original author(s) and
the copyright owner(s) are credited and that the or smallest long-term population density with the same initial density distribution)
original publication in this journal is cited, in was seen in the shade coffee-dominated landscape. Additionally, population growth
accordance with accepted academic practice. rate and damping ratio was higher in forest compared with pasture, with lower
No use, distribution or reproduction is
permitted which does not comply with convergence time for the forest subsites. These demographic patterns reveal the
these terms. contrasting levels of population resilience of this orchid in different landscape
structures with the more connected shade-coffee dominated landscape permitting
some healthier populations with greater population growth and survival in forest than
pasture. This study highlights that twig epiphyte colonization of isolated phorophytes
in pastures should not be interpreted as a sign of a healthy population but as a
temporal transitory period.
KEYWORDS
demography, landscape, matrix models, resiliency, Rodriguezia granadensis, tropical Andes,
reproductive success, PPM
Introduction
Habitat fragmentation threatens the survival of populations and species in two main
ways. Firstly, smaller, isolated populations in habitats with high fragmentation are more
vulnerable to stochastic events (Fischer and Lindenmayer, 2007). These may be the result
of environmental catastrophes, particularly in the context of increasingly extreme climatic
events, random genetic processes, with the loss of evolutionary potential through genetic
Frontiers in Ecology and Evolution 10213 frontiersin.org
fevo-11-1135316 February 8, 2023 Time: 15:42 # 2
Ospina-Calderón et al. 10.3389/fevo.2023.1135316
drift (Lienert, 2004; Honnay and Jacquemyn, 2007) or demographic terrestrial. Those species that are exclusively epiphytic are often
processes, with year-to-year variability in reproductive success limited to a particular ecological niche in a restricted zone of the
(Tomimatsu and Ohara, 2010; Jacquemyn et al., 2012). Secondly, with architecture of the tree (Catling et al., 1986; Medeiros, 2010). The
the reduction in area of natural habitat, the abiotic conditions of so-called twig epiphytes use as their substrate the smallest branch
the surrounding landscape change, with consequent negative impact size, most often located in the outer fringe of the tree canopy
on plant reproduction (Aguilar et al., 2006; Aguilar et al., 2019) and (Ventre-Lespiaucq et al., 2017). Obligate twig epiphytes often are
biotic interactions (Brosi, 2009; Briggs et al., 2013). characterized by their accelerated life cycle, psigmoid or terete leaves,
The impact of habitat transformation (Ritchie and Roser, and thickened seed testa (Chase, 1987; Zotz, 2007).
2013; Winkler et al., 2021) on ecological characteristics of species Orchids specialized as twig epiphytes, while numerous, are
has been well-studied from diverse perspectives, such as life phylogenetically restricted to Oncidiinae and Vandeae clade (Chase,
histories characteristics (Kolb and Diekmann, 2005; Bruna et al., 1987; Gravendeel et al., 2004). A limited amount of information of
2009), extinction probabilities (Fréville et al., 2007), plant animal the life history of these species is available [Tolumnia variegata (Sw.)
interactions (Benítez-Malvido and Arroyo-Rodríguez, 2008; Benitez- Braem, Calvo and Horvitz, 1990; Ackerman et al., 1996; Erycina
Malvido et al., 2016) and reproductive success (Brudvig et al., 2015; crista-galli (Rchb.f.) N.H.Williams and M.W.Chase, Mondragón
Vellend et al., 2017). et al., 2007; Ionopsis utricularioides (Sw.) Lindl., García-González and
Following the theory of island biogeography, fragments of native Riverón-Giró, 2013] and most of the supposed advantages of being
habitat can be considered as islands within a “sea” of transformed a twig epiphyte are circumstantial. It is commonly assumed that the
terrain. In a mosaic of patches of different land use, the dominant, advantage in being in the outer rim of the canopy twig epiphytes is
usually non-native background in a transformed landscape is known the higher availability of light. However, this advantage may have
as the matrix (Fischer and Lindenmayer, 2007). Spatial matrix types, tradeoffs (Ventre-Lespiaucq et al., 2017) including a greater risk of
such as cattle pasture, different agricultural systems, or urbanization, dehydration (Chase, 1987). A study in T. variegata found that plants
have differential impacts on the connectivity between native habitat located on twigs at the canopy edge had a reproductive disadvantage
fragments, altering resource availability, as well as the activity of compared with those located within the tree canopy (Tremblay et al.,
pollinators, seed dispersers, and herbivores (Jules and Shahani, 2003; 2010). Increased light availability may result in higher reproductive
Debinski, 2006). potential (flower production) but could also result in lower survival of
For example, fragmentation and loss of habitat quality affect the smaller individuals (for example because of desiccation), resulting
pollinator communities, including the so-called “orchid bees” in an overall decrease in the long-term persistence of the population.
(Apidae: Euglossini), impacting home ranges (Brosi, 2009) and Twig epiphytic orchids are often transitory pioneer species and
reproductive success (Newman et al., 2013). Small, isolated plant frequently colonize phorophytes in transformed habitats. Given
populations are expected to have lower reproductive success when the tolerance of twig epiphytes to higher light intensities (Ventre-
dependent on non-resident pollinators (Murren, 2002). Lespiaucq et al., 2017), such populations on trees in transformed,
The tropical Andes represent a hotspot of biodiversity (Myers open environment may be perceived to be as healthy as those
et al., 2000; Liang et al., 2022) and endemism (Gentry, 1982; in undisturbed forest habitat. However, this perspective may be
Olson and Dinerstein, 1998) and at the same time is one of the misleading as the number of individuals may be temporary.
geographical areas with the highest rate of anthropogenic habitat Our study aims to evaluate the impact of habitat transformation
transformation (Etter and van Wyngaarden, 2000; Etter et al., 2006). on the twig epiphytic species, Rodriguezia granadensis (Lindl) Rchb.f.
These anthropogenic changes can influence demographic dynamics This orchid is commonly distributed across Andean premontane
(Rodríguez-Echeverry and Leiton, 2021), survivorship or persistence and montane forests. Although endemic to Colombia, its natural
of populations (Philpott et al., 2008). Nonetheless, our understanding tendency to colonize isolated phorophytes in open pastures is a
of the impact of habitat transformation in this biodiverse region major contributing factor to its classification of least concern (LC) in
is limited (Hoang and Kanemoto, 2021; Winkler et al., 2021). In national red-list evaluations (Calderón-Sáenz, 2007; López-Gallego
the neotropics, the influence of modified landscape mosaics on the and Morales, 2021).
diversity of birds, bats (Harvey and González Villalobos, 2007), We used an approach which includes the complete life history of
insects (Vandermeer et al., 2019), and trees (Philpott et al., 2008) have the species, following individuals in a mark-recapture approach and
been documented, however there are only a few studies focused on population projection matrices (PPM), comparing the demographic
epiphytes (Richards et al., 2020), including epiphytic orchids (García- structure and dynamics in native forest fragments and on isolated
González and Riverón-Giró, 2013; Raventós et al., 2018), with only phorophytes in pastures across two contrasting landscapes. We aim
one study in the Andes (Parra Sánchez et al., 2016). to understand the potential different demographic responses in each
Epiphytic plants grow on the trunk, branches, twigs, and even the landscape and land use, thereby drawing inferences on the dynamics
leaves (Alvarenga and Pôrto, 2007) of a plant host, the phorophyte, of orchid twig epiphyte populations in varying anthropogenically
enabling growth in higher light conditions. Vascular epiphytes modified environments. We hope our findings may inform landscape
are one of the most dominant guilds of species in the tropics management practices to promote orchid conservation in this
and are potentially highly impacted and endangered by habitat biodiverse region.
transformation (Hernández-Pérez and Solano, 2015; Osie et al., As a null model, we would expect no differences in population
2022). The distribution and survival of epiphytes is influenced by the dynamics between the two landscapes, a matrix dominated by either
landscape structure, phorophyte diversity, the age of the forest and coffee crop or sugar cane and cattle grassland, nor between native
tree size (Hietz, 1999). forest and pastureland cover populations. If the main driver for the
The main plant families with epiphytic species are Bromeliaceae niche occupancy of twig epiphytes in the outer tree canopy is to
and Orchidaceae (Zotz, 2013). Some epiphytic species can also be maximize exposure to light, it could be expected that populations on
rupiculous, growing on rock substrate, while others may also be isolated trees in an open environment such as pastures would present
Frontiers in Ecology and Evolution 10224 frontiersin.org
fevo-11-1135316 February 8, 2023 Time: 15:42 # 3
Ospina-Calderón et al. 10.3389/fevo.2023.1135316
more favorable demographic parameters, with a higher population Puertas-Orozco et al., 2011). Our three field sites were (1) Hondonada
growth rate and stability. Specifically, we aim to determine whether (H) (3◦ 49.896′ N, 76◦ 26.043′ W), (2) Lilas (L) (3◦ 50.986′ N, 76◦
population dynamics of twig epiphytes are similar in native forest and 26.344′ W) and (3) the National Forest Reserve of Yotoco (Y) (3◦
pastures in the two regions across 2 years of survey considering the 52.712′ N, 76◦ 26.291′ W). The three field sites within each region
following parameters (1) deterministic population growth rate, (2) had pairwise geographic distances between 5 and 15 km (Figure 1).
transient dynamics and (3) non-linear elasticity (transfer function)
of the different life stages, (4) reproductive potential (fruit set), and
(5) recruitment. Survey
At each of the six field sites we surveyed plants of R. granadensis
Materials and methods in two sub-sites of contrasting land cover: native forest (continuous
canopy) and pasture (grassland with isolated trees), for a total of
Study species 12 sampled sub-sites. From here on, we refer to three differentanalysis levels: Region comparing Cauca with Valle, Site Calibío (Cl),
Cajibío (Cj), Piendamó (Pi) in Cauca, and Hondonada (H), Lilas
Rodriguezia granadensis (Lindl.) Rchb.f. is widely distributed at (Li), and Yotoco (Y) in Valle and Sub-site, contrasting landcover,
mid-elevation (700–1,900 m.a.s.l.) in Andean Forest. This orchid forest or pasture. Within sub-sites all individual orchids present
frequently colonizes coffee or fruit tree plantations. The species is in each phorophyte (host tree) were marked and counted until
common and widely distributed and consequently an excellent model reaching 300 at the first census. The position of each phorophyte
species to study the impact of changing landscapes in the northern sampled was registered with a GPS Global Positional System (Garmin
Andes on epiphytic population dynamics (Ventre-Lespiaucq et al., Oregon 750), and the minimum convex polygon for each sub-site was
2017). calculated to report phorophyte distance and density (QGIS 3.26).
Rodriguezia granadensis has two flowering seasons a year Individual orchid plants were marked with permanent Dymo tags for
(March–April and October–November), which coincide with monitoring over consecutive censuses.
bimodal peaks of rainfall (Calderón-Sáenz, 2007). It is pollinated While density per meter square is a common metric used to
by euglossine bee species–Eulaema meriana Oliver, E. cingulata describe the dispersion pattern of many plant species, it is not always
Fabricius, and Exaerete frontalis Guérin-Méneville–that forage for an adequate description of the dispersion pattern of epiphytic orchids
nectar in a melitophilly syndrome behavior (Ospina-Calderón et al., (Tremblay, 1997). Because epiphytic orchids are dependent on the
2015). presence of the host tree, measuring density per host tree is a more
realistic index, and we took both of these variables into account.
We surveyed plants at four different times, resulting in three
Study sites transition matrices between consecutive post-breeding censuses for
each site. In the first census in March 2017, we tagged the first plants
Populations of R. granadensis were studied in three field sites in each sub-site. For the subsequent three censuses (Oct. 2017, March
each in two regions of the tropical Andes in Colombia, in the 2018, and Oct. 2018) when additional individuals were detected, these
departments of Cauca, and Valle del Cauca (from here on “Valle”). were tagged too and included in the population analyses. Thus, an
These two regions, separated by approximately 150 km, are located at additional 30 to 50 plants were registered per sub-site per survey for
the same elevation (approximately 1,700 m.a.s.l.) on the eastern slope a total of between 348 and 440 plants per sub-site and a grand total of
of western cordillera of the Andes (Figure 1). 4,650 unique individual plants in the study (Supplementary Table 1).
The Cauca region is in the Colombian massif of the Popayán For each plant we registered the number of live pseudobulbs,
plateau with sun and shade coffee crops in an agroforestry mosaic, inflorescences, flowers, and fruits as an index of reproductive
with mixed and forestry crops, small fragments of forest and potential in addition to survival among time periods. The
riparian forests (Criollo and Bastidas, 2011; Arenas-Clavijo and reproductive potential for a specific stage was estimated as the
Armbrecht, 2018). Although the coffee landscape is increasing in number of fruits/number of flowers in the time period (Sabat and
agricultural intensity, to the detriment of biodiversity (Armbrecht, Ackerman, 1996). The expected number of recruits at time t is
2003; Philpott et al., 2008; Harvey et al., 2021), it continues to host assumed to be proportional to the fruit set at time t-1, consequently,
more diversity as an agroecosystem (Letourneau et al., 2011) than recruitment does not include the seed stage or dormancy of seeds
extensive monocultures such as the sugar cane and cattle ranching (Tremblay and Hutchings, 2002).
model in the Valle del Cauca department to the north (Marull et al.,
2018; Sardi et al., 2018). In the latter, we find a few isolated forests in a
predominantly pasture matrix, where the landscape and biodiversity Data analysis
has been dramatically affected (Torres et al., 2012; Vélez-Torres et al.,
2019). We reviewed the distribution of all the demographic and
The southern region in Cauca, has a mean annual precipitation reproductive variables per region, site, and subsite. After conducting
of ± 2,120 mm, and temperature of 15◦C (IDEAM, 2010; Puertas- assumption tests with Shapiro Wilks and without transformation,
Orozco et al., 2011). The three field sites in this region were: (1) we ran an analysis of variance ANOVA to test for differences in
Calibío (Cl) (2◦ 37.446′ N, 76◦ 33.525′W); (2) Cajibío (Cj) (2◦ 38.888′ demography and reproductive variables between region and site. For
N, 76◦ 32.328′ W); and (3) Piendamó (Pi) (2◦ 41.126′ N, 76◦ 33.710′ subsites, we ran a paired t-test.
W). The region to the north, in Valle del Cauca, has a mean annual For the population projection matrices (PPM), we applied the
precipitation ± 1,480 mm and temperature of 18◦C (IDEAM, 2010; life cycle structure previously determined for this species, based on
Frontiers in Ecology and Evolution 10235 frontiersin.org
fevo-11-1135316 February 8, 2023 Time: 15:42 # 4
Ospina-Calderón et al. 10.3389/fevo.2023.1135316
FIGURE 1
Geographical location study sites and experimental design showing sampling organization over three scales, Region: Cauca and Valle. Site: Cl, Calibío;
Cj, Cajibío; Pi, Piendamó; H, Hondonada; Li, Lilas; Y, Yotoco. Sub-site: F, forest; P, pasture.
pseudobulb and inflorescence number (Ospina-Calderón, 2009), and specific stages (Tremblay et al., 2021). For the Bayesian analysis, prior
the methodology for PPM developed for orchids, following Tremblay data for the matrix (Table 1) were selected from a previous census
and Hutchings (2002) and Martorell et al. (2022). Our simplified undertaken in the population of Yotoco in 2008 (Ospina-Calderón,
life history of R. granadensis is based on four size classes describing 2009) with an effective sample size of n = 1. Consequently, this weak
the life stages: (1) Seedling (S), individuals lacking pseudobulbs; (2) prior has little impact on the transition probabilities when sample
Juvenile (J) with 1–2 pseudobulbs; (3) Small adults–stage 1 (A1), sizes are large. This choice of an effective sample size yields posterior
possessing 3–6 pseudobulbs and no more than one inflorescence; parameters that are dominated by the data.
and 4. Large adults–stage 2 (A2) for plants possessing more than 7
pseudobulbs with one or more inflorescences. Transitions between
life stages from one census to the next were recorded as growth (G); Population growth rates
fecundity (Fe), stasis within the same life stage (L); and reversal (R)
(Figure 2). We employed population projection matrix (PPM) analysis to
evaluate the asymptotic populations growth rate, lambda. When
Estimating transition probabilities lambda is equal to one (including the credible intervals, CrI)
populations are considered to be stable, while lambda values either
smaller or larger than one (with the CrI) indicate a decreasing
From the data registered during the four censuses (March and
or increasing population size, respectively. The median population
October, 2017 and 2018) we calculated parameters for 36 transition
growth rate and the CrI were calculated with 15,000 simulations
matrices (time period x sub-sites). Each matrix corresponds to a time
corresponding to the posterior lambda values and the CrI (Tremblay
period: Time 1–March to October 2017; Time 2–October 2017 to
March 2018; Time 3–March to October 2018. Thus, three matrices for et al., 2021).
each of the 12 sub-sites were constructed. The transition probabilities
were estimated using a Bayesian approach (Tremblay et al., 2021).
This analysis is more appropriate for the current dataset for two Transient dynamics, transfer function
reasons. Firstly, it resolves issues for estimating the parameters of
some transitions with small sample sizes (for example, seedlings were Transient dynamics analysis and the indices described by Stott
scarce or not detected in some populations). Secondly, the parameter et al. (2012a) are mathematical approaches to study the short-term
estimates (transitions, survival, death, and stasis) follow the required effect of ecological disturbances or perturbation on the population
beta distribution and the credible intervals are bounded between 0 structure of a species in addition to understanding the impact of
and 1. With this Bayesian approach infrequent transitions can be population structure not at equilibrium (Stott et al., 2011). This
estimated while avoiding improbable values that may be generated innovative approach to understanding short–term dynamics has been
with small sample sizes or few observed transitions for some of the applied broadly in plants (McDonald et al., 2016), for example, plant
Frontiers in Ecology and Evolution 10246 frontiersin.org
fevo-11-1135316 February 8, 2023 Time: 15:42 # 5
Ospina-Calderón et al. 10.3389/fevo.2023.1135316
FIGURE 2
Life cycle diagram of Rodriguezia granadensis. S, seedling; J, juvenile; A1, adult, stage 1; A2, adult, stage 2. Transitions: G, growth; F, fecundity; L, stasis in
the same stage; R, reverse.
invasions (Iles et al., 2016) and orchids (Raventós et al., 2015; Ortiz- TABLE 1 Priors for the transition matrix for estimating the posterior
Rodríguez et al., 2020). Transient dynamics PPM models are time- transition probabilities.
invariant, however by varying the starting demographic distribution, Stages S J A1 A2
and modeling demographic stochasticity whether of biotic, abiotic,
and anthropogenic origin, transient dynamics may result in a stage S 0.30 0 0 0
distribution that differs from the stable stage distribution. J 0.09 0.35 0.001 0
The different starting scenarios lead to either a short-term A1 0.01 0.10 0.60 0.06
increase in population size and density (amplifications) or a short- A2 0 0.05 0.07 0.84
term decrease (attenuation). If no other perturbations or disturbances
are present, then the transient dynamics models are expected to Data from a previous study on Rodriguezia granadensis and Yotoco population (Ospina-Calderón, 2009). S, seedling; J, juvenile; A1, adult, stage 1; A2, adult, stage 2.
stabilize to the stable stage distribution. The time to reach the
stable stage is the transient period (Stott et al., 2011). One of
the most useful measurements of transient population density Results
and growth are reactivity and inertia. These indices and their
bounds describe the majority of variation in transient population We surveyed a total of 4,650 plants, in 12 sub-sites of
density with biological interpretations because they describe shorts both native forest and pasture land cover sub-sites in three
term changes (Stott et al., 2011). In general, orchid populations sites each in two regions with differing landscape composition:
are not at stable stage distribution (Schödelbauerová et al., 2010; shade-coffee dominated, and pasture-dominated (Supplementary
Tremblay et al., 2015), however a comprehensive review is Table 1). A total of 1,636 individuals died across the survey period
still lacking. (Supplementary Table 2).
Transfer function is an approach for evaluating the non-linear
effect of perturbation on population dynamics. The traditional
approach has been to evaluate the elasticities of the parameters of Spatial distribution and fruit set
the matrix (Caswell, 2000), with the limitation that elasticities are
assumed to be linear and consequently are usually more applicable The spatial distribution of R. granadensis plants and phorophytes
when perturbation is small (Stott et al., 2012a). The advantage of varied among landcover sub-sites in a similar way in both regions.
transfer function analysis is that it can elucidate the possible impact In the native forest populations, plants were found over areas from
across a wider range of perturbation without assuming that the 1,830 to 6,218 m2, while the number of phorophytes varied from
response is linear. nine to 96. The number of orchids per phorophyte varied between
four and 10 in each forest sub-site. In the pasture sub-sites in both
regions, the distribution areas were half to five times less (582 to
Software 1,127 m2) with a range of nine to 35 phorophytes and between 10
and 46 individuals per phorophyte (Supplementary Table 1). Thus,
All the analysis was performed in the R 4.2.0 environment. population density was 2 ind./m2 (sd = 0.51) in forest sub-sites,
The PPM parameters based on a Bayesian approach were evaluated and 10 ind./m2 (sd = 5.16) in pasture sub-sites. A more aggregated
using the raretrans package (Tremblay et al., 2021). The asymptotic distribution in the isolated phorophytes in the pasture matrix was
population growth rates, transient dynamics, and transfer function observed. The number of plants per phorophyte were significantly
were attained using the popdemo R package (Stott et al., 2012b) lower in forest sub-sites with fewer plants per tree in forest (6.93,
using the posterior matrices. Data were visualized, contrasted, and ANOVA sd = 2.22, p = 0.01) than in pastures (25.67, sd = 13.95).
wrangled using the ggplot2 and tidyverse packages (Wickham, 2016; The average fruit set was 0.055 fruits/flowers (sd = 0.035) for
Wickham et al., 2019). all sub-sites and seasons (Supplementary Table 3). However, over
Frontiers in Ecology and Evolution 10257 frontiersin.org
fevo-11-1135316 February 8, 2023 Time: 15:42 # 6
Ospina-Calderón et al. 10.3389/fevo.2023.1135316
TABLE 2 Summary of Rodriguezia granadensis fruit set for forest (F) and showing decline and tendency to extinction in 5 to 10 years (Figure 4;
pasture (P) sub-sites in two regions, Cauca (C) and Valle (V), Colombia. Supplementary Figure 1). Additionally, higher values for inertia,
Region Sub-site Fruit set Sd Mean Number Number reactivity and damping ratio indicate greater resilience for forest than
number of fruits of pasture sub-sites (Supplementary Table 5).
of fruits flowers The upper value for inertia upper was for the forest subsites in the
C F 0.0733 0.0464 18.58 223 3151 sites Calibío, Cl (Cauca) and Hondonada, H (Valle) and for inertia
low, the lowest was Calibío pasture and Lilas, Li (Valle) pasture.
V F 0.0625 0.0362 13.75 165 2861
Reactivity confirms this pattern of more resilient plants in the forest
C P 0.0383 0.0248 12.50 150 4733 sub-sites, with a higher register for Calibío forest and lower for Lilas
V P 0.0433 0.0235 15.58 187 3841 (Valle) pasture (Supplementary Table 5).
Mean and total number of fruits over four censuses in 3 sub-sites of each land cover type in each Transient population dynamics simulation for 50 flowering
region. Sd, standard deviation. seasons (25 years) revealed a greater tendency for populations to
decline and possible extinction in 5 to 10 years in pasture compared
both regions, fruit set was significantly greater for forest sub-sites with forest populations (Figure 5; Supplementary Figure 2). In four
compared with pasture sub-sites (Forest mean = 0.067, sd = 0.041; of the six pasture sub-sites over both regions, simulations indicated
Pasture mean = 0.040, ANOVA sd = 0.023, p = 0.01) (Table 2). The probable population decline tending to extinction in 5 to 10 years
exception was in the Hondonada pasture (Valle), which presented a (Supplementary Figure 2). In contrast, for the forest sub-sites, these
greater number of fruits than the forest (Supplementary Table 4). simulations suggested population growth for four of the six sub-sites,
with only Cajibío (Cauca) and Lilas (Valle) indicating a decrease in
population size.
Population projection matrices (PPM),
asymptotic population grow rate Perturbation analysis
A total of 36 transition matrices representing three time periods Perturbation analysis using the non-linear elasticities approach,
for each of the 12 subsites were constructed: three subsites each of transfer function, revealed a non-linear relation on relative
forest or pasture within each of the two regions Cauca and Valle. The importance of the influence of perturbation for each stage
most common transition detected for all stages was for stasis (L), with (Supplementary Figure 3). It is evident that perturbation results in
plants remaining in the same stage through at least two consecutive non-linear response of population growth rate as a function of the
censuses (Figure 2). Over both regions, in forest sub-sites the most amount of perturbation in almost all of the parameters. This is most
common transition was for the Adult 1 stasis, L33, and in pasture evident in the stasis stage (the diagonal of the matrix) where most
sub-sites for Adult 2 stasis, L44 (Supplementary Table 2). have a narrow peak with a rapid decrease and increase around an
Overall, the intrinsic population growth rates (lambda) in all 36 optimum. Increases in reproductive success (fruits/flowers) show a
matrices ranged from a minimum of 0.742 to a maximum of 1.268 near linear response in almost all cases. While transitions to the
(Figure 3; Supplementary Table 5). A striking difference was seen next life stage results in a “U” shape response in some cases, the
between the forest and pasture subsites over both regions. In the pattern is inconsistent across sites and time periods showing how the
forest sub-sites 12 of the 18 PPM yielded a lambda greater than one population is likely to respond if there is an increase or decrease in
(increasing population), with two less than one. In contrast, in the the parameters and how that would affect growth rate.
pasture subsites, 12 PPM yielded a lambda less than one, with two
being greater than one. The distribution of population reduction,
stability and growth was not equal among the forest sub-sites (Fisher’s Discussion
exact test = 12.42, df = 2, p = 0.002), however it was independent of
regions, although forest populations may be of slightly better health This is the first study to compare the demographic patterns
in both Cauca and Valle. of a twig epiphytic orchid between populations in native and
Among sub-sites and time periods, variation was seen in the transformed habitat matrices in the tropical Andes. Our findings of
population growth rates, from reductions of close to 30% to increases less favorable population dynamics for R. granadensis in transformed
of 25%. The sub-site with the largest population reduction was at compared with forest land covers, and between two contrasting
Cajibío pasture, in Cauca (Time 2, lambda 0.742; 95% CrI 0.667– landscape structures has important implications for the evaluation
0.819). The two sub-sites which had the largest increase were in forest, of the conservation status of this species and will inform landscape
in Valle, Hondonada (Time 1, lambda 1.268, 95%CrI 1.187–1.349) management practices to promote conservation of other similar twig
and at Cauca, Calibío (Time 1, lambda 1.199; 95%CrI 1.147–1.253). epiphytic orchids in this biodiverse region.
Previous studies of tropical epiphytic communities have shown
that species diversity and abundance decrease over gradients of
Transient dynamics increasing human impact (Larrea and Werner, 2010; Hylander and
Nemomissa, 2017). However, in the present study, the populations
The transient dynamics indices revealed that the convergence of R. granadensis colonizing isolated fruit or shade trees within
times to stable structure tended to be smaller in forest than in pasture a transformed pasture matrix had a higher density of plants per
sub-sites for all sites, with the exception of Cajibío in the coffee- phorophyte compared with the forest sub-sites (Supplement Table 1).
dominated landscape in Cauca. The shadow diagram confirms that This may be partly explained as isolated trees in open pastures tend
Cj Forest is more likely to grow than pasture, with darker zones to grow larger and wider crowns (Elias et al., 2021).
Frontiers in Ecology and Evolution 10268 frontiersin.org
fevo-11-1135316 February 8, 2023 Time: 15:42 # 7
Ospina-Calderón et al. 10.3389/fevo.2023.1135316
FIGURE 3
Median posterior asymptotic population growth rate and 95% credible intervals estimated for Rodriguezia granadensis population at two subsites, Forest
and Pasture in six Sites (Cauca sites: Cl, Calibío; Cj, Cajibío; Pi, Piendamó. Valle sites: H, Hondonada; Li, Lilas; Y, Yotoco), over three time periods; Time 1,
Mar.–Oct. 2017; Time 2, Mar. 2017–Oct. 2018; Time 3, Mar.–Oct. 2018. Red are populations/time periods where lambda was significantly smaller than 1,
green for lambda equal to one (not significantly different from stability), and blue for lambda significantly larger than one.
In Andean human-transformed landscape, Köster et al. (2011) Population growth simulation across multiple time periods suggests
found that tree traits explain 60% of the epiphytic community that the forest populations are less likely to go extinct as compared to
composition in an Ecuadorian cloud forest, where the isolated trees pasture sub-sites. The likelihood of extinction of forest sites within
act as steppingstones that permit some persistence of epiphytes a 20-year period is 20% (lambda mean 1.021), while pasture sites
outside of the forest in a changing landscape mosaic (Köster et al., have a 45% (lambda mean 0.938) probability of extinction (Figure 2;
2009; Elias et al., 2021). In these circumstances some species become Supplementary Table 5). According to Criterion C for the IUCN
denser and more abundant outside of the forest, possibly as a Red List evaluations, a species may be categorized as vulnerable with
response to scarce available of phorophytes (Larrea and Werner, less than 1,000 mature individuals in each subpopulation and/or a
2010), as well as changing abiotic conditions. probability of extinction of 10% in 10 years. While it is likely that
Nonetheless, studies in other epiphytic species have shown that R. granadensis has more than a total of 10,000 individuals across its
in transformed habitat, population density initially increases quickly, range (the central aspect of criterion C), our findings indicate that
only to later decrease, often leading to extinction, depending on time the likelihood of subpopulation extinction is high even in the forest
and distance to the forest source of seeds (Pellegrino et al., 2015; environment.
Hylander and Nemomissa, 2017). Thus, the high-density populations Only a small number of multi-period censuses of orchid
of R. granadensis in open pasture found in this study may be of population dynamics have been undertaken in the tropics, and these
a transient nature. Continued population monitoring over a longer have similarly found a negative impact on orchid populations in
period is needed to gauge this temporal effect. landscapes with anthropogenic activity. In a study of three epiphytic
The abundance of reproductive adults and fruit set was greater species growing on coffee trees, Oncidium poikilostalix (Kraenzl.)
in the forest environment (Supplementary Table 3). Native forests M.W.Chase and N.H.Williams, Lepanthes acuminata Schltr. and
likely comprise a more suitable ecological niche and adequate Telipogon helleri (L.O.Williams) N.H.Williams and Dressler in
pollinator community compared to isolated trees within a pasture Chiapas México, lambda was greater in populations in unmanaged
matrix. With increasing isolation of phorophytes from the native coffee plantations compared with managed plantations (García-
forests, a reduction in the number of Euglossine pollinators visiting González et al., 2017; Raventós et al., 2018). In the terrestrial
these isolated patches has been observed (Briggs et al., 2013). tropical invasive species, Oeceoclades maculata (Lindl.) Lindl. a
Our study reveals that the demographic health of orchid twig higher population growth rate was noted within a Mexican forest
epiphytes is negatively influenced in transformed environments; than in a managed coffee plantation (Riverón-Giró et al., 2019).
hence, in both landscape structures, the forest sub-sites showed While in Phaius flavus (Blume) Lindl. in southeast China, Li et al.
higher asymptotic population growth rate with greater resilience (2022) found that populations tended to decrease, and this change
(inertia, reactivity) and a lower short-term population decline. was attributed to the low germination rate in the wild and the
Frontiers in Ecology and Evolution 10279 frontiersin.org
fevo-11-1135316 February 8, 2023 Time: 15:42 # 8
Ospina-Calderón et al. 10.3389/fevo.2023.1135316
FIGURE 4
Visualizations of transient dynamics for four selected sub-sites at different time periods showing the diversity of plausible transient responses. Data from;
Cl1, Calibío Time period 1; Cj2, Cajibío Time period 2, (F) Forest, (P) Pasture. Decreasing Cj2P, stable, Cl1F and increasing Cj2F. y Cl1P the figure
represents the diversity of transient dynamics as a function of the initial starting vector (the number of distribution of individuals at each stage). Darker
shading indicates a higher probability of the population size in that range.
loss of adult individuals caused by anthropogenic disturbances. Population convergence time to a stable state distribution was
In this current study, individual plants tended to remain in the lower for forest sites than in pastures, which suggest that forest
same stage from one census to the next (Supplementary Table 2). habitat may be beneficial for population stability and promoting
Such stasis as the predominant life history process has also been higher population resiliency (Supplementary Table 5). Furthermore,
registered in other neotropical epiphytic orchids (Tremblay and values for inertia and reactivity, amplification and attenuation had
Hutchings, 2002; Crain et al., 2019). In general, in iteroparous forest wider intervals for forest populations, and so greater resilience
plants with long lifespans, multi-year reproductive adult stages, and in the face of changing environmental circumstances, including
generation overlap, populations often consist of a preponderance habitat transformations or climate change. Rapid fluctuations in the
of adults of varying sizes that remain in the same stage and population size through time (Figure 4) could be advantageous if a
contribute to population recruitment through the reproductive (Fe) population can increase rapidly after a size reduction due to stochastic
stage (Silvertown et al., 1996). In contrast, iteroparous plants in phenomena. However, it may also suggest vulnerability if the
open habitat plants typically exhibit populations with predominantly fluctuation results in a rapid decrease in population size, as noted in
growth (G) and reproductive (Fe) transitions (Silvertown et al., 1993; Lepanthes caritensis Tremblay and Ackerman, Dendrophylax lindenii
Franco and Silvertown, 2004). Our data show that R. granadensis (Lindl.) Benth. ex Rolfe, Broughtonia cubensis Cogn (Raventós et al.,
populations in an open habitat retain the forest strategy, with 2015,b; Tremblay et al., 2015; Crain et al., 2019).
persistence of adults, lower generational turnover, fewer seedlings Most of the transient dynamics simulations for R. granadensis
and juveniles that survive to adulthood, slower growth rates, all reflect the tendency for rapid reduction and high probability of
leading to declining populations. extinction in about five to 10 years. Although some subsites
Frontiers in Ecology and Evolution 10380 frontiersin.org
fevo-11-1135316 February 8, 2023 Time: 15:42 # 9
Ospina-Calderón et al. 10.3389/fevo.2023.1135316
FIGURE 5
Rodriguezia granadensis transient dynamic simulation for 50 seasons, 25 years for tree of six study sites, Li, Lilas; Cl, Calibío; Cj, Cajibío. Subsites: F,
forest; P, pasture. Bias S1, S2, S3, S4, stage-biased demographic vectors.
showed population growth (Cl, Pi, H, Y forest and Pi, H pastures), example, in Central American and Caribbean orchid populations, the
the remaining subsites were near “equilibrium” without increased growth rate and high intensity fluctuations are mediated by stochastic
tendency to growth for more than 10 years. Even though populations disturbance due to large storms or hurricanes (Crain et al., 2019;
are near equilibrium this does not necessarily guarantee that these Ortiz-Rodríguez et al., 2020; Raventós et al., 2021).
populations will persist. A number of studies have shown that even Perturbation analysis allows us to identify the effects of probable
when population sizes fluctuate, they are vulnerable to extinction changes in each transition on the growth rate (Stott et al., 2011).
when stochastic events are common (Raventós et al., 2015b; Crain In R. granadensis the stasis stages may be the most elastic as small
et al., 2019). Twig epiphytes may be highly vulnerable to stochastic changes the parameters could result in large, non-linear changes in
events, as loss of small branches as a consequence of the architectural population growth rates, most often showing a pattern close to a
growth of trees and competition with surrounding trees may result in narrow inverted “U.”
reduced niche availability for these obligate small branch epiphytes. Our analyses show that R. granadensis populations have lower
Inherent fluctuations of epiphytic and twig epiphytic habit survival probability when colonizing phorophytes dispersed in
represent important constrictions for population growth, structure a pasture matrix as compared to forest sites. The diminished
and distribution of R. granadensis. Transient dynamics are highly persistence of this orchid in a modified landscape can likely be
influenced by the initial vector and therefore linked to explosion considered an extinction debt. Colonization of isolated trees may
or extinction and stochastic phenomena (García-González et al., prevent extinction in the short term, but the persistence of these
2017; Raventós et al., 2018). The pattern and intensity of fluctuation sites may depend on the dynamics of the sink-source and the
in population size may be exacerbated by natural phenomena. For distance from a more suitable forest fragment (Pellegrino et al., 2015;
Frontiers in Ecology and Evolution 10391 frontiersin.org
fevo-11-1135316 February 8, 2023 Time: 15:42 # 10
Ospina-Calderón et al. 10.3389/fevo.2023.1135316
Hylander and Nemomissa, 2017). While orchid populations within Funding
the forest and pasture landscape may function as a sink-source
dynamic, the importance of the source vs. sink processes are presently This study was supported by Internal call project Pontificia
little understood. Apposite questions include: how important is this Universidad Javeriana Seccional Cali (PUJ) project no. 020100486
process for the persistence of pasture populations, and is the direction and Doctorate in Sciences-Biology Universidad del Valle, National
of the seed source always from forest to pastures sites? Moreover, Doctorates Scholarship, COLCIENCIAS call 647.
stochastic processes linked with the natural population dynamics
of the species are drivers of epiphyte presence and persistence, and
subsequently their interaction in land transformation and habitat
fragmentation need to be considered (Armbrecht, 2003; Rivera- Acknowledgments
Pedroza et al., 2019; Zewdie et al., 2022).
This work collects partial data from the Doctorate in Biology
Conclusion Sciences thesis of the Universidad del Valle-Cali, with honors (Res.
154 Nov. 2020), with the support of UNAL Palmira and RNFBY,
Payanesa Orchids Association, Popayán, Julia Calderón and Luis
The endemic twig epiphyte, R. granadensis, is present
Fernando Ospina, Cristian Delgado, Ana Isabel Parra, Constanza
in anthropogenically-transformed land covers, but analysis of
asymptotic and transitory dynamics indicates that these populations Álvarez, María Mercedes Valencia Falla, Adriana Villalba, Fabián and
have lower viability than those in native forest fragments. Populations Luis Paz, Valentín Hidalgo, Beatriz Vásquez, Álvaro José Botero, and
on isolated trees have lower generational turnover, fewer seedlings García Ayala family.
and juveniles that survive to adulthood, slower growth rates,
and, in general, declining populations. Our data suggests that the
demographic dynamics of epiphytic orchids are of a fluctuating
nature, which makes them more vulnerable to disturbances and Conflict of interest
stochastic events. Since R. granadensis is a species categorized as
of least concern (LC) according to IUCN Red List criteria, a more The authors declare that the research was conducted in the
hopeful pattern in its population dynamics was expected, especially absence of any commercial or financial relationships that could be
since it is found more or less frequently in disturbed landscapes. This construed as a potential conflict of interest.
contradiction in the health of a species when comparing observed
long-term population growth rates and IUCN criteria may in part
be that IUCN criteria used are those which do not explicitly include
the ecology and long-term dynamics of the species but a snapshot Publisher’s note
of the population based on multiple assumptions which may not be
predictors of the future health for some species. All claims expressed in this article are solely those of the
authors and do not necessarily represent those of their affiliated
organizations, or those of the publisher, the editors and the reviewers.
Data availability statement Any product that may be evaluated in this article, or claim that may
be made by its manufacturer, is not guaranteed or endorsed by the
The original contributions presented in this study are included publisher.
in this article/Supplementary material, further inquiries can be
directed to the corresponding author.
Supplementary material
Author contributions
The Supplementary Material for this article can be found online
NO-C collected the data and ran the analysis. All authors at: https://www.frontiersin.org/articles/10.3389/fevo.2023.1135316/
contributed equally on writing and approved the submitted version. full#supplementary-material
References
Ackerman, J. D., Sabat, A., and Zimmerman, J. K. (1996). Seedling establishment in Alvarenga, L. D. P., and Pôrto, K. C. (2007). Patch size and isolation effects on epiphytic
an epiphytic orchid: an experimental study of seed limitation. Oecologia 106, 192–198. and epiphyllous bryophytes in the fragmented Brazilian Atlantic forest. Biol. Conservat.
doi: 10.1007/BF00328598 134, 415–427. doi: 10.1016/j.biocon.2006.08.031
Aguilar, R., Ashworth, L., Galetto, L., and Aizen, M. A. (2006). Plant reproductive Arenas-Clavijo, A., and Armbrecht, I. (2018). Gremios y diversidad de
susceptibility to habitat fragmentation: review and synthesis through a meta-analysis. hormigas (Hymenoptera: Formicidae) en tres usos del suelo de un paisaje
Ecol. Lett. 9, 968–980. doi: 10.1111/j.1461-0248.2006.00927.x cafetero del Cauca-Colombia. Rev. Biol. Trop. 66, 48–57. doi: 10.15517/rbt.v66i1.
Aguilar, R., Cristóbal-Pérez, E. J., Balvino-Olvera, F. J., de Jesús Aguilar-Aguilar, M., 30269
Aguirre-Acosta, N., Ashworth, J. A., et al. (2019). Habitat fragmentation reduces plant Armbrecht, I. (2003). Habitat changes in Colombian coffee farms under increasing
progeny quality: a global synthesis. Ecol. Lett. 22, 1163–1173. doi: 10.1111/ele.13272 management intensification. Endang. Spec. Update 20, 163–180.
Frontiers in Ecology and Evolution 11302 frontiersin.org
fevo-11-1135316 February 8, 2023 Time: 15:42 # 11
Ospina-Calderón et al. 10.3389/fevo.2023.1135316
Benítez-Malvido, J., and Arroyo-Rodríguez, V. (2008). Habitat fragmentation, edge Hietz, P. (1999). Diversity and conservation of epiphytes in a changing environment.
effects and biological corridors in tropical ecosystems. Oxford: Eolss Publishers. Pure Appl. Chem. 70, 1–11.
Benitez-Malvido, J., Dattilo, W., Martinez-Falcon, A. P., Duran-Barron, C., Valenzuela, Hoang, N. T., and Kanemoto, K. (2021). Mapping the deforestation footprint of nations
J., Lopez, S., et al. (2016). The multiple impacts of tropical forest fragmentation on reveals growing threat to tropical forests. Nat. Ecol. Evol. 5, 845–853. doi: 10.1038/s41559-
arthropod biodiversity and on their patterns of interactions with host plants. PLoS One 021-01417-z
11:e0146461. doi: 10.1371/journal.pone.0146461 Honnay, O., and Jacquemyn, H. (2007). Susceptibility of common and rare plant
Briggs, H. M., Perfecto, I., and Brosi, B. J. (2013). The role of the agricultural matrix: species to the genetic consequences of habitat fragmentation. Conserv. Biol. 21, 823–831.
coffee management and euglossine bee (Hymenoptera: Apidae: Euglossini) communities doi: 10.1111/j.1523-1739.2006.00646.x
in southern Mexico. Environ. Entomol. 42, 1210–1217. doi: 10.1603/EN13087
Hylander, K., and Nemomissa, S. (2017). Waiving the extinction debt: Can shade from
Brosi, B. J. (2009). The effects of forest fragmentation on euglossine bee communities coffee prevent extinctions of epiphytic plants from isolated trees? Div. Distribut. 23,
(Hymenoptera: Apidae: Euglossini). Biol. Conservat. 142, 414–423. doi: 10.1016/j.biocon. 888–897. doi: 10.1111/ddi.12579
2008.11.003
IDEAM (2010). Promedios Climatológicos 1981-2010. Available online at: http://www.
Brudvig, L. A., Damschen, E. I., Haddad, N. M., Levey, D. J., and Tewksbury, J. J. (2015). ideam.gov.co/web/tiempo-y-clima/clima (accessed October 27, 2022).
The influence of habitat fragmentation on multiple plant–animal interactions and plant
reproduction. Ecology 96, 2669–2678. doi: 10.1890/14-2275.1 Iles, D. T., Salguero-Gómez, R., Adler, P. B., and Koons, D. N. (2016). Linking transient
dynamics and life history to biological invasion success. J. Ecol. 104, 399–408.
Bruna, E. M., Fiske, I. J., and Trager, M. D. (2009). Habitat fragmentation and plant
populations: Is what we know demographically irrelevant? J. Veg. Sci. 20, 569–576. Jacquemyn, H., De Meester, L., Jongejans, E., and Honnay, O. (2012). Evolutionary
changes in plant reproductive traits following habitat fragmentation and their
Calderón-Sáenz, E. (2007). Libro rojo de plantas de Colombia. Vol. 6. Orquídeas, consequences for population fitness. J. Ecol. 100, 76–87. doi: 10.1111/j.1365-2745.2011.
primera parte. Serie Libros Rojos de Especies Amenazadas de Colombia. Bogotá: Instituto 01919.x
Alexander von Humboldt-Ministerio de Ambiente, Vivienda y Desarrollo Territorial.
Jules, E. S., and Shahani, P. (2003). A broader ecological context to habitat
Calvo, R. N., and Horvitz, C. C. (1990). Pollinator limitation, cost of reproduction, and fragmentation: why matrix habitat is more important than we thought. J. Veg. Sci. 14,
fitness in plants: a transition-matrix demographic approach. Am. Natur. 136, 499–516. 459–464. doi: 10.1111/j.1654-1103.2003.tb02172.x
Caswell, H. (2000). Matrix Population Models, Vol. 1. Sunderland: Sinauer. Kolb, A., and Diekmann, M. (2005). Effects of life-history traits on responses of plant
Catling, P. M., Brownell, V. R., and Lefkovitch, L. P. (1986). Epiphytic orchids in species to forest fragmentation. Conserv. Biol. 19, 929–938.
a Belizean grapefruit orchard: distribution, colonization and association. Lindleyana 1, Köster, N., Friedrich, K., Nieder, J., and Barthlott, W. (2009). Conservation of epiphyte
194–202. diversity in an Andean landscape transformed by human land use. Conserv. Biol. 23,
Chase, M. W. (1987). Obligate twig epiphytism in the Oncidiinae and other neotropical 911–919. doi: 10.1111/j.1523-1739.2008.01164.x
orchids. Selbyana 10, 24–30. Köster, N., Nieder, J., and Barthlott, W. (2011). Effect of host tree traits on epiphyte
Crain, B. J., Tremblay, R. L., and Ferguson, J. M. (2019). Sheltered from the storm? diversity in natural and anthropogenic habitats in Ecuador. Biotropica 43, 685–694. doi:
Population viability analysis of a rare endemic under periodic catastrophe regimes. 10.1111/j.1744-7429.2011.00759.x
Populat. Ecol. 61, 74–92. doi: 10.1002/1438-390X.1002 Larrea, M. L., and Werner, F. A. (2010). Response of vascular epiphyte diversity to
Criollo, Y., and Bastidas, S. (2011). Análisis del Paisaje y de su Relación con different land-use intensities in a neotropical montane wet forest. For. Ecol. Manage. 260,
Características Florísticas de los Bosques Naturales de la Compañía Smurfit Kappa Cartón 1950–1955. doi: 10.1016/j.foreco.2010.08.029
de Colombia, en el Núcleo Meseta, Municipios de Popayan y Cajibio, Departamento del Letourneau, D. K., Armbrecht, I., Salguero Rivera, B., Montoya Lerma, J., Jiménez
Cauca. Doctoral thesis. Popayán: Universidad del Cauca. Carmona, B., Daza, M. C., et al. (2011). Does plant diversity benefit agroecosystems? A
Debinski, D. M. (2006). Forest fragmentation and matrix effects: the matrix does synthetic review. Ecol. Appl. 21, 9–21.
matter. J. Biogeogr. 33, 1791–1792. doi: 10.1111/j.1365-2699.2006.01596.x Li, Z., Wang, Y., and Mu, L. (2022). How does deforestation affect the growth
Elias, J. P., Mortara, S. R., Nunes-Freitas, A. F., van den Berg, E., and Ramos, F. N. of cypripedium (Orchidaceae) species? A simulation experiment in Northeast China.
(2021). Host tree traits in pasture areas affect forest and pasture specialist epiphyte species Forests 13:166. doi: 10.3390/f13020166
differently. Am. J. Bot. 108, 598–606. doi: 10.1002/ajb2.1634 Liang, J., Gamarra, J. G., Picard, N., Zhou, M., Pijanowski, B., Jacobs,
Etter, A., McAlpine, C., Wilson, K., Phinn, S., and Possingham, H. (2006). Regional D. F., et al. (2022). Co-limitation towards lower latitudes shapes global forest
patterns of agricultural land use and deforestation in Colombia. Agric. Ecosyst. Environ. diversity gradients. Nat. Ecol. Evol. 6, 1423–1437. doi: 10.1038/s41559-022-
114, 369–386. doi: 10.1016/j.agee.2005.11.013 01831-x
Etter, A., and van Wyngaarden, W. (2000). Patterns of landscape transformation in Lienert, J. (2004). Habitat fragmentation effects on fitness of plant populations–a
Colombia, with emphasis in the Andean region. Ambio 29, 432–439. review. J. Nat. Conserv. 12, 53–72. doi: 10.1016/j.jnc.2003.07.002
Fischer, J., and Lindenmayer, D. B. (2007). Landscape modification and habitat López-Gallego, C., and Morales, M. P. A. (2021). Rodriguezia Granadensis. The
fragmentation: a synthesis. Glob Ecol. Biogeogr. 16, 265–280. doi: 10.1111/j.1466-8238. IUCN Red List of Threatened Species 2021: e.T118267887A203443808. Available
2007.00287.x online at: https://dx.doi.org/10.2305/IUCN.UK.2021-3.RLTS.T118267887A203443808.
es (accessed October 26, 2022).
Franco, M., and Silvertown, J. (2004). A comparative demography of plants based upon
elasticities of vital rates. Ecology 85, 531–538. Martorell, C., Flores-Martínez, A., and Franco, M. (2022). Conceptual and
methodological issues in structured population models of plants. Bot. Sci. 100, 110–136.
Fréville, H., McConway, K., Dodd, M., and Silvertown, J. (2007). Prediction of
extinction in plants: Interaction of extrinsic threats and life history traits. Ecology 88, Marull, J., Delgadillo, O., Cattaneo, C., La Rota, M. J., and Krausmann, F. (2018).
2662–2672 Socioecological transition in the Cauca river valley, Colombia (1943–2010): towards an
energy–landscape integrated analysis. Reg. Environ. Change 18, 1073–1087. doi: 10.1007/
García-González, A., Damon, A., Raventós, J., Riverón-Giró, F. B., Mújica, E., and s10113-017-1128-2
Solís-Montero, L. (2017). Impact of different shade coffee management scenarios, on
a population of Oncidium poikilostalix (Orchidaceae), in Soconusco, Chiapas, Mexico. McDonald, J. L., Stott, I., Townley, S., and Hodgson, D. J. (2016). Transients drive
Plant Ecol. Div. 10, 185–196. doi: 10.1080/17550874.2017.1315840 the demographic dynamics of plant populations in variable environments. J. Ecol. 104,
306–314. doi: 10.1111/1365-2745.12528
García-González, A., and Riverón-Giró, F. B. (2013). Organización espacial y
estructura de una población de Ionopsis utricularioides (Orchidaceae) en un área Medeiros, T. D. S. (2010). Distribuição Vertical de Orquídeas Epífitas e Relação com os
suburbana de Pinar del Río, Cuba. Lankesteriana Int. J. Orchidol. 13, 419–427. Forófitos na apa Ilha do Combú, Belém, Pará, Brasil. Brasil: Universidade Federal Rural
Gentry, A. H. (1982). Neotropical floristic diversity: phytogeographical connections da Amazônia.
between Central and South America, Pleistocene climatic fluctuations, or an accident of Mondragón, D., Maldonado, C., and Aguilar-Santelises, R. (2007). Life history and
the Andean orogeny? Ann. Missouri Bot. Gard. 69, 557–593. demography of a twig epiphyte: a case study of Erycina crista-galli (Orchidaceae).
Gravendeel, B., Smithson, A., Slik, F. J., and Schuiteman, A. (2004). Epiphytism and Selbyana 28, 137–144.
pollinator specialization: drivers for orchid diversity? Philos. Trans. R. Soc. Lond B Biol. Murren, C. J. (2002). Effects of habitat fragmentation on pollination: pollinators,
Sci. 359, 1523–1535. doi: 10.1098/rstb.2004.1529 pollinia viability and reproductive success. J. Ecol. 90, 100–107.
Harvey, C. A., and González Villalobos, J. A. (2007). Agroforestry systems conserve Myers, N., Mittermeier, R. A., Mittermeier, C. G., Da Fonseca, G. A., and Kent, J.
species-rich but modified assemblages of tropical birds and bats. Biodiv. Conserv. 16, (2000). Biodiversity hotspots for conservation priorities. Nature 403, 853–858.
2257–2292. Newman, B. J., Ladd, P., Brundrett, M., and Dixon, K. W. (2013). Effects of
Harvey, C. A., Pritts, A. A., Zwetsloot, M. J., Jansen, K., Pulleman, M. M., Armbrecht, habitat fragmentation on plant reproductive success and population viability at the
I., et al. (2021). Transformation of coffee-growing landscapes across Latin America. landscape and habitat scale. Biol. Conserv. 159, 16–23. doi: 10.1016/j.biocon.2012.
A review. Agron. Sustain. Dev. 41, 1–19. doi: 10.1007/s13593-021-00712-0 10.009
Hernández-Pérez, E., and Solano, E. (2015). Effects of habitat fragmentation on the Olson, D. M., and Dinerstein, E. (1998). The Global 200: a representation approach to
diversity of epiphytic orchids from a montane forest of southern Mexico. J. Trop. Ecol. conserving the Earth’s most biologically valuable ecoregions. Conserv. Biol. 12, 502–515.
31, 103–115. doi: 10.1017/S0266467414000662 doi: 10.1046/j.1523-1739.1998.012003502.x
Frontiers in Ecology and Evolution 11313 frontiersin.org
fevo-11-1135316 February 8, 2023 Time: 15:42 # 12
Ospina-Calderón et al. 10.3389/fevo.2023.1135316
Ortiz-Rodríguez, I. A., Raventós, J., Mújica, E., González-Hernández, E., Vega-Peña, Silvertown, J., Franco, M., and Menges, E. (1996). Interpretation of elasticity matrices
E., Ortega-Larrocea, P., et al. (2020). Spatiotemporal effects of Hurricane Ivan on an as an aid to the management of plant populations for conservation. Conserv. Biol. 10,
endemic epiphytic orchid: 10 years of follow-up. Plant Ecol. Div. 13, 29–45. doi: 10.1080/ 591–597. doi: 10.1046/j.1523-1739.1996.10020591.x
17550874.2019.1673495 Silvertown, J., Franco, M., Pisanty, I., and Mendoza, A. N. A. (1993). Comparative plant
Osie, M., Shibru, S., Dalle, G., and Nemomissa, S. (2022). Habitat fragmentation demography–relative importance of life-cycle components to the finite rate of increase in
effects on vascular epiphytes diversity in Kafa biosphere reserve and nearby coffee woody and herbaceous perennials. J. Ecol. 81, 465–476.
Agroecosystem, Southwestern Ethiopia. Trop. Ecol. 63, 1–11. doi: 10.1007/s42965-022-
00223-3 Stott, I., Hodgson, D. J., and Townley, S. (2012a). Beyond sensitivity: nonlinearperturbation analysis of transient dynamics. Methods Ecol. Evol. 3, 673–684. doi: 10.1111/
Ospina-Calderón, N. H. (2009). Selección de Rasgos Florales en Rodriguezia j.2041-210X.2012.00199.x
Granadensis (Lindl.) Rchb.f. (Orchidaceae): Estudio de la Eficacia Biológica en Una Especie
Polimórfica. Bogotá: Universidad Nacional de Colombia. Stott, I., Hodgson, D. J., and Townley, S. (2012b). Popdemo: an R package forpopulation demography using projection matrix analysis. Methods Ecol. Evol. 3, 797–802.
Ospina-Calderón, N. H., Duque-Buitrago, C. A., Tremblay, R. L., and Tupac Otero, J. doi: 10.1111/j.2041-210X.2012.00222.x
(2015). Pollination ecology of Rodriguezia granadensis (Orchidaceae). Lankesteriana 15,
129–139. doi: 10.15517/lank.v15i2.20745 Stott, I., Townley, S., and Hodgson, D. J. (2011). A framework for studying transientdynamics of population projection matrix models. Ecol. Lett. 14, 959–970. doi: 10.1111/j.
Parra Sánchez, E., Armenteras, D., and Retana, J. (2016). Edge influence on diversity of 1461-0248.2011.01659.x
orchids in Andean cloud forests. Forests 7:63. Tomimatsu, H., and Ohara, M. (2010). Demographic response of plant populations to
Pellegrino, G., Bellusci, F., and Palermo, A. M. (2015). Effects of population structure habitat fragmentation and temporal environmental variability. Oecologia 162, 903–911.
on pollen flow, clonality rates and reproductive success in fragmented Serapias lingua doi: 10.1007/s00442-009-1505-8
populations. BMC Plant Biol. 15:222. doi: 10.1186/s12870-015-0600-8 Torres, A. M., Adarve, J. B., Cárdenas, M., Vargas, J. A., Londoño, V., Rivera, K., et al.
Philpott, S. M., Arendt, W. J., Armbrecht, I., Bichier, P., Diestch, T. V., Gordon, C., et al. (2012). Dinámica sucesional de un fragmento de bosque seco tropical del Valle del Cauca,
(2008). Biodiversity loss in Latin American coffee landscapes: review of the evidence on Colombia. Biota Colomb. 13, 66–85.
ants, birds, and trees. Conserv. Biol. 22, 1093–1105. doi: 10.1111/j.1523-1739.2008.01029.
x Tremblay, R. L. (1997). Distribution and dispersion patterns of individuals in ninespecies of Lepanfhes (Orchidaceae) 1. Biotropica 29, 38–45.
Puertas-Orozco, O. L., Carvajal Escobar, Y., and Quintero Angel, M. (2011). Estudio de
tendencias de la precipitación mensual en la cuenca alta-media del río Cauca, Colombia. Tremblay, R. L., Ackerman, J. D., and Pérez, M. E. (2010). Riding across the selection
Dyna 78, 112–120. landscape: Fitness consequences of annual variation in reproductive characteristics.Philos. Trans. R. Soc. B 365, 491–498.
Raventós, J., García-González, A., Riverón-Giró, F. B., and Damon, A. (2018).
Comparison of transient and asymptotic perturbation analyses of three epiphytic orchid Tremblay, R. L., and Hutchings, M. J. (2002). Population Dynamics in Orchid
species growing in coffee plantations in Mexico: effect on conservation decisions. Plant Conservation: a Review of Analytical Methods Based on the Rare Species Lepanthes
Ecol. Div. 11, 133–145. doi: 10.1080/17550874.2018.1444110 eltoroensis. Orchid conservation. Kota Kinabalu: Natural History Publications (Borneo),183–204.
Raventós, J., González, E., Mújica, E., and Bonet, A. (2015). Transient population
dynamics of two epiphytic orchid species after Hurricane Ivan: implications for Tremblay, R. L., Raventos, J., and Ackerman, J. D. (2015). When stable-stage
management. Biotropica 47, 441–448. doi: 10.1111/btp.12231 equilibrium is unlikely: integrating transient population dynamics improves asymptoticmethods. Ann. Bot. 116, 381–390. doi: 10.1093/aob/mcv031
Raventós, J., González, E., Mújica, E., and Doak, D. F. (2015b). Population viability
analysis of the epiphytic ghost orchid (Dendrophylax lindenii) in Cuba. Biotropica 47, Tremblay, R. L., Tyre, A. J., Pérez, M. E., and Ackerman, J. D. (2021). Population
179–189. doi: 10.1111/btp.12202 projections from holey matrices: using prior information to estimate rare transitionevents. Ecol. Model. 447:109526. doi: 10.1016/j.ecolmodel.2021.109526
Raventós, J., Mújica, E., Gonzalez, E., Bonet, A., and Ortega-Larrocea, M. P. (2021).
The effects of hurricanes on the stochastic population growth of the endemic epiphytic Vandermeer, J., Armbrecht, I., De la Mora, A., Ennis, K. K., Fitch, G., Gonthier, D. J.,
orchid Broughtonia cubensis living in Cuba. Populat. Ecol. 63, 302–312. doi: 10.1002/ et al. (2019). The community ecology of herbivore regulation in an agroecosystem:
1438-390X.12098 lessons from complex systems. BioScience 69, 974–996. doi: 10.1093/biosci/biz127
Richards, J. H., Luna, I. M. T., and Waller, D. M. (2020). Tree longevity drives
conservation value of shade coffee farms for vascular epiphytes. Agric. Ecosyst. Environ. Vélez-Torres, I., Varela, D., Cobo-Medina, V., and Hurtado, D. (2019). Beyond
301:107025. doi: 10.1016/j.agee.2020.107025 property: rural politics and land-use change in the Colombian sugarcane landscape.J. Agrarian Change 19, 690–710. doi: 10.1111/joac.12332
Ritchie, H., and Roser, M. (2013). “Land Use”. Published Online at
OurWorldInData.org. Available online at: ‘https://ourworldindata.org/land-use’ [Online Vellend, M., Baeten, L., Becker-Scarpitta, A., Boucher-Lalonde, V., McCune, J. L.,
Resource] (accessed September 9, 2013). Messier, J., et al. (2017). Plant biodiversity change across scales during the Anthropocene.Annu. Rev. Plant Biol. 68, 563–586. doi: 10.1146/annurev-arplant-042916-04
Rivera-Pedroza, L. F., Escobar, F., Philpott, S. M., and Armbrecht, I. (2019). The 0949
role of natural vegetation strips in sugarcane monocultures: ant and bird functional
diversity responses. Agric. Ecosyst. Environ. 284:106603. doi: 10.1016/j.agee.2019. Ventre-Lespiaucq, A. B., Delgado, J. A., Ospina-Calderón, N. H., Otero, J. T., Escudero,
106603 A., Sánchez, M. A., et al. (2017). A tropical epiphytic orchid uses a low-light interceptionstrategy in a spatially heterogeneous light environment. Biotropica 49, 318–327. doi:
Riverón-Giró, F. B., Raventós, J., Damon, A., García-González, A., and Mújica, 10.1111/btp.12425
E. (2019). Spatio-temporal dynamics of the invasive orchid Oeceoclades maculata
(Orchidaceae), in four different habitats in southeast Chiapas, Mexico. Biol. Invasions Wickham, H. (2016). Ggplot2: Elegant Graphics for Data Analysis. New York, NY:
21, 1905–1919. doi: 10.1007/s10530-019-01945-7 Springer-Verlag.
Rodríguez-Echeverry, J., and Leiton, M. (2021). State of the landscape and dynamics Wickham, H., Averick, M., Bryan, J., Chang, W., McGowan, L., François, R., et al.
of loss and fragmentation of forest critically endangered in the tropical andes hotspot: (2019). Welcome to the tidyverse. J. Open Source Softw. 4:1686. doi: 10.21105/joss.01686
implications for conservation planning. J. Landsc. Ecol. 14, 73–91. doi: 10.2478/jlecol- Winkler, K., Fuchs, R., Rounsevell, M., and Herold, M. (2021). Global land use changes
2021-0005 are four times greater than previously estimated. Nat. Commun. 12, 1–10. doi: 10.1038/
Sabat, A. M., and Ackerman, J. D. (1996). Fruit set in a deceptive orchid: the effect of s41467-021-22702-2
flowering phenology, display size, and local floral abundance. Am. J. Bot. 83, 1181–1186. Zewdie, B., Tack, A. J., Ayalew, B., Wondafrash, M., Nemomissa, S., and Hylander,
Sardi, A., Torres, A. M., and Corredor, G. (2018). Diversidad florística en un paisaje K. (2022). Plant biodiversity declines with increasing coffee yield in Ethiopia’s coffee
rural del piedemonte de los Farallones de Cali, Colombia. Colomb. For. 21, 142–160. agroforests. J. Appl. Ecol. 59, 1198–1208. doi: 10.1111/1365-2664.14130
doi: 10.14483/2256201x.10866 Zotz, G. (2007). Johansson revisited: the spatial structure of epiphyte assemblages.
Schödelbauerová, I., Tremblay, R. L., and Kindlmann, P. (2010). Prediction vs. reality: J. Veg. Sci. 18, 123–130. doi: 10.1111/j.1654-1103.2007.tb02522.x
Can a PVA model predict population persistence 13 years later? Biodiv. Conserv. 19, Zotz, G. (2013). The systematic distribution of vascular epiphytes–a critical update.
637–650. doi: 10.1007/s10531-009-9724-1 Bot. J. Linn. Soc. 171, 453–481. doi: 10.1111/boj.12010
Frontiers in Ecology and Evolution 11324 frontiersin.org
TYPE Original Research
PUBLISHED 24 March 2023
DOI 10.3389/fevo.2023.1085938
Floral and genetic divergence 
OPEN ACCESS across environmental gradients is 
EDITED BY
Tiiu Kull,  moderated by inter-population 
Estonian University of Life Sciences,  
Estonia gene flow in Platanthera dilatata 
REVIEWED BY
Caroline Turchetto,  
Federal University of Rio Grande do Sul,  (Orchidaceae)
Brazil
Jeremie Benjamin Fant,  
Chicago Botanic Garden,  Lisa E. Wallace 1* and Marlin L. Bowles 2
United States
Kadri Tali,  1 Department of Biological Sciences, Old Dominion University, Norfolk, VA, United States, 2 The Morton 
Estonian University of Life Sciences,  Arboretum (retired), Lisle, IL, United States
Estonia
*CORRESPONDENCE
Lisa E. Wallace  Understanding how natural selection acts on intraspecific variation to bring 
 lewallac@odu.edu about phenotypic divergence is critical to understanding processes of 
SPECIALTY SECTION evolutionary diversification. The orchid family is well known for pollinator-
This article was submitted to  mediated selection of floral phenotypes operating among species and along 
Conservation and Restoration Ecology,  environmental or geographic gradients. Its effectiveness at small spatial scales 
a section of the journal  
Frontiers in Ecology and Evolution is less understood, making the geographic scale at which intraspecific floral 
31 October 2022 variation is examined important to evaluating causes of phenotypic divergence. RECEIVED 
ACCEPTED 02 March 2023 In this study, we quantified phenotypic variation in the orchid Platanthera dilatata 
PUBLISHED 24 March 2023 across 26 populations in coastal Southeast Alaska and compared this to edaphic 
CITATION and genetic variation at microsatellite loci. We sought to determine (1) if flower 
Wallace LE and Bowles ML (2023) Floral and morphological variation is structured at smaller geographic scales, (2) the extent 
genetic divergence across environmental 
gradients is moderated by inter-population of genetic divergence in relation to phenotypic divergence, (3) the scale at which 
gene flow in Platanthera dilatata (Orchidaceae). inter-population gene flow occurs, and (4) the relative importance of geographic 
Front. Ecol. Evol. 11:1085938. distance and abiotic factors on population genetic structure. Two morphological 
doi: 10.3389/fevo.2023.1085938
groups were found to separate based on lip and spur length and are restricted to 
COPYRIGHT different habitats. Small-flowered forms occur in muskeg bogs, whereas large-
© 2023 Wallace and Bowles. This is an open-
access article distributed under the terms of flowered forms occur in fens and meadows, and rarely in sub-alpine habitat. 
the Creative Commons Attribution License Genetic analyses were concordant with the morphological clusters, except for 
(CC BY). The use, distribution or reproduction four small-flowered populations that were genetically indistinguishable from 
in other forums is permitted, provided the 
original author(s) and the copyright owner(s) large-flowered populations and considered to be  introgressed. In fact, most 
are credited and that the original publication in populations exhibited some admixture, indicating incomplete reproductive 
this journal is cited, in accordance with isolation between the flower forms. Pollinators may partition phenotypes but 
accepted academic practice. No use, 
distribution or reproduction is permitted which also facilitate gene flow because short-tongued Noctuidae moths pollinate both 
does not comply with these terms. phenotypes, but longer-tongued hawkmoths were only observed pollinating 
the large-flowered phenotype, which may strengthen phenotypic divergence. 
Nevertheless, pollinator movement between habitats could have lasting effects 
on neutral genetic variation. At this small spatial scale, population genetic 
structure is only associated with environmental distance, likely due to extensive 
seed and pollinator movement. While this study corroborates previous findings of 
cryptic genetic lineages and phenotypic divergence in P. dilatata, the small scale 
of examination provided greater understanding of the factors that may underlie 
divergence.
KEYWORDS
cryptic divergence, flower variation, genetic structure, gene flow, soils, isolation by 
distance, isolation by environment, Platanthera dilatata
Frontiers in Ecology and Evolution 01 frontiersin.org
135
Wallace and Bowles 10.3389/fevo.2023.1085938
1. Introduction questions: (1) Is flower morphological variation structured at smaller 
geographic scales, (2) Are floral phenotypes genetically divergent, (3) 
Understanding the origin and maintenance of intraspecific Does gene flow occur across morphologically distinct populations, 
variation is of central importance to evolutionary biology because they and (4) How do geographic distance and environmental differences 
inform our understanding of diversification across space and time and influence population genetic structure? We predicted strong isolation 
illuminate the process of speciation (Pinheiro et al., 2018). Substantial by distance at the regional scale (i.e., encompassing all study 
intraspecific phenotypic variation may indicate the maintenance of populations) because of limitations on gene flow via seeds and 
polymorphisms over otherwise connected populations (Nobarinezhad selection on flowers by pollinators, but at a local scale (i.e., less than 
and Wallace, 2022), or it could indicate the presence of evolutionarily 50 km between populations), we predicted that environmental factors 
divergent cryptic lineages that exhibit parallel ecological responses would more strongly influence genetic structure because seeds should 
(Kahl et al., 2021). Whereas a polymorphic species is expected to be  capable of dispersal over these distances but may differ in 
experience spatially and temporally heterogeneous gene flow among adaptation to habitats and pollinators.
populations, cryptic lineages exhibiting genetic divergence should 
be isolated from one another (Surveswaran et al., 2018). While genetic 
tools have been especially useful for identifying cryptic lineages, 2. Materials and methods
integrated approaches involving multiple data types and widespread 
sampling of populations provide not only the identification of cryptic 2.1. Study area
lineages but also clues about their divergence and geographic spread 
(Surveswaran et al., 2018; Liu et al., 2022). When examined deeply, This study took place in Southeast Alaska, United States. which 
many species have been found to comprise cryptic lineages (Pinheiro comprises an 800 km mountainous coastline and adjacent island chain 
et al., 2018). Linking such divergence with pollinator selection of floral along the northwest coast of North America (Figure 1). The climate of 
traits is critical to understanding how it integrates with co-evolutionary this region is primarily wet maritime, averaging over 300 cm annual 
processes in determining ecological speciation (Van der Niet precipitation. The average maximum temperature reaches 18°C in 
et al., 2014). July, and the average minimum temperature reaches -4°C in January 
The orchid genus Platanthera (L.) Rich. contains many (Shulski and Wendler, 2007; Bienek et al., 2012). The predominant 
phenotypically polymorphic species (e.g., Robertson and Wyatt, 1990; coastal vegetation is northern rainforest; about 17% of the area is 
Wallace, 2003a; Bateman and Sexton, 2008; Bateman et  al., 2013; non-forested shrubland and peatland (Kirchoff et al., 2016). This area 
Adhikari and Wallace, 2014) and potentially cryptic lineages (Wettewa was glaciated <10,000 years BP; as a result, climate and post-glacial 
et  al., 2020). As in many orchids, this phenotypic variation is migration strongly affect vegetation composition (Andersen, 1955; 
frequently attributed to pollinator-mediated selection (Hapeman and Mathewes, 1985), but glacial refugia present during the late Wisconsin 
Inoue, 1997; Van der Niet et al., 2014). Such selection has been shown glaciation also may have allowed persistence and recolonization of 
to operate even within species (Robertson and Wyatt, 1990). At larger vegetation within this region (Carrara et al., 2007).
geographic scales or along environmental gradients, pollinator- In the study area, P. dilatata is most abundant in open bog and fen 
mediated selection is a reasonable hypothesis for morphological peatlands, coastal, lakeshore, and riverine meadows, and 
polymorphism if pollinators exhibit habitat preferences or have anthropogenic-disturbed roadsides (Figure 2). Bogs, also known as 
distributional limits. However, at smaller geographic scales, other muskeg, are usually ombrotrophic and develop at low to 
factors must also be  considered to explain the maintenance of mid-elevations but grade into subalpine conditions with less organic 
phenotypic variation in Platanthera species. Characterizing the matter. These habitats usually comprise sapric to hemic peat, and 
geographic scale of phenotypic variation within species is important support plant species of open bogs, including Sphagnum L. sp., Carex 
for distinguishing among competing factors in the maintenance of L. sp., and Ericaceous shrubs (Neiland, 1971). Fens usually occur at 
this variation. low elevations along drainage ways and range from weak to moderately 
Platanthera dilatata (Pursh) Lind. ex L.C. Beck is distributed minerotrophic, receiving greater nutrient input than bogs (Fellman 
across the northern U.S. and Canada, reaching as far south as New and D’Amore, 2007; Fellman et al., 2008; D’Amore et al., 2010, 2015). 
Mexico and as far north as Alaska. This species has been treated as They comprise floating or solid mats of hemic to fibric peat and 
representing three varieties based on nectar spur length, which are support a subset of bog and meadow vegetation. Coastal meadows, 
thought to partition pollinators by corresponding proboscis lengths also termed uplift meadows, are developed in fine-textured glacial 
(Sheviak, 2002). However, as noted by Sheviak (2002), “the recognized outwash and lacustrine deposits and are undergoing isostatic uplift 
varieties of P. dilatata are evidentially merely endpoints in a very following glaciation. They are dominated by broad-leaved herbs, with 
complex variation pattern,” leading to unanswered questions as to why a minor component of graminoid species, and may zonate along 
polymorphism in this species exists. tidelands (Stone, 1993). Anthropogenic roadsides have mineral soils 
In this study, we examined phenotypic and genotypic divergence developed from grading and gravel deposition and tend to represent 
among populations of P. dilatata in Southeast Alaska and across a subset of meadow vegetation that tolerates disturbances such as 
elevational, climatic, and edaphic gradients. In the study area, seasonal mowing.
P. dilatata populations do not readily fit into the varieties outlined by The mycorrhizal fungi Ceratobasidium sp. and Tulasnella sp. have 
Sheviak (2002). Thus, we  sampled across an area covering many been identified in P. dilatata root samples from the study area. Two of 
habitats and flower types to quantify variation in soil characteristics, three Ceratobasidium isolates were from muskeg, while 10 of 11 
climatic variables, flower morphological traits, and genetic variation Tulasnella samples were from fen, meadow, or anthropogenic habitat 
at microsatellite loci. We used these data to address the following (Melton, 2020; M. McCormick, pers. comm.; L. Zettler, pers. comm.). 
Frontiers in Ecology and Evolution 02 frontiersin.org
136
Wallace and Bowles 10.3389/fevo.2023.1085938
A B
C
FIGURE 1
Location of (A) study area in North America, (B) Southeast Alaska collection sites of Platanthera dilatata sampled for morphology, genetic markers, and 
soil characteristics, and (C) distribution of populations shown in the box in panel B and referenced as central in landscape genetic analyses. In panels B 
and C, shapes and colors indicate the six combinations of habitat x flower size x genetic cluster observed in this study. Site names follow those in 
Table 1 and Supplementary Table S1.
Pollinators of P. dilatata include Noctuidae moths, the hawkmoth >10 flowering plants at each site, and regional distribution to 
Hyles gallii (Rottemburg, 1775), and the butterfly Pieris marginalis maximize sampling in morphologically diverse populations and 
Scudder, 1861 (Figure 3); pollinia were deposited on the proboscises environmentally variable sites.
of these insects (Bowles and Armstrong, 2021). Noctuidae moths 
appear to be primary pollinators across all habitats, but hawkmoths 
may be most frequent in fens and meadows. 2.3. Morphological data collection and 
analysis
2.2. Site selection Lip and spur length are the most important variables for 
distinguishing among varieties of P. dilatata (Adhikari and Wallace, 
The 26 study sites represented 12 muskeg bogs, six meadows, four 2014). These metrics were obtained from single flowers selected from 
fens and four anthropogenic roadsides, spanning ca. 500 km from 10–28 (mean = 17.4, se = 1.01) inflorescences from each study 
north to south (Figure 1; Table 1; Supplementary Table S1). Although population. Flowers were collected in 2018–2019. Flower collection 
fen, meadow, and anthropogenic habitats may occur in southern was stratified to represent the range of inflorescence sizes present; 
Southeast Alaska, study sites for these habitats were restricted to flowers were collected from the lower third of inflorescences to avoid 
northern Southeast Alaska. Sites were selected based on accessibility, nectar spurs that were not fully developed. Flowers were stored in 
lack of anthropogenic disturbance (excluding roadsides) presence of zip-lock plastic bags at 4°C, and measured within 48 h. Each flower 
Frontiers in Ecology and Evolution 03 frontiersin.org
137
Wallace and Bowles 10.3389/fevo.2023.1085938
TABLE 1 Genetic diversity at eight microsatellite loci across sampled cluster analysis was performed in NCSS statistical software (Hintze, 
locations of Platanthera dilatata in Southeast Alaska.
2013). This test evaluated 1–5 clusters, using 5 random starts to 
Site Habitat N Na %P HO HE FIS produce an optimum solution in which within-cluster sum of squares 
name is minimized. A goodness of fit comparison of the percent variation in 
Large-flowered populations each within-cluster group sum of squares relative to one group was 
FM Anthropogenic 18.4 3.0 100 0.413* 0.497 0.106 used to evaluate which number of clusters had the greatest reduction 
in variation (Hintze, 2013). The Duda and Hart (1973) test was also 
HM Anthropogenic 20.0 3.4 100 0.500 0.496 0.031
used to evaluate whether single or multiple clusters better fit the data, 
ND Anthropogenic 23.6 4.5 100 0.516 0.604 0.034 followed by application of the Calinski and Harabasz (1974) index to 
UAK Anthropogenic 19.0 3.7 100 0.507 0.501 0.033 further evaluate the most likely number of clusters beyond one.
BS Meadow 22.0 3.6 88 0.381* 0.458 0.113 We calculated PST (Brommer, 2011) to estimate the degree of 
differentiation in lip and spur length among populations. PST was then BBN Meadow 15.9 3.6 100 0.557 0.528 0.039
compared to FST estimated from the microsatellite data (see below) to 
BBS Meadow 22.7 3.9 100 0.460* 0.513 0.088 evaluate the relative potential for selection and genetic drift to drive 
ERT Meadow 15.0 3.6 100 0.467* 0.511 0.068 the observed differences in floral traits. PST was estimated separately 
ML Meadow 9.0 3.5 88 0.542 0.542 0.054 for lip length and spur length using the R package Pstat (Blondaeu Da 
PBM Meadow 9.0 3.0 100 0.403* 0.428 0.133 Silva and Da Silva, 2018). Data were subjected to Atchinson 
transformation and the value of c/h2 was set to 1; bootstrap analysis 
AHR Fen 12.6 3.6 100 0.544 0.513 0.090 with 1,000 replicates was used to calculate 95% confidence intervals 
PC Fen 22.6 3.7 100 0.432* 0.501 0.078 for PST and this was compared to our estimate of FST based on 
AM Fen 24.0 3.6 100 0.458 0.462 0.043 microsatellite loci.
PBF Fen 10.0 2.9 100 0.400 0.376 0.051
HMA Muskeg bog 10.0 1.5 50 0.288 0.226 0.025 2.4. Soil data collection and analysis
Small-flowered populations
DMa Muskeg bog 24.0 4.0 100 0.641 0.578 0.013 Soil samples were collected from each study site in 2018–2022. 
PBMKa Muskeg bog 20.7 3.5 100 0.459 0.491 0.040 Each sample comprised multiple excavations made to rooting depth 
with a hand trowel, which were combined into a single collection for 
CLa Muskeg bog 23.0 4.2 100 0.565 0.600 0.035
each site. Samples were analyzed by Waypoint Analytical (Richmond, 
GIa Muskeg bog 16.0 2.7 100 0.539 0.470 0.033 Virginia, USA) for percent organic matter (POM); parts per million 
ELO Muskeg bog 20.9 3.2 100 0.431* 0.508 0.069 (PPM) Ca, K, Mg, and P; percent base saturation (PBS) Ca, K and Mg; 
EUP Muskeg bog 19.0 2.5 75 0.270 0.284 0.053 percent H saturation (PHS); and cation exchange capacity (CEC, 
meq/100 g). Analytic methods followed Horton (2011).
HMM Muskeg bog 21.7 2.2 88 0.319 0.364 0.032
Soils data were analyzed with ANOVA and multivariate statistics. 
IR Muskeg bog 23.9 2.7 100 0.350 0.385 0.031 One-way ANOVA was used to test whether soils variables differed 
MJ Muskeg bog 21.0 2.7 88 0.310 0.316 0.043 among muskeg, fen, meadow, and anthropogenic habitat groups, 
BM Muskeg bog 14.6 3.1 100 0.434 0.436 0.044 which supported different orchid phenotypes (see below). For these 
SM Muskeg bog 19.0 2.6 100 0.395 0.451 0.071 tests, transformations were used to approximate normality for POM 
and PBS K (arcsin transformation), PPM P (log transformation), PPM 
Mean–large-flowered 16.9 3.4 95 0.458 0.477 0.066 Ca and Ca (square root transformation). Non-metric 
populations Multidimensional Scaling (NMS) was used on PCORD (McCune and 
Mean–small-flowered 20.0 2.7 93 0.358 0.392 0.049 Mefford, 2011) to ordinate habitat groups using POM, pH, CEC, PPM 
without hybrid populationsa P, PBS K, PBS Mg, PBS Ca, and PHS as metrics. A relative Euclidian 
T-testb P -- 2.486 0.369 3.074 2.206 1.228 distance measure with a random seed starting configuration and 100 
0.022 0.72 0.006 0.039 0.233 runs with real data were used to project three axes using a Varimax 
N = mean number of individuals sampled across all loci, Na = mean number of alleles per rotation, for which stability was tested with a randomization test. 
locus, % P = percentage of polymorphic loci, HO = observed heterozygosity, HE = expected Relationships of each metric with the first and second NMS axis were 
heterozygosity, FIS = inbreeding coefficient. tested with correlation analysis. A Multi-Response Permutation 
*Significant deviation from Hardy–Weinberg equilibrium (P < 0.05).
aSmall-flowered populations suspected of having introgression from large-flowered Procedures (MRPP) test was used on PCORD to assess whether 
populations. habitats differed in their multivariate distributions based on soils 
bT-tests were conducted without the inclusion of hybrid small populations. metrics. Because of skewed metrics in anthropogenic habitat soils, the 
MRPP test was repeated with this group excluded from the analysis.
was dissected to remove the lip and spur, and their lengths were 
measured to the nearest 0.5 mm. Most spurs were falcate, and they 
were flattened under a flexible sheet of transparent plastic for 2.5. Genetic data collection and analysis
linear measurement.
Population means (+ se) were calculated for flower and lip length. Leaf samples used in genetic analyzes were collected in 2018–
To assess whether morphological groups could be identified, a k-means 2019. A 5 cm length of fresh leaf tissue was removed from one leaf 
Frontiers in Ecology and Evolution 04 frontiersin.org
138
Wallace and Bowles 10.3389/fevo.2023.1085938
FIGURE 2
Variation in habitats and soils occupied by Platanthera dilatata. (A) anthropogenic roadside, (B) fen, (C) uplift meadow, (D) and muskeg bog. Panels (E,F) 
show the NMS ordination of habitat vegetation types in relation to soil characteristics and flower group as indicated by the K-means clustering analysis 
of floral traits (F). Ordination final stress = 3.4891, final instability = 0.0; probability of final stress obtained by chance (Axis 1 p = 0.002, Axis 2 p = 0.044). 
Cumulative correlations between ordination distances and distances in the original n-dimensional space: Axis 1 r2 = 0.828, Axis 2 r2 = 0.990. MRPP: all 
habitats (A = 0.41613688, p < 0.0001); anthropogenic habitats excluded (A = 0.26071168, p < 0.0001). See Supplemental Table S4 for soils variables axis 
correlation statistics.
from 9–26 (mean = 18.8 se = 1.3) plants from each study site. Leaf genotyped at nine microsatellite loci that were developed from a 
samples were stored in zip-lock plastic bags at 4°C. These samples transcriptome library of P. dilatata (Wallace, unpublished data). 
were dried within 24 h. by placing them in folded aluminum foil Primer sequences are provided in Supplementary Table S2. The nine 
containing silica gel crystals and then sealed within double zip-lock loci were amplified using a multiplex PCR with the Kapa 2G Fast 
plastic bags. DNA was extracted from dried leaves using the multiplex PCR kit (Roche Sequencing and Life Science, Wilmington, 
SYNERGY 2.0 Plant DNA extraction kit (OPS Diagnostics, Lebanon, Massachusetts, USA) and fluorescent labeled primers following 
New Jersey, USA) and stored in 1X TE buffer. DNA samples were protocols in Culley et al. (2013). Each fluorescently labeled primer 
standardized to 10 ng/μl for use in PCR. Each sampled plant was contained a sequence that matched a tag sequence located on the 5′ 
Frontiers in Ecology and Evolution 05 frontiersin.org
139
Wallace and Bowles 10.3389/fevo.2023.1085938
FIGURE 3
Variation in flower traits and pollinators of Platanthera dilatata in the study area of Southeast Alaska. Differences in (A) inflorescence size and (B) the 
lengths of spurs (in boxes) between plants grouped by flower size and (C) mean (+ se) lip and spur length of flowers, with the ellipses indicating 95% 
concentrations of populations within large and small-flowered groups identified by k-means cluster analysis. Pollinators observed on P. dilatata flowers 
in the study populations include (D) Autographa corusca Strecker, 1885 on small-flowered phenotype (photo: R. H. Armstrong), (E) Actebia fennica 
(Tauscher, 1806) on small-flowered phenotype (photo: G. Bayluss), (F) Plusia sp. Ochsenheimer 1816 on the large-flowered phenotype (photo: R. H. 
Armstrong), (G) Autographa corusca Strecker, 1885 on large-flowered phenotype, and (H), Hyles gallii (Rottemburg, 1775) on large-flowered 
phenotype (photo: R. H. Armstrong); D-G are Noctuidae species, and H is a Sphingidae species. Large-flowered group: lip mean = 8.43 (se = 0.13), spur 
mean = 10.31 (se = 0.23), t-test of lip and spur lengths: t = −7.089, p < 0,001; small-flowered group: lip mean = 6.49 (se = 0.173), spur mean = 8.11 (se = 0.25),), 
t-test of lip and spur lengths: t = −5.385, p < 0.001. One-way ANOVA between morphological groups: Lip F1,24 = 85.45, p < 0.0001, Spur F1,24 = 40.50, 
p < 0.0001. Lip-spur correlations: among groups (r = 0.6409, p = 0.0004); small-flowered group (r = −0.3877, p = 0.237); large-flowered group (r = −0.0581, 
p = 0.8435).
Frontiers in Ecology and Evolution 06 frontiersin.org
140
Wallace and Bowles 10.3389/fevo.2023.1085938
end of the locus-specific forward primer. For each sample, each We evaluated population genetic structure according to the 
multiplex reaction was performed in a final volume of 10 μl in the groups identified by morphological analyzes and NMS of the soil 
presence of 10 ng of template DNA, 100 μmole of each of the reverse variables, that is, between large-flowered and small-flowered groups 
and tagged fluorescently labeled primers and 10 μmole of tagged and separately among the four habitat types (Table 1) using analysis 
forward primer using KAPA 2G Fast Multiplex PCR mix. The thermal of molecular variance (AMOVA) (Excoffier et al., 1992), conducted in 
cycler program used to amplify loci included 3 min at 95°C, 30 cycles GenAlEx version 6.503 (Peakall and Smouse, 2012). Statistical 
of 15 s at 95°C, 30 s at 60°C, and 30 s at 72°C, and a final extension step significance of AMOVA was assessed by 9,999 permutations. We used 
of 1 min at 72°C. Amplified products were genotyped at the Institute the Bayesian clustering approach implemented in STRUCTURE v. 
of Biotechnology at Cornell University with LIZ 500 size standard, and 2.3.4 (Pritchard et al., 2000) to test for admixture and assignment of 
individual alleles were sized using GeneMarker (SoftGenetics, State individuals to distinct genetic clusters. These analyzes were conducted 
College, Pennsylvania, United States). using an admixture model with correlated allele frequencies, a ‘burn-
The presence of null alleles in each locus and population was in’ period of 50,000 MCMC replicates, sampling 100,000 replicates, 
checked using the program FreeNA (Chapuis and Estoup, 2007). and eight iterations of each K value, from one to 13. This range of K 
Null allele frequencies <0.2 are not expected to greatly influence the values was used in the final run because an initial analysis of four 
results of population genetic analyzes (Dakin and Avise, 2004; iterations each for K values from 1 to 25 under similar run parameters 
Carlsson, 2008). Thus, we considered further only loci exhibiting a indicated low probability of a K value greater than five. For the final 
null allele frequency > 0.2, which occurred at three loci, 72267, analysis multiple posterior probability values (log likelihood (lnL) 
99945, and 107223, in eight, three, and three populations, values) for the eight iterations of each K were generated, and the most 
respectively. Locus 72267 was removed from the dataset because of likely number of clusters was determined using STRUCTURE 
the extensive occurrence of potential null alleles. We  further HARVESTER (Earl and vonHoldt, 2012) and Delta K- (Evanno et al., 
investigated inbreeding as a potential cause of null alleles for the 2005). CLUMPP (Jakobsson and Rosenberg, 2007) was used to 
other two loci. Heterozygote deficiency, which is a potential sign of aggregate individual assignment probabilities from the eight iterations 
null allele presence, has often been reported in association with for the selected K. STRUCTURE PLOT (Ramasamy et al., 2014) was 
significant FIS in other orchids (Chung et al., 2004; Alcantara et al., used to generate plots of individual assignment from the CLUMPP 
2006; Andriamihaja et al., 2021). For each of the five populations output file.
suspected of having null alleles, we compared a model based on the We estimated the potential for admixture in populations using 
inclusion of null alleles, inbreeding, and genotyping errors (i.e., nfb) several methods. Identify scores (Q-matrix scores) from the 
with one lacking inbreeding (i.e., nb) using the software INEST v. 2.2 STRUCTURE analysis were used to infer if individuals were of pure 
(Chybicki and Burczyk, 2009). These analyzes were implemented ancestry or contained an admixed background. An identity 
using a Bayesian approach with 1 million MCMC cycles, keeping score < 0.9  in a single cluster was used to assign an individual as 
every 100th result, and a burn-in of 10,000 prior to summarizing the admixed. NewHybrids v1.1 (Anderson and Thompson, 2002) was 
results. DIC values were compared between the two models to used to assign each individual to one of six genotypic classes (i.e., pure 
evaluate the impact of inbreeding on observed diversity. For large-flowered, pure small-flowered, F1, F2, backcross with large-
population PC, the full model had a substantially lower DIC than the flowered, or backcross with small- flowered). This analysis was 
model without inbreeding. For the other four populations (i.e., conducted without specifying individuals to a particular class, and all 
HMM, IR, SM, and ND), the difference in DIC between the two individuals were analyzed. We ran the analysis using a Jeffreys prior, 
models was less than 1.5. As these results suggest that inbreeding 10,000 burn-in replicates, and 1 million sweeps before assignment 
may account for the lack of heterozygous individuals in these probabilities were determined. No individual was assigned to any of 
populations at the suspected loci, we chose to retain data for these the hybrid classes with probably >0.7, so we only considered a hybrid 
locus-population combinations for further analysis of genetic group, rather than F1, F2, or backcross generations. Furthermore, 
diversity and structure. we used a cut-off probability of >0.9 to assign individuals into one of 
Within each population, we tested for significant departures from the pure parental groups, rather than the hybrid group.
Hardy–Weinberg expectations using a global test of heterozygote BayesAss (Wilson and Rannala, 2003) and Geneclass 2 (Piry 
deficiency in GENEPOP version 3.2 (Raymond and Rousset, 1995; et al., 2004) were used to estimate the proportion of immigrants and 
Rousset, 2008). Genotypic linkage disequilibrium was measured for non-immigrants. Whereas BayesAss (Wilson and Rannala, 2003) is 
each pair of loci in each population and tested through Fisher’s exact better able to detect older instances of movement, Geneclass (Piry 
test using GENEPOP version 3.2 (Raymond and Rousset, 1995; et al., 2004) more aptly identifies first generation immigrants. For 
Rousset, 2008) and applying a Bonferroni correction (Holm, 1979). these analyzes we assigned populations to one of three groups, large-
Genetic diversity within populations was assessed as number of alleles flowered populations, small-flowered populations, and hybrid 
per locus (Na), observed heterozygosity (HO), expected heterozygosity populations, after considering the morphological groupings and 
(HE), and percent of polymorphic loci (% P) using GenAlEx version genetic assignments suggested by STRUCTURE and NewHybrids 
6.503 (Peakall and Smouse, 2012). Inbreeding coefficients were (see results, Supplementary Table S4; Figures  3C, 4A). Hybrid 
calculated in INEST (Chybicki and Burczyk, 2009) as described above. populations were identified by their conflicting placement into 
To determine if genetic diversity varies between large and small- groups based on morphological and genetic variation (i.e., small-
flowered populations, as identified in the morphological K-means flowered plants that were genetically similar to large-flowered plants). 
clustering, we compared mean values of Na, HO, HE, and FIS using BayesAss analysis was conducted using 50 million iterations, a 
t-tests in SPSS v. 27 (IBM Corp, 2020). p < 0.05 was used to identify burn-in of 1 million, and sampling every 5,000 generations. The 
significant differences in genetic diversity between the flower groups. Geneclass analysis was conducted to identify first generation 
Frontiers in Ecology and Evolution 07 frontiersin.org
141
Wallace and Bowles 10.3389/fevo.2023.1085938
A
B
FIGURE 4
Admixture proportions for all samples of Platanthera dilatata from Southeast Alaska based on analysis with (A) STRUCTURE and (B) NewHybrids. For a, 
the Q-matrix scores for each individual according to the solution K = 2 (blue and pink clusters) are indicated. For (B), the estimated probability that an 
individual is from one of the pure parental groups (blue or pink) or a hybrid (black) is shown. Populations are arranged in the order shown in Table 1. 
See Figure 2E for alignment of habitat groups along soil gradients.
immigrants only using the criterion of Rannala and Mountain (1997) (R Core Team, 2022). These geographic distances were 
with 10,000 simulated individuals under the simulation algorithm of log-transformed to reduce the impact of the largest distances. For the 
Paetkau et  al. (2004). A p < 0.05 was used to identify significant environmental dataset we considered elevation, four soil factors that 
immigration events. were important in the NMS ordination (i.e., POM, pH, Pppm, and 
Past studies (e.g., Balkenhol et al., 2009; Kierepka and Latch, PBSK), and five uncorrelated (r2 < 0.70) climate variables (i.e., 
2015) have found that different methods for assessing the role of precipitation in the warmest quarter, precipitation seasonality, mean 
geography, environment, or other factors on population genetic temperature of the coldest quarter, mean temperature of the driest 
structure show only moderate agreement and recommend choosing quarter, and mean temperature of the wettest quarter) sampled from 
multiple statistical approaches when testing for isolation by distance 30-s layers of the WorldClim data set (Fick and Hijmans, 2017) for 
(IBD), isolation by environment (IBE), or other factors. Thus, all locations considered in this study. The environmental dataset was 
we  used multiple matrix regression (MMR) (Wang, 2013) and subjected to principal component analysis (PCA) using R (R Core 
distance-based RDA (dbRDA) (Legendre and Anderson, 1999) to test Team, 2022). Values for each site along the first two axes, which 
for IBD and IBE. MMR compares pairwise population genetic accounted for 49% of the observed variation, were used to construct 
distance against distance matrices based on explanatory variables an environmental distance matrix based on Euclidean distances in 
using regression, whereas dbRDA assesses selected explanatory Passage 2 (Rosenberg and Anderson, 2011). Multivariate MMR was 
variables directly as predictors of population genetic distance. conducted using the MMRR script of Wang (2013) in R (R Core 
We  conducted these analyzes at a regional scale, i.e., on all Team, 2022) using the explanatory matrixes of geographic distance 
populations, as well as a local scale, i.e., the centrally located and environmental distance and the response matrix of genetic 
populations near Juneau (Figure  1B) to evaluate whether distances. Significance was tested using 9,999 permutations.
environmental factors are more predictive of genetic structure at the For dbRDA, the geographic distance matrix was used in a 
local scale and geographic distance at the regional scale, given that principal coordinates of neighbor matrices (PCNM; Borcard and 
orchid seeds may be  capable of dispersing over several hundred Legendre, 2002; Borcard et al., 2004) in R (R Core Team, 2022) with 
kilometers (Arditti and Ghani, 2000; Phillips et al., 2012). default threshold values to generate a set of independent variables 
For all analyzes, pairwise population genetic distances were reflecting spatial relationships among the populations. The positive 
generated using the distance metric of Cavalli-Sforza and Edwards PCNM axes were retained and tested as predictors of genetic distance 
(1967), with correction by the INA method implemented in FreeNA in dbRDA. To test for IBE, we used the first two axes from a principal 
and described in Chapuis and Estoup (2007). For MMR, distance component analysis (PCA) of the environmental variables, as 
matrixes reflecting geography and environmental features were described above, as predictors of pairwise genetic distances. The 
created in the following manner. The pairwise population geographic explanatory variables were assessed independently in marginal tests 
distance matrix was created using GPS coordinates and the distGeo() and conditioned on geographic distance to account for potential 
function in the geosphere package v. 1.5 (Hijmans et al., 2021) in R correlation between environmental factors and geography. The 
Frontiers in Ecology and Evolution 08 frontiersin.org
142
Wallace and Bowles 10.3389/fevo.2023.1085938
dbRDA analyzes were conducted using the capscale () function of the Non-metric Multidimensional Scaling reached a stable solution 
Vegan package (Oksanen et al., 2020) in R (R Core Team, 2022). An for ordination of two axes after 88 iterations (Figure 2E). Only CEC 
analysis of variance was used to evaluate significance of each model. and PBS Mg were not significantly correlated with either ordination 
The varpart() function in R (R Core Team, 2022) was used to assess axis (Supplementary Table S4). On Axis 1, muskeg bog habitat was 
the contribution of each environmental and geographic variables to strongly associated with positive axis scores and highly correlated 
genetic distances. with increasing POM and PHS, while anthropogenic roadside habitat 
was strongly associated with negative axis scores and highly 
correlated with increasing pH and PBS Ca. Meadow habitat was most 
3. Results strongly associated with Axis 2 and greater PBS K, but also tended to 
be associated with increasing PPM P along Axis 1. Fen habitat was 
3.1. Flower morphology centrally located and intermediate with respect to soil chemistry and 
nutrients. With MRPP, all habitats had significantly different 
In the k-means cluster analysis, a two-group solution reduced multivariate distributions, which remained different with 
percent variation of within-sum of squares to 29.6%. Subsequent anthropogenic habitat excluded from the model (Figure 2E).
clusters further reduced variation by <10%. The Duda-Hart test 
indicated that the optimal clustering solution contained more than 
one group (DHK = 0.3555, alpha = 0.99), and the Calinski-Harabasz 3.3. Genetic variation and population 
index also was greater for two clusters. Thus, two clusters were structure
selected as the optimal solution.
The two-cluster analysis separated populations with Among the 728 inter-locus comparisons, there were six instances 
significantly different lip and spur lengths, which were correlated of significant genetic disequilibrium identified in three populations. 
among, but not within, groups (Figure 3C). In both groups, spurs No locus pairs exhibited significant disequilibrium in multiple 
were significantly longer (by >20%) than lips, however, one group populations, but one population (i.e., ELO) did have four loci out of 
had 20% longer lips and spurs than did the other group, and thus equilibrium. Overall, these results suggest that the loci are genetically 
larger flowers (Figures 3A,B). A comparison of inflorescence size independent, and that instances of linkage disequilibrium are likely 
among three populations of small-flowered plants and four due to demographic factors unique to the affected populations. Seven 
populations of large-flowered plants (among three habitats) found populations had a deficiency of heterozygotes consistent with 
that small-flowered plants also had smaller inflorescences (N = 37, deviation from Hardy–Weinberg equilibrium (Table 1). Mean genetic 
mean = 26.95, se = 1.60) than did large-flowered populations diversity was significantly higher in populations assigned to the large-
(N = 52, mean = 38.85, se = 2.25); nested ANOVA of ln-transformed flowered group than those assigned to the small-flowered group 
data: F2,82 = 28.77, p = 0.033. The correspondence between habitats (excluding the four hybrid populations) when considering Na, HO, 
and flower morphology is shown in Figure  2F. The group with and HE but not for % P or the inbreeding coefficient (Table 1).
smaller flowers comprised populations occurring only in muskeg AMOVA assigned most of the observed genetic structure 
bog habitat (Figure 2D). The group with larger flowers included all within populations (79–80%), then among populations (15–16%), 
fen, meadow, and anthropogenic habitats, as well as a single and between the small and large-flowered groups (6%) or among 
muskeg site that occurred at high elevation and is a transition to the habitats (5%; Table  2). All F-statistics were significant 
alpine habitat (Figures 2A–D). (p < 0.01). The optimal number of groups, based on Bayesian 
Populations showed strong differentiation in spur length and lip analysis in STRUCTURE of the genetic variation, was two clusters 
length as PST values were 0.94 (95% CI: 0.930–0.959) for spur length (Figure  4A). The two clusters primarily align with large and 
and 0.97 (95% CI: 0.969–0.980) for lip length. These estimates were small-flowered populations, although samples from four small-
robust to variation in our selection of the value for c/h2. The critical flowered populations (i.e., PBMK, CL, BM, and DM) were placed 
value of PST, whereby quantitative traits are more strongly reflective in the cluster with large-flowered populations. Most individuals 
of selection than genetic drift, occurred at c/h2 < 0.5 (80%) were assigned to a single cluster with Q-matrix 
(Supplementary Figure S1). values >0.95.
Our analyzes indicated admixture between the phenotypic 
groups. This was most extensive for the four small-flowered 
3.2. Soil chemistry and fertility populations that are genetically-like large-flowered populations as all 
but seven samples from these populations were assigned to this 
Most soils variables differed significantly among habitat groups cluster with Q-matrix scores >0.9 by STRUCTURE, but other 
(Supplementary Table S3). Percent organic matter had the strongest instances of admixture were also noted for most populations 
differentiation. It was significantly greater in muskeg (64%) (Figure 4A). In fact, based on a cut-off of 0.9 in Q-matrix scores from 
intermediate in fen (37%), and lower (< 10%) in meadow and STRUCTURE, all but four populations (i.e., HA, HMA, HMM, and 
anthropogenic habitats. Other significant variables (pH, CEC, ppm ML) contained at least one admixed individual, resulting in ca. 10% 
Ca, and PBS Ca) were greater in anthropogenic habitat and not of all individuals assigned as admixed. NewHybrids produced nearly 
different among other habitats. Ca was about 400% higher in identical results to those from STRUCTURE (Figure 4B), although 
anthropogenic habitat, where it reached 1783.5 PPM. Though it did seven populations were not predicted to contain admixed individuals 
not differ significantly (p = 0.115), PBS K tended to be  higher in by this analysis. NewHybrids also estimated more extensive 
meadow habitat, where it reached 4.9%. hybridization in the small-flowered IR population compared to 
Frontiers in Ecology and Evolution 09 frontiersin.org
143
Wallace and Bowles 10.3389/fevo.2023.1085938
TABLE 2 Results from an analysis of molecular variance based on allelic TABLE 3 Results from distance-based Redundancy Analyzes (dbRDA) 
diversity among populations of Platanthera dilatata from Southeast testing the effects of geographic distance (Geo) and environmental 
Alaska. factors (Env) on genetic distance among the populations of Platanthera 
dilatata surveyed in Southeast Alaska.
Source df SS MS Percent of 
variance Marginal test Conditional test
Flower groups Variable F P % F P % 
Variation Variation
Among groups 1 92.542 92.542 6
Among 24 383.237 15.968 15 Full
populations Geo 1.904 0.01 26.62
Within 936 1830.080 1.955 79 Enva 2.83 0.001 19.75 2.13 0.017 18.30
populations Central
Total 961 2305.860 2.494 100 Geo 1.341 0.141 20.09
FRT = 0.063, P < 0.001; FSR = 0.163, P < 0.001; FST = 0.216, p < 0.001 Enva 3.126 0.003 26.89 2.334 0.026 25.00
Habitat aFor conditional tests, the contribution of environmental factors was considered after 
removing the covariate effects of geographic distance. The analyzes were run on the full set of 
Among groups 3 122.659 40.886 5 populations (Full) and the centrally located populations only (Central).
Among 22 353.120 16.051 16
populations
Within 936 1830.080 1.955 80 4. Discussion
populations
Total 961 2305.860 100 No previous assessment of P. dilatata has concurrently 
FRT = 0.045, P < 0.01; F  = 0.165, p < 0.01; F  = 0.202, p < 0.01
examined genetic structure, morphological diversity, and habitat 
SR ST
characteristics in populations with a shared regional geography. 
Groups were designated by (a) flower size and (b) habitat. Examination of populations at this scale provided greater 
understanding of intraspecific variation for this species and of the 
STRUCTURE, which identified admixed genotypes in only four environmental factors that may influence morphological and 
individuals in this population. genetic variation. We  note several novel results: (1) flower 
Further support of two genetically divergent groups was found phenotypes are strongly associated with habitats, (2) there is a deep 
in the estimated immigration rates, which were extremely low genetic divergence between small-flowered and large-flowered 
between large-flowered and small-flowered populations. Both forms, (3) nevertheless, admixture has occurred between 
BayesAss and Geneclass suggested that at least 97% of the populations harboring different phenotypes and introgression is 
individuals in each of these groups originated within their assigned deeply rooted in some populations, and (4) whereas IBD and IBE 
group (Supplementary Table S5). BayesAss, but not Geneclass, both contribute to significant population genetic structure at 
indicated strongly unidirectional immigration from the large- regional scales, among closely spaced populations, environmental 
flowered group into the hybrid populations. Only a low level of factors are stronger determinants of genetic structure.
immigration was detected into the hybrid group from the small-
flowered group, even though these populations share a common 
habitat type and are morphologically similar. Additionally, low 4.1. Phenotypic variation in relation to 
levels of recent immigration were detected between large-flowered environmental factors
and small-flowered groups and from the hybrid group 
by GeneClass. Platanthera dilatata has long been recognized as a 
Landscape genetic analyzes indicated that at the regional level, morphologically variable species (Luer, 1975; Sheviak, 2002). 
both geographic distance and environmental factors are predictive Within Southeast Alaska, P. dilatata populations have variable 
of genetic structure. In MMR analysis at this scale, the multivariate flower morphology, yet the phenotypes are partitioned by habitat 
model placed geographic distance as the strongest factor, with (Figures 1, 2). Plants with inflorescences containing fewer flowers 
environmental distance slightly less important but still significant. and flowers with shorter lips and spurs have a narrow habitat niche 
The overall r-square for this model is 0.27, but it is significant as they are restricted to muskeg bogs, whereas plants with larger 
(p = 0.0002). Comparable results were obtained with dbRDA, with inflorescences and flowers with longer lips and spurs have a broader 
geographic distance explaining slightly more variation than habitat niche, occurring across a habitat gradient that is exclusive 
environmental factors when considered independently (Table 3). of muskeg bogs except at extremely high elevations.
Despite a correlation between geographic distance and The high estimates of PST for lip and spur lengths relative to FST 
environmental distance, environmental factors do remain is suggestive of divergent selection on flower morphology. When 
significant in the dbRDA conditioned on geographic distance. At morphological variation is partitionable across populations, 
a smaller local scale, geographic distance was not a significant selection by pollinators has been documented as an underlying 
factor explaining genetic structure among populations, but mechanism promoting its retention in Platanthera bifolia (Boberg 
environmental factors were in both MMRR and dbRDA (Table 3; et al., 2014), Disa draconis Sw. (Johnson and Steiner, 1997), and 
Supplementary Figure S3). Gymnadenia odoratissima (L.) Rich. (Sun et  al., 2014). Though 
Frontiers in Ecology and Evolution 10 frontiersin.org
144
Wallace and Bowles 10.3389/fevo.2023.1085938
flowers of P. dilatata fit the primitive “settling moth” syndrome concentrations, most nutrients are bound in OM in peat soils and 
characteristic of Noctuidae moths (Hapeman and Inoue, 1997), are not available for plant uptake, especially under acidic conditions 
regional differences in primary and secondary pollinators readily (Verhoeven, 1986; Vitt and Chee, 1990). Indeed PHS, which was 
occur. This variation includes seven Noctuidae species and a relatively high in muskeg (Figure 2E), was negatively correlated 
Hesperiidae butterfly in Newfoundland, Canada (Boland, 1993), a with CEC (r = −0.4046, p = 0.0403). Mycorrhizal fungi may increase 
Noctuidae species in Oregon, United States (Larson, 1992), three efficiency of mineral uptake in peat soils and could provide a 
Bombus Latreille bumblebee species, a Noctuidae, and a competitive advantage as well as a favorable germination site for 
Nymphalidae butterfly in British Columbia, Canada (Van der Voort orchids in these habitats (Rasmussen, 1995). Unlike Ceratobasidium, 
et  al., 2022), and three Noctuidae, a Sphingidae moth, and a which can utilize N from both ammonium and nitrate, Tulasnella 
butterfly in Southeast Alaska (this study; Bowles and Armstrong, requires ammonium as a N source (Fochi et al., 2017). Ammonium 
2021). Such variation in primary and secondary pollinator types is the predominant form of N in dissolved nutrient concentrations 
and abundance could drive selection for variable flower morphology in bogs and fens our study area, but it is much more highly 
at regional and local scales. concentrated in fens (Fellman et al., 2008). This could explain the 
Nectar spur length in Platanthera determines whether an insect greater presence of Tulasnella in fens in our study area and might 
can access nectar and how pollinia are attached and pollen are suggest that the larger phenotype uses these fungi. However, it is 
deposited on the stigma, and selection should shift spur length unknown whether obligate relationships exist between the large and 
toward pollinators that maximize fitness (Boberg et al., 2014). Such small P. dilatata phenotypes and different fungal species. Other 
selection could be rapid if pollinators remain constant and gene habitats associated with the large-flowered phenotype also tended 
flow from other populations is infrequent. In this study, Noctuidae to have greater fertility and association with the Tulasnella fungus. 
moths were the most commonly observed pollinators on both the If these mycorrhizal fungi occur in different environments because 
small- and large-flowered forms at all elevations. The hawkmoth of nutrient availability (Fochi et al., 2017; Thixton et al., 2020) and 
Hyles gallii (Sphingidae) was observed only on large-flowered plants have strong relationships with P. dilatata phenotypes, then they 
at low elevations and appeared to carry greater pollen loads than could reinforce their habitat selection. Given the importance of 
did Noctuidae pollinators. Because hawkmoths have a longer mycorrhizae to orchid life history, research to understand the 
proboscis (ca. 25 mm; Miller, 1997) than Noctuidae moths (< potential for mycorrhizae to impose selection on orchid phenotypes 
11 mm; Zenker et al., 2011; Zhang et al., 2021), they may be more would also be useful.
effective pollinators for longer-spurred orchids than are Noctuidae 
(Tao et al., 2018). Based on iNaturalist observations (n = 27), the 
median elevation at which H. gallii has been observed in SE Alaska 4.2. Concordance between morphological 
is less than 50 m (range 1–285 m), much lower than the median differentiation and genetic differentiation
elevation (215 m, range 10–600 m) of muskegs in the study area. If 
H. gallii is prevalent in non-muskeg habits, then selection for longer Whereas previous studies identified significant morphological 
spurs is expected to drive flower morphology to match the most and genetic divergence in P. dilatata at broad geographic scales, 
efficient pollinator (Johnson and Steiner, 1997; Boberg et al., 2014; they have not previously correlated genetic differentiation with 
Sun et al., 2014) despite counter-selection from the more frequent phenotypic divisions (Wallace, 2003a; Adhikari and Wallace, 2014). 
Noctuidae moth pollinators across the study area. By contrast, if The allelic variation reported in this study provides the strongest 
plants in muskeg bogs are visited only by shorter tongued Noctuidae indication yet that a shared evolutionary history connects 
species, then selection is expected to drive flowers toward shorter phenotypically similar populations and distinguishes these from 
spurs and lead to adaptation of those plants to muskeg habitat. phenotypically dissimilar populations. The genetic dataset has also 
Absence or rarity of Sphingidae moths at higher elevations could revealed some major differences between the phenotypic groups. 
limit gene flow across an altitudinal gradient. Further work is For example, small-flowered populations that are not strongly 
needed to characterize pollinators and their selection for flower size admixed have lower allelic variation and heterozygosity and greater 
and nectar spur length for P. dilatata across its distribution to test population differentiation compared to large-flowered populations. 
this hypothesis of localized selection. These results suggest greater isolation, which could occur due to 
A strong correspondence between phenotype and soil lower density of populations and reduced gene flow over widely 
conditions has not been previously noted for this species and spaced muskeg habitats, as non-forest habitats cover only 17% of 
suggests the possibility that other factors might also influence the landscape in Southeast Alaska.
phenotypic variation. While phenotypic plasticity often underlies The deep genetic divergence in phenotypic groups may also 
phenotypic variation that aligns with environmental differences reflect historical divergence, perhaps associated with Pleistocene 
(Schlichting, 1986), it has rarely been documented for flower traits refugia in this area (Carrara et al., 2007; Marr et al., 2008; Geml 
(Sultan, 2000; Pélabon et al., 2011), and the alpine population HMA et  al., 2010; Shafer et  al., 2010). The Alexander Archipelago of 
also occurs in a muskeg-like habitat at high elevation yet retains a Southeast Alaska contains more than 2,000 individual islands and 
large-flowered phenotype similar to populations at lower elevations. stretches across 16,000 km of coastline (Carrara et al., 2007), giving 
This suggests that flower size is not a plastic trait, and that soil the region’s extensive topographical and geographical complexity 
fertility does not influence flower size. An alternative hypothesis is that undoubtedly influences gene flow and population isolation. The 
that the small-flowered phenotype is a stress tolerant poor impact of the last glacial period was heterogenous across Southeast 
competitor that is adapted to the skewed soil chemistry of muskeg Alaska, with numerous refugia proposed along the western edges of 
habitat. Although bog habitats may appear to have adequate base the Alexander Archipelago and exposed areas of the continental 
Frontiers in Ecology and Evolution 11 frontiersin.org
145
Wallace and Bowles 10.3389/fevo.2023.1085938
shelf (Carrara et al., 2007). More recently, successional changes in pollinators to facilitate repeated introgression and backcrossing 
coastal vegetation were associated with uplift following the Little Ice with newly colonized small-flowered populations in the area. All 
Age between 1770–1790 (Motyka, 2003). With isostatic changes but one of the hybrid populations sampled (GI) are located within 
continuing to occur, extensive uplift meadows may have rapidly 250 km of a large-flowered population in the study area, which is 
developed in the area (e.g., Auffret and Cousins, 2018). While these consistent with the maximum distance for seed dispersal that was 
habitats could represent an earlier successional stage relative to suggested by Phillips et al. (2012).
muskeg, they may harbor older genetic lineages if they persisted While seed dispersal may have led to colonization of small-
during glaciation. Phylogeographic studies would be  useful to flowered forms in areas containing large-flowered populations, 
understand the evolutionary and historical connections among pollinators must be the agents of gene flow leading to introgression. 
populations in the study area and the presence of multiple refugia Different spur lengths are expected to reduce cross-pollination 
within the Alexander Archipelago or dispersal from other refugia between phenotypes, but not to prevent it. Because hawkmoths are 
in northwestern North America. strong fliers that may easily cross between habitats, gene flow may 
be more easily mediated from larger flowers to smaller flowers. 
Noctuidae moths can also transport pollen long distances (Hendrix 
4.3. Hybridization between divergent et al., 1987), but this may be more likely during migration. Whereas 
phenotypes pollinaria adhere to the proboscis of both pollinators, they would 
adhere closer to the eyes of Noctuidae moths visiting longer-
While most populations we studied have at least one admixed spurred flowers of P. dilatata (Figure 3) than for hawkmoths. In 
sample (Figures 4A,B), the extensive and cryptic introgression that short-spurred flowers, positioning of pollinaria closer to the 
characterized several small-flowered populations was unexpected proboscis tip for hawkmoths might facilitate contact with the 
as these populations are morphologically similar to other small- column leading to successful cross-pollination. Thus, even 
flowered populations sampled in muskeg bogs. While other studies occasional visits to these populations by hawkmoths carrying 
have reported cryptic introgression, for example in Protea pollinia from long-spurred populations could have long-lasting 
L. (Mitchell and Holsinger, 2018) and in Lomatia R. Br. (McIntosh impacts because an orchid pollinarium contains enough pollen to 
et al., 2014), these studies also found hybrids with both genetic and fertilize thousands of ovules. Many inter-specific hybrids are known 
morphological intermediacy. In our study, morphologically within Platanthera (Wallace, 2003b; Brown, 2004; Alcantara et al., 
intermediate populations were not readily detected when averaged 2006; Brown et al., 2008; Wettewa et al., 2020; Hartvig et al., 2022), 
across samples. Nevertheless, at the individual level, statistical indicating that spur length does not consistently prevent cross-
outliers representing larger flowers were observed in muskeg pollination and pollinators readily move pollen between species.
populations and could indicate admixed individuals due to 
pollinator-mediated gene flow from large-flowered populations 
(M.L. Bowles, unpublished data). 4.4. Genetic structure and factors 
Phillips et  al. (2012) suggested that seed dispersal between influencing gene flow
populations at regional scales (e.g., < 250 km) is likely common, but 
gene flow might be more limited at larger geographic scales. Our Factors determining genetic structure may vary across the 
analyzes indicated that genetic distance among populations reflects landscape and across spatial scales. We expected that across the 
isolation by distance at large scales but not at small scales, consistent extent of the study area, which is nearly 500 km, geographic distance 
with the patterns described by Phillips et  al. (2012). The low would be  important because of limited gene flow. By contrast, 
incidence of admixture detected by STRUCUTRE and NewHybrids within the areas where seed dispersal can occur over shorter 
in the four hybrid populations suggests that introgression may have distances or pollinators are capable of flying between sites, 
occurred swiftly and early in their history. Given the commonality environmental factors are expected to be  more important 
of small-flowered populations in muskeg, we  suggest it is more determinants of genetic structure. In the study area, both elevation 
likely that these small-flowered populations were colonized from and habitat differences might influence gene flow at varying scales. 
other small-flowered populations, rather than large-flowered When considering all populations, both geographic distance and 
populations. If gene flow occurred early in their establishment and habitat (i.e., elevation, soils, and climate) are significant predictors 
was not maladaptive, then it would persist in the growing of genetic distance. Nevertheless, consistent with our hypothesis, 
population. The alternative explanation for the genetic similarity of geographic distance explained more of the observed variation in 
hybrid populations to large-flowered populations, that they genetic distances than environmental factors did (Table  3; 
originated from large-flowered colonizers of muskeg bogs that Supplementary Figure S2). The extensive topographic variation of 
subsequently evolved smaller flowers, seems less likely in the Southeast Alaska could impose barriers to gene flow if orchid seeds 
absence of a functional basis for variation in flower size due to are not able to move between mountains and the orchids are 
climate or soils. adapted to soil types or interact with other organisms, e.g., 
The contemporary presence of large-flowered populations in mycorrhizae or pollinators, that are themselves restricted by 
more diverse habitats may indicate greater historical abundance environmental factors.
across the landscape compared to small-flowered populations and At a smaller spatial scale encompassing the central populations, 
muskeg habitats (Auffret and Cousins, 2018). To produce extensive ca. 50 km north-to-south, we found that geographic distance was 
and cryptic introgression in the small-flowered populations, large- not predictive of genetic distance. This suggests that seeds and/or 
flowered plants would need to be  nearby and accessible to pollinators readily move about populations at this scale. By contrast, 
Frontiers in Ecology and Evolution 12 frontiersin.org
146
Wallace and Bowles 10.3389/fevo.2023.1085938
environmental factors were found to significantly influence genetic scales, which is needed to evaluate the cohesiveness of P. dilatata 
distance, indicating the presence of habitat barriers to successful populations and to quantify the geographic scale of discord in 
movement. This is expected if the small and large-flowered morphological and genetic divergence.
populations are adapted to different habitats or are limited by 
symbiotic partners that are themselves adapted to these differing 
habitats as noted for larger geographic scales. The difference in the 5. Conclusion
pattern of genetic structure across spatial scales demonstrates a 
complexity of landscapes in how they influence population By studying genetic, morphological and habitat diversity at the 
connectivity. While these results suggest that orchid seeds and or regional scale in P. dilatata we have identified novel patterns, yet 
pollen may readily move about, we are unable to discern the relative consistency with previous studies on this species. Strong genetic 
importance of these factors for gene flow. Additionally, the complex divergence between flower groups suggests the presence of distinct 
history of this region has undoubtedly impacted the patterns evolutionary lineages within Southeast Alaska. Evidence of 
observed today, but without a phylogeographic context we  also bidirectional gene flow between flower forms, nevertheless, indicates 
cannot account for how historical factors have influenced the that they are not reproductively isolated. Although orchid seeds are 
genetic structure of P. dilatata in this region. Future studies that test considered capable of long-distance gene flow, our results indicate 
hypotheses about the locations of glacial refugia are important foci that gene flow most readily occurs only at shorter geographic 
for future studies of this species across western North America. distances, perhaps <50 km. Environmental factors also contribute 
significantly to genetic structure and could reflect adaptations of the 
orchids themselves to these habitats or adaptations of their symbiotic 
4.5. Taxonomic implications partners. Further studies are needed to understand the evolution of 
adaptation in this species and its phylogeographic history. Platanthera 
With four habitats and two phenotypic groups, eight unique dilatata should be considered a model system for understanding the 
combinations could characterize P. dilatata in Southeast Alaska. Yet, process of diversification in temperate orchids.
we found only five of these combinations as small-flowered plants 
are restricted to muskeg bogs and large-flowered plants are rarely 
found in these habitats. The deep genetic divergence between Data availability statement
groups of populations supports the inference that there are multiple 
evolutionary lineages in the study area. Nevertheless, placing these The original contributions presented in the study are included 
lineages within the current taxonomy of this species is difficult. in the article/Supplementary files, further inquiries can be directed 
Plants from Southeast Alaska have a mean spur length that exceeds to the corresponding author.
lip length (Figure  3A), which would place all of them in var. 
leucostachys. Yet, the range of spur and lip lengths measured on 
plants in the study area (spurs: 6.5–9 mm for small-flowered and Author contributions
9–12 mm for large-flowered; lips: 5.25–7.25 mm for small-flowered 
and 7–9.5 mm for large-flowered) encompasses or exceeds the LW and MB conceived of the study, collected data, wrote the 
lengths described for the three varieties by previous authors manuscript, and critically reviewed the manuscript. All authors 
(Sheviak, 2002; Wallace, 2003a; Sears, 2008; Adhikari and Wallace, contributed to the article and approved the submitted version.
2014; Supplementary Table S6) but lie primarily within vars. 
dilatata and leucostachys. These morphological measurements are 
not consistent with varietal circumscriptions by Sheviak (2002) or Funding
the suggestion that three varieties occur in Southeast Alaska.
An additional consideration in metric comparisons among Funding was provided through the Robert Stiffler Endowment 
studies is the presence of artifacts associated with measurement through Old Dominion University.
methods. It is difficult to measure nectar spur length because they 
are falcate; thus, intact spurs will appear shorter than flattened 
spurs. The source of the flowers for measurement (i.e., fresh, dried, Acknowledgments
or spirit-preserved) also influences measurements as preservation 
can introduce distortions (Bateman et al., 2013), and spur length We thank Naghmeh Moghimi for aid in extracting DNA from 
has been reported to increase over the flowering season in leaf samples, Robert Armstrong, Matt Goff, Judy Hall Jacobson, and 
individual plants (Sheviak, 2002). Kris Larson for aid in field sampling, Elizabeth Esselman for review 
Taxonomic revision of P. dilatata is warranted because the and for sharing data on mycorrhizal isolates from the studied 
division of three varieties is inadequate to explain the variation species, and three reviewers for comments that improved the 
encountered in many areas of the distribution. Furthermore, manuscript. Collection permits were graciously provided by the US 
ecological or pollination studies should explicitly include Forest Service to access sites within the Tongass National Forest. 
morphological measurements of samples, rather than simply giving The herbaria of the Juneau Botanical Club located at the Alaska 
a varietal designation, as this would provide more transparency in State Museum, US Forest Service Forestry Research Lab (Juneau, 
morphological variability of studied populations. Such data would AK), and the University of Alaska contributed digitized herbarium 
also contribute to a greater ability to synthesize variation at local records through ARCTOS that were helpful in this study.
Frontiers in Ecology and Evolution 13 frontiersin.org
147
Wallace and Bowles 10.3389/fevo.2023.1085938
Conflict of interest organizations, or those of the publisher, the editors and the 
reviewers. Any product that may be evaluated in this article, or 
The authors declare that the research was conducted in the claim that may be made by its manufacturer, is not guaranteed or 
absence of any commercial or financial relationships that could endorsed by the publisher.
be construed as a potential conflict of interest.
Supplementary material
Publisher’s note
The Supplementary material for this article can be found online 
All claims expressed in this article are solely those of the authors at: https://www.frontiersin.org/articles/10.3389/fevo.2023.1085938/
and do not necessarily represent those of their affiliated full#supplementary-material
References
Adhikari, B., and Wallace, L. E. (2014). Does the Platanthera dilatata (Orchidaceae) Brown, P. M., Smith, C., and Shriver, J. S. (2008). A new species of fringed Platanthera 
complex contain cryptic species or continuously variable populations? Plant Syst. Evol. from the Central Appalachian Mountains of Eastern North America. N. Am. Native 
300, 1465–1476. doi: 10.1007/s00606-013-0974-8 Orchid J. 14, 238–254.
Alcantara, S., Semir, J., and Solferini, V. N. (2006). Low genetic structure in an Calinski, R. B., and Harabasz, J. (1974). A dendrite method for cluster analysis. 
epiphytic Orchidaceae (Oncidium hookeri) in the Atlantic rainforest of South-Eastern Commun. Stat. 3, 1–27.
Brazil. Ann. Bot. 98, 1207–1213. doi: 10.1093/aob/mcl202 Carlsson, J. (2008). Effects of microsatellite null alleles on assignment testing. J. Hered. 
Andersen, H. E.. (1955). Climate in Southeast Alaska in Relation to Tree Growth. 99, 616–623. doi: 10.1093/jhered/esn048
Station Paper No. 3, Alaska Forest Research Center, Juneau, Alaska. Carrara, P. E., Ager, T. A., and Baichtal, J. F. (2007). Possible refugia in the Alexander 
Anderson, E. C., and Thompson, E. A. (2002). A model-based method for identifying Archipelago of southeastern Alaska during the late Wisconsin glaciation. Can. J. Earth 
species hybrids using Multilocus genetic data. Genetics 160, 1217–1229. doi: 10.1093/ Sci. 44, 229–244. doi: 10.1139/e06-081
genetics/160.3.1217 Cavalli-Sforza, L. L., and Edwards, A. W. F. (1967). Phylogenetic analysis–models and 
Andriamihaja, C. F., Ramarosandratana, A. V., Grisoni, M., Jeannoda, V. H., and estimation procedures. Am. J. Hum. Genet. 19, 233–257.
Besse, P. (2021). Drivers of population divergence and species differentiation in a recent Chapuis, M. P., and Estoup, A. (2007). Microsatellite null alleles and estimation of 
group of indigenous orchids (Vanilla spp.) in Madagascar. Ecol. Evol. 11, 2681–2700. doi: population differentiation. Molec. Biol. Evol. 24, 621–631. doi: 10.1093/molbev/msl191
10.1002/ece3.7224
Chung, M. Y., Nason, J. D., and Chung, M. G. (2004). Spatial genetic structure in 
Arditti, J., and Ghani, A. K. A. (2000). Numerical and physical properties of orchid populations of the terrestrial orchid Cephalanthera longibracteata (Orchidaceae). Am. J. 
seeds and their biological implications. New Phytol. 146:569. Bot. 91, 52–57. doi: 10.3732/ajb.91.1.52
Auffret, A. G., and Cousins, S. A. O. (2018). Land uplift creates important meadow Chybicki, I. J., and Burczyk, J. (2009). Simultaneous estimation of null alleles and 
habitat and a potential original niche for grassland species. Proc. R. Soc. B 285:20172349. inbreeding coefficients. J. Hered. 100, 106–113. doi: 10.1093/jhered/esn088
doi: 10.1098/rspb.2017.2349 Culley, T. M., Stamper, T. I., Stokes, R. L., Brzyski, J. R., Hardiman, N. A., 
Balkenhol, N., Waits, L. P., and Dezzani, R. J. (2009). Statistical approaches in Klooster, M. R., et al. (2013). An efficient technique for primer development and 
landscape genetics: an evaluation of methods for linking landscape and genetic data. application that integrates fluorescent labeling and multiplex PCR. Appl. Plant Sci. 
Ecography 32, 818–830. doi: 10.1111/j.1600-0587.2009.05807.x 1:1300027. doi: 10.3732/apps.1300027
Bateman, R. M., Rudall, P. J., and Moura, M. (2013). Systematic revision of Platanthera D’Amore, D. V., Edwards, R. T., Herendeen, P. A., Hood, E., and Fellman, J. B. (2015). 
in the Azorean archipelago: not one but three species, including arguably Europe’s rarest Dissolved organic carbon fluxes from hydropedologic units in Alaskan coastal temperate 
orchid. PeerJ 1:e218. doi: 10.7717/peerj.218 rainforest watersheds. Soil Sci. Soc. Am. J. 79, 378–388. doi: 10.2136/sssaj2014.09.0380
Bateman, R. M., and Sexton, R. (2008). Is spur length of Platanthera species in the D’Amore, D. V., Fellman, J. B., Edwards, R. T., and Hood, E. (2010). Controls on 
British Isles adaptively optimized or an evolutionary red herring? Watsonia 28, 1–21. dissolved organic matter concentrations in soils and streams from a forested wetland 
and sloping bog in Southeast Alaska. Ecohydrology 3, 249–261. doi: 10.1002/eco.101
Bienek, P. A., Bhatt, U. S., Thoman, R. L., Angeloff, H., Partin, J., Papineau, J., et al. 
(2012). Climate divisions for Alaska based on objective methods. J. Appl. Meteorol. Dakin, E. E., and Avise, J. C. (2004). Microsatellite null alleles in parentage analysis. 
Climatol. 51, 1276–1289. doi: 10.1175/JAMC-D-11-0168.1 Heredity 93, 504–509. doi: 10.1038/sj.hdy.6800545
Blondaeu Da Silva, S. B., and Da Silva, A. (2018). Pstat: An R package to assess Duda, R. O., and Hart, P. E.. (1973). Pattern Classification and Scene Analysis. New 
population differentiation in phenotypic traits. R J. 10, 447–454. doi: 10.32614/ York: Wiley.
RJ-2018-010 Earl, D. A., and vonHoldt, B. M. (2012). Structure harvester: a website and program 
Boberg, E., Alexandersson, R., Jonsson, M., Maad, J., Ågren, J., and Nilsson, L. A. for visualizing structure output and implementing the Evanno method, version 2.3.4. 
(2014). Pollinator shifts and the evolution of spur length in the moth-pollinated orchid Conservation Genet. Resour. 4, 359–361. doi: 10.1007/s12686-011-9548-7
Platanthera bifolia. Ann. Bot. 113, 267–275. doi: 10.1093/aob/mct217 Evanno, G., Regnaut, S., and Goudet, J. (2005). Detecting the number of clusters of 
individuals using the software STRUCTURE: a simulation study. Molec. Ecol. 14, 
Boland, J. T. (1993). The Floral Biology of Platanthera dilatata (Pursh) Lindl. 2611–2620. doi: 10.1111/j.1365-294X.2005.02553.x
(Orchidaceae) [M.S. Thesis]. St. John’s (Newfoundland, Canada): Memorial University.
Excoffier, L., Smouse, P. E., and Quattro, J. M. (1992). Analysis of molecular variance 
Borcard, D., and Legendre, P. (2002). All-scale spatial analysis of ecological data by inferred from metric distances among DNA haplotypes: application to human 
means of principal coordinates of neighbour matrices. Ecol. Model. 153, 51–68. doi: mitochondrial DNA restriction data. Genetics 131, 479–491. doi: 10.1093/
10.1016/S0304-3800(01)00501-4 genetics/131.2.479
Borcard, D., Legendre, P., Avois-Jacquet, C., and Tuomisto, H. (2004). Dissecting the Fellman, J. V., and D’Amore, D. V. (2007). Nitrogen and phosphorus mineralization in 
spatial structure of ecological data at multiple scales. Ecology 85, 1826–1832. doi: three wetland types in Southeast Alaska, USA. Wetlands 27, 44–53. doi: 
10.1890/03-3111 10.1672/0277-5212(2007)27[44:NAPMIT]2.0.CO;2
Bowles, M. L., and Armstrong, R. H. (2021). Pollination, pilfering, and predation in Fellman, J. B., D’Amore, D. V., Hood, E., and Boone, R. D. (2008). Fluorescence 
an orchid pollinator network in the Juneau area of Southeast Alaska. Newslett. Alaska characteristics and biodegradability of dissolved organic matter in forest and wetland 
Entomol. Soc. 14, 12–25. soils from coastal temperate watersheds in Southeast Alaska. Biogeochemistry 88, 
Brommer, J. E. (2011). Whither Pst? The approximation of Qst by Pst in evolutionary 169–184. doi: 10.1007/s10533-008-9203-x
and conservation biology. J. Evol. Biol. 24, 1160–1168. doi: Fick, S. E., and Hijmans, R. J. (2017). WorldClim 2: new 1km spatial resolution climate 
10.1111/j.1420-9101.2011.02268.x surfaces for global land areas. Int. J. Climatol. 37, 4302–4315. doi: 10.1002/joc.5086
Brown, P. M. (2004). Understanding Platanthera chapmanii (Orchidaceae), its origins Fochi, V., Chitarra, W., Kohler, A., Voyron, S., Singan, V. R., Lindquist, E. A., et al. 
and hybrids. Sida 21, 853–859. (2017). Fungal and plant gene expression in the Tulasnella calospora–Serapias vomeracea 
Frontiers in Ecology and Evolution 14 frontiersin.org
148
Wallace and Bowles 10.3389/fevo.2023.1085938
symbiosis provides clues about nitrogen pathways in orchid mycorrhizas. New Phytol. Mitchell, N., and Holsinger, K. E. (2018). Cryptic natural hybridization between two 
213, 365–379. doi: 10.1111/nph.14279 species of Protea. S. Afr. J. Bot. 118, 306–314. doi: 10.1016/j.sajb.2017.12.002
Geml, J., Tulloss, R. E., Laursen, G. A., Sazanoa, N. A., and Taylor, D. L. (2010). Motyka, R. J. (2003). Little Ice Age subsidence and post Little Ice Age uplift in Juneau, 
“Ectomycorrhizal Basidiomycete, Amanita muscaria, suggest forest refugia in Alaska Alaska, inferred from dendrochronology and geomorphology. Quat. Res. 59, 300–309. 
during the Last Glacial Maximum” in Relict Species. eds. J. C. Habel and T. Assmann doi: 10.1016/S0033-5894(03)00032-2
(Berlin: Springer)
Neiland, B. J. (1971). The forest-bog complex of Southeast Alaska. Vegetatio 22, 1–64. 
Hapeman, J. R., and Inoue, K. (1997). “Plant-pollinator interactions and floral doi: 10.1007/BF01955719
radiation in Platanthera (Orchidaceae)” in Molecular Evolution and Adaptive Radiation. 
eds. T. J. Givnish and K. J. Sytsma (Cambridge, UK: Cambridge University Press), Nobarinezhad, M. H., and Wallace, L. E. (2022). Genetic structure and phenotypic 
433–454. diversity in Chamaecrista fasciculata (Michx.) Greene (Fabaceae) at its range edge. 
Castanea 87, 129–146. doi: 10.2179/0008-7475.87.1.129
Hartvig, I., Evans, S., Ligon, J., Eserman, L., Coffey, E., and McCormick, M. (2022). 
Quantitative morphological variation in the orange and white fringed orchids Oksanen, J., Blanchet, F. G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., et al. 
(Platanthera) in the eastern U.S. Orchids 91, 532–539. (2020). Vegan: Community Ecology Package. R Package Version 2.5–7. https://CRAN.R-
project.org/package=vegan/. (Accessed October 26, 2022).
Hendrix, W. H. III, Mueller, T. F., Phillips, J. R., and Davis, O. K. (1987). Pollen as an 
indicator of long-distance movement of Heliothis zea (Lepidoptera: Noctuidae). Environ. Paetkau, D., Slade, R., Burden, M., and Estoup, A. (2004). Direct, real-time estimation 
Entomol. 16, 1148–1151. doi: 10.1093/ee/16.5.1148 of migration rate using assignment methods: a simulation-based exploration of accuracy 
and power. Mol. Ecol. 13, 55–65. doi: 10.1046/j.1365-294X.2004.02008.x
Hijmans, R. J., Karney, C., Williams, E., and Vennes, C. (2021). Geosphere: Spherical 
Trigonometry. R Package Version 1.5–14. Available at: https://CRAN.R-project.org/ Peakall, R., and Smouse, P. E. (2012). GenAlEx 6.5: genetic analysis in Excel. 
package=geosphere (Accessed October 26, 2022). Population genetic software for teaching and research-an update. Bioinformatics 28, 
2537–2539. doi: 10.1093/bioinformatics/bts460
Hintze, J.. (2013). NCSS 9. Kaysville, Utah: NCSS, LLC.
Pélabon, C., Armbruster, W. S., and Hansen, T. F. (2011). Experimental evidence for 
Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scand. J. Stat. the Berg hypothesis: vegetative traits are more sensitive than pollination traits to 
6, 65–70. environmental variation. Funct. Ecol. 25, 247–257. doi: 10.1111/j.1365-2435.2010.01770.x
Horton, M. L.. (2011). Recommended soil testing procedures for the northeastern Phillips, R. D., Dixon, K. W., and Peakall, R. (2012). Low population genetic 
United  States. Revised Third Edition Northeastern Regional Publication No. 493. differentiation in the Orchidaceae: implications for the diversification of the family. Mol. 
Agricultural Experiment Stations of Connecticut, Delaware, Maine, Maryland, Ecol. 21, 5208–5220. doi: 10.1111/mec.12036
Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, 
Vermont, and West Virginia. Pinheiro, F., Dantas-Queiroz, M. V., and Palma-Silva, C. (2018). Plant species 
complexes as models to understand speciation and evolution: A review of South 
IBM Corp. (2020). IBM SPSS Statistics for Windows (Version 27). Armonk, New York: American studies. CRC Crit. Rev. Plant Sci. 37, 54–80. doi: 
IBM Corp. 10.1080/07352689.2018.1471565
Jakobsson, M., and Rosenberg, N. A. (2007). CLUMPP: a cluster matching and Piry, S., Alapetite, A., Cornuet, J.-M., Paetkau, D., Baudouin, L., and Estoup, A. (2004). 
permutation program for dealing with label switching and multimodality in analysis of GENECLASS2: A software for genetic assignment and first-generation migrant 
population structure. Bioinformatics 23, 1801–1806. doi: 10.1093/bioinformatics/ detection. J. Hered. 95, 536–539. doi: 10.1093/jhered/esh074
btm233
Johnson, S. D., and Steiner, K. E. (1997). Long-tongued fly pollination and evolution Pritchard, J. K., Stephens, M., and Donnelly, P. (2000). Inference of population 
of floral spur length in the Disa draconis complex (Orchidaceae). Evolution 51, 45–53. structure using multilocus genotype data. Genetics 155, 945–959. doi: 10.1093/
doi: 10.2307/2410959 genetics/155.2.945
Kahl, S. M., Kappel, C., Joshi, J., and Lenhard, M. (2021). Phylogeography of a widely R Core Team. (2022). R: A Language and Environment for Statistical Computing. 
distributed plant species reveals cryptic genetic lineages with parallel phenotypic Vienna, Austria: R Foundation for Statistical Computing. Available at: http://www.R-
responses to warming and drought conditions. Ecol. Evol. 11, 13986–14002. doi: project.org/. (Accessed October 26, 2022).
10.1002/ece3.8103 Ramasamy, R. K., Ramasamy, S., Bindroo, B. B., and Naik, V. G. (2014). STRUCTURE 
Kierepka, E. M., and Latch, E. K. (2015). Performance of partial statistics in individual- PLOT: a program for drawing elegant STRUCTURE bar plots in user friendly interface. 
based landscape genetics. Mol. Ecol. Resour. 15, 512–525. doi: 10.1111/1755-0998.12332 SpringerPlus 3:431. doi: 10.1186/2193-1801-3-431
Kirchoff, M., Smith, M., and Walker, N. (2016). “Biological setting: land cover and Rannala, B., and Mountain, J. L. (1997). Detecting immigration by using multilocus 
forest vegetation” in Ecological Atlas of Southeast Alaska. ed. M. A. Smith (Alaska: genotypes. Proc. Natl. Acad. Sci. U. S. A. 94, 9197–9201. doi: 10.1073/pnas.94.17.9197
Audubon), 44–50. Rasmussen, H. N. (1995). Terrestrial Orchids: From Seed to Mycotrophic Plant. 
Larson, R. J. (1992). Pollination of Platanthera dilatata var. dilatata in Oregon by the Cambridge, United Kingdom: Cambridge University Press.
noctuid moth Discestra oregonica. Madrono 39, 236–237. Raymond, M., and Rousset, F. (1995). GENEPOP: population genetics software for 
Legendre, P., and Anderson, M. J. (1999). Distance-based redundancy analysis: testing exact tests and ecumenicism, version 1.2. J. Hered. 86, 248–249. doi: 10.1093/
multispecies responses in multifactorial ecological experiments. Ecol. Monogr. 69, 1–24. oxfordjournals.jhered.a111573
doi: 10.1890/0012-9615(1999)069[0001:DBRATM]2.0.CO;2 Robertson, J. L., and Wyatt, R. (1990). Evidence for pollination ecotypes in the yellow-
Liu, R., Wang, H., Yang, J. B., Corlett, R. T., Randle, C. P., Li, D. Z., et al. (2022). fringed orchid, Platanthera ciliaris. Evolution 44, 121–133. doi: 10.2307/2409528
Cryptic species diversification of the Pedicularis siphonantha complex (Orobanchaceae) Rosenberg, M. S., and Anderson, C. D. (2011). PASSaGE: Pattern analysis, spatial 
in the mountains of Southwest China since the Pliocene. Front. Plant Sci. 13:811206. doi: statistics, and geographic exegesis. Version 2. Methods Ecol. Evol. 2, 229–232. doi: 
10.3389/fpls.2022.811206 10.1111/j.2041-210X.2010.00081.x
Luer, C. A.. (1975). The Native Orchids of the United States and Canada, Excluding Rousset, F. (2008). Genepop’007: a complete reimplementation of the Genepop 
Florida. New York: Botanical Garden. software for Windows and Linux. Molec. Ecol. Resour. 8, 103–106. doi: 
Marr, K. L., Allen, G. A., and Hebda, R. J. (2008). Refugia in the Cordilleran ice sheet 10.1111/j.1471-8286.2007.01931.x
of western North America: chloroplast DNA diversity in the Arctic–alpine plant Oxyria Schlichting, C. D. (1986). The evolution of phenotypic plasticity in plants. Annu. Rev. 
digyna. J. Biogeogr. 35, 1323–1334. doi: 10.1111/j.1365-2699.2007.01879.x Ecol. Syst. 17, 667–693. doi: 10.1146/annurev.es.17.110186.003315
Mathewes, R. W. (1985). “Paleobotanical evidence for climatic change in southern Sears, C. J. (2008). Morphological discrimination of Platanthera aquilonis, P. 
British Columbia during late-glacial and Holocene time” in Climatic Change in Canada huronensis, and P. dilatata (Orchidaceae) herbarium specimens. Rhodora 110, 389–405. 
5: Critical Periods in the Quaternary Climatic History of Northern North America. ed. C. doi: 10.3119/07-24.1
R. Harington, vol. 55 (Ottawa, Canada: National Museums of Canada, National Museum 
of Natural Sciences, Syllogeus Series), 397–442. Shafer, A. B. A., Cullingham, C. I., Côté, S. D., and Coltman, D. W. (2010). Of glaciers 
and refugia: a decade of study sheds new light on the phylogeography of northwestern 
McCune, B., and Mefford, M. J.. (2011). PC-ORD. Multivariate Analysis of Ecological North America. Mol. Ecol. 19, 4589–4621. doi: 10.1111/j.1365-294X.2010.04828.x
Data. Version 6.08. Gleneden Beach, Oregon: MjM Software Design.
Sheviak, C. J. (2002). “Platanthera” in Flora of North America: Magnoliophyta: Liliidae: 
McIntosh, E. J., Rossetto, M., Weston, P. H., and Wardle, G. M. (2014). Maintenance Liliales and Orchidales: Editorial Committee, vol. 26 (New York: University Press), 
of strong morphological differentiation despite ongoing natural hybridization between 551–571.
sympatric species of Lomatia (Proteaceae). Ann. Bot. 113, 861–872. doi: 10.1093/aob/
mct314 Shulski, M., and Wendler, G. (2007). The Climate of Alaska. Fairbanks, Alaska: 
Melton, C. (2020). Symbiotic germination of Spiranthes cernua and S. vernalis in Illinois University of Alaska Press.
and identification of mycorrhizal fungus associates of Platanthera dilatata in Alaska. [M.S. Stone, C. S. (1993). Vegetation of coastal marshes near Juneau, Alaska. Northwest Sci. 
Thesis]. Edwardsville, IL: Southern Illinois University. 67, 215–230.
Miller, W. E. (1997). Diversity and evolution of tongue length in hawkmoths Sultan, S. E. (2000). Phenotypic plasticity for plant development, function, and life 
(Sphingidae). J. Lepid. Soc. 51, 9–31. history. Trends Plant Sci. 5, 537–542. doi: 10.1016/S1360-1385(00)01797-0
Frontiers in Ecology and Evolution 15 frontiersin.org
149
Wallace and Bowles 10.3389/fevo.2023.1085938
Sun, M., Gross, K., and Schiestl, F. P. (2014). Floral adaptation to local pollinator guilds Vitt, D. H., and Chee, W. L. (1990). The relationships of vegetation to surface water 
in a terrestrial orchid. Ann. Bot. 113, 289–300. doi: 10.1093/aob/mct219 chemistry and peat chemistry in fens of Alberta, Canada. Vegetatio 89, 87–106. doi: 
Surveswaran, S., Gowda, V., and Sun, M. (2018). Using an integrated approach to 10.1007/BF00032163
identify cryptic species, divergence patterns, and hybrid species in Asian ladies’ tresses Wallace, L. E. (2003a). An evaluation of taxonomic boundaries in Platanthera dilatata 
orchids (Spiranthes, Orchidaceae). Mol. Phylogenet. Evol. 124, 106–121. doi: 10.1016/j. (Orchidaceae) based on morphological and molecular variation. Rhodora 105, 322–336.
ympev.2018.02.025 Wallace, L. E. (2003b). Molecular evidence for allopolyploid speciation and recurrent 
Tao, Z. B., Ren, Z. X., Bernhardt, P., Wang, W. J., Liang, H., Li, H. D., et al. (2018). origins in Platanthera huronensis (Orchidaceae). Int. J. Plant Sci. 164, 907–916. doi: 
Nocturnal hawkmoth and noctuid moth pollination of Habenaria limprichtii 10.1086/378658
(Orchidaceae) in sub-alpine meadows of the Yulong Snow Mountain (Yunnan, China). Wang, I. J. (2013). Examining the full effects of landscape heterogeneity on spatial 
Bot. J. Linn. Soc. 187, 483–498. doi: 10.1093/botlinnean/boy023 genetic variation: a multiple matrix regression approach for quantifying geographic and 
Thixton, H. L., Esselman, E. J., Corey, L., and Zettler, L. W. (2020). Further evidence ecological isolation. Evolution 67, 3403–3411. doi: 10.1111/evo.12134
of Ceratobasidium D. P. Rogers (Basidiomycota) serving as the ubiquitous fungal Wettewa, E., Bailey, N., and Wallace, L. E. (2020). Comparative analysis of genetic and 
associate of Platanthera leucophaea (Orchidaceae) in the North American tallgrass morphological variation in the Platanthera hyperborea complex (Orchidaceae). Syst. Bot. 
prairie. Bot. Stud. 61:12. doi: 10.1186/s40529-020-00289-z 45, 767–778. doi: 10.1600/036364420X16033962925303
Van der Niet, T., Peakall, R., and Johnson, S. D. (2014). Pollinator-driven ecological Wilson, G. A., and Rannala, B. (2003). Bayesian inference of recent migration rates 
speciation in plants: new evidence and future perspectives. Ann. Bot. 113, 199–212. doi: using multilocus genotypes. Genetics 163, 1177–1191. doi: 10.1093/genetics/163.3.1177
10.1093/aob/mct290
Zenker, M. M., Penz, C., de Paris, M., and Specht, A. (2011). Proboscis morphology 
Van der Voort, G. E., Gilmore, S. R., Gorrell, J. C., and Janes, J. K. (2022). Continuous and its relationship to feeding habits in noctuid moths. J. Insect Sci. 11:42. doi: 
video capture, and pollinia tracking, in Platanthera (Orchidaceae) reveal new insect 10.1673/031.011.0142
visitors and potential pollinators. PeerJ. 10:e13191. doi: 10.7717/peerj.13191 Zhang, C. M., Niu, Y., Hu, G. L., and Lu, J. Q. (2021). Ultramorphological comparison 
Verhoeven, J. T. A. (1986). Nutrient dynamics in minerotrophic peat mires. Aquat. of proboscis and associated sensilla of Scotogramma trifolii and Protoschinia scutosa 
Bot. 25, 117–137. doi: 10.1016/0304-3770(86)90049-5 (Lepidoptera: Noctuidae). Insects 12:992. doi: 10.3390/insects12110992
Frontiers in Ecology and Evolution 16 frontiersin.org
150
TYPE Original Research
PUBLISHED 20 April 2023
DOI 10.3389/fevo.2023.1057940
Diversity and specificity of orchid 
OPEN ACCESS mycorrhizal fungi in a leafless 
EDITED BY
Tiiu Kull,  epiphytic orchid, Dendrophylax 
Estonian University of Life Sciences, Estonia
REVIEWED BY lindenii and the potential role of 
Raymond L. Tremblay,  
University of Puerto Rico, Puerto Rico
María Isabel Mujica,  fungi in shaping its fine-scale 
Pontificia Universidad Católica de Chile, Chile
*CORRESPONDENCE distribution
Lynnaun J. A. N. Johnson  
 Lynnaunjohnson2018@u.northwestern.edu
SPECIALTY SECTION Lynnaun J. A. N. Johnson 1,2*, Michael E. Kane 3, 
This article was submitted to  Lawrence W. Zettler 4 and Gregory M. Mueller 2
Conservation and Restoration Ecology,  
a section of the journal  1 Northwestern University, Evanston, IL, United States, 2 Negaunee Institute for Plant Conservation 
Frontiers in Ecology and Evolution Science and Action, Chicago Botanic Garden, Glencoe, IL, United States, 3 Environmental Horticulture 
RECEIVED 30 September 2022 Department, University of Florida, Gainesville, FL, United States, 
4 Department of Biology, Illinois College, 
ACCEPTED 29 March 2023 Jacksonville, IL, United States
PUBLISHED 20 April 2023
CITATION Orchids grow in diverse habitats worldwide with most (approximately 69%) growing 
Johnson LJAN, Kane ME, Zettler LW and 
Mueller GM (2023) Diversity and specificity of on trees as epiphytes. Although orchid mycorrhizal fungi have been identified as 
orchid mycorrhizal fungi in a leafless epiphytic potential drivers for terrestrial orchid distribution, the influence of these fungi on 
orchid, Dendrophylax lindenii and the potential the fine-scale distribution of epiphytic orchids is poorly understood. In this study, 
role of fungi in shaping its fine-scale 
distribution. we investigated the mycorrhizal fungal community and fine-scale distribution of 
Front. Ecol. Evol. 11:1057940. Dendrophylax lindenii, a rare and endangered epiphytic orchid that is leafless when 
doi: 10.3389/fevo.2023.1057940 mature. We used amplicon sequencing to investigate the composition of orchid 
COPYRIGHT mycorrhizal fungi in the roots of 39 D. lindenii individuals in their natural habitat, 
© 2023 Johnson, Kane, Zettler and Mueller. 
This is an open-access article distributed under the swamps of Florida. We compared the orchid mycorrhizal fungi of D. lindenii 
the terms of the Creative Commons Attribution to those of co-occurring epiphytic orchids, as well as to the orchid mycorrhizal 
License (CC BY). The use, distribution or fungal communities of bark from potential host trees, with and without D. lindenii. 
reproduction in other forums is permitted, 
provided the original author(s) and the Our results show that D. lindenii has a high specificity for a single Ceratobasidium 
copyright owner(s) are credited and that the species, which is widely distributed on phorophytes and detected in both wet and 
original publication in this journal is cited, in dry periods in the orchid’s habitat. This Ceratobasidium species was mostly absent 
accordance with accepted academic practice. 
No use, distribution or reproduction is or only recorded in low frequency in the roots of co-occurring epiphytic orchids. 
permitted which does not comply with these Phylogenetic analysis documented that this Ceratobasidium was conspecific 
terms. with the strain that is used to germinate D. lindenii ex-situ. However, our findings 
suggest that laboratory germinated adult D. lindenii transplanted into the field had 
lower read abundances of this Ceratobasidium compared to naturally occurring 
plants. These findings suggest that this orchid mycorrhizal fungus may play a 
significant role in the fine-scale distribution of naturally occurring D. lindenii.
KEYWORDS
conservation, Ceratobasidium, host tree specificity, amplicon sequencing, ghost orchid
Introduction
Mycorrhizal fungi are well known mutualists that are essential for their plant partners’ 
abundance and spatial distribution (Smith and Read, 2010; McCormick and Jacquemyn, 2014). 
While ca. 69% of orchid species are tropical epiphytes (Zotz, 2016), little is known about the 
orchid mycorrhizal fungi (OMF) they associate with compared to temperate terrestrial orchids. 
Frontiers in Ecology and Evolution 01 frontiersin.org
151
Johnson et al. 10.3389/fevo.2023.1057940
Epiphytic orchids, like their terrestrial counterparts, enlist OMF to 
facilitate seed germination and seedling development, but it remains 
unclear to what degree epiphytes continue to utilize OMF into 
maturity (Dearnaley et al., 2012; Rasmussen et al., 2015; Selosse et al., 
2022). Stable isotope work by Gebauer et al. (2016) revealed that a 
greater number than previously thought of orchids are likely reliant 
on OMF, and are functioning as myco-heterotrophs even though they 
are photosynthetic as adults. This finding of likely orchid dependence 
on OMF as adults, especially epiphytic orchids, raises the question of 
the potential role that OMF play in driving their fine-scale 
spatial distribution.
The drivers of fine scale epiphyte spatial distribution and host tree 
(phorophyte) specificity have been debated within the literature for 
FIGURE 1
over a century since the writings of Schimper (1888) see review by (A) Flowers of Dendrophylax lindenii (photo by Larry W. Richardson). 
Wagner et al. (2015). Debate has focused on the role of various abiotic (B) Dendrophylax lindenii roots growing on the tree trunk of a 
factors (e.g., microclimate and host bark characteristics) and biotic phorophyte, Fraxinus caroliniana.
factors (e.g., symbiotic fungi and co-occurrence with moss). Research 
by McCormick et al. (2018) has demonstrated that while OMF may 
restrict terrestrial orchid distributions at local scales, at broad during two periods, flooded and not flooded in the area of its natural 
geographic scales terrestrial orchids are not constrained by OMF. Most distribution in the United States (Supplementary Figures S1, S2).
of these findings were established for terrestrial orchids, with Dendrophylax lindenii, also known as the Ghost Orchid, is 
investigations of epiphytic orchids still pending (Li et  al., 2021). restricted to southwestern Florida and the western tip of Cuba (Brown, 
Recently, studies have investigated fungal communities in the bark of 2002) where it remains vulnerable to poaching and environmental 
phorophytes of epiphytic orchids, which providing insights into changes (Mújica et  al., 2018, 2021). In Florida, less than 1,500 
phorophyte specificity and spatial distribution of epiphytic orchids individuals are thought to remain (Haaland et al., 2022), and in Cuba, 
(Izuddin et al., 2019; Eskov et al., 2020; Pecoraro et al., 2021; Petrolli the number is even fewer [<500; (Mújica et al., 2018)]. In the Florida 
et al., 2021, 2022). Eskov et al. (2020) further explored OMF and Panther National Wildlife Refuge where about 1/3rd of the state’s 
revealed that fungi colonizing epiphytic orchid roots were significantly Ghost Orchids are found, Mújica et  al. (2018) calculated that 
different from the phorophytes’ branches. Pecoraro et  al. (2021) D. lindenii numbers will decline by 20% during the next decade. 
studied the phorophyte specificity of two epiphytic orchid species, as Consequently, the species is now a candidate for U.S. Federal 
well as the environmental factors influencing the relationship between protection under the Endangered Species Act (Haaland et al., 2022). 
the orchids and their phorophytes. They concluded that the orchid The Florida habitats of D. lindenii consist of cypress domes and strand 
phorophyte associations were influenced by the phorophyte bark’s swamps in the Big Cypress Basin. According to a study by Mújica et al. 
OMF communities and potentially its pH and water holding capacity. (2018) in 2015, 69% of the growth of D. lindenii in Florida is found on 
Recent studies have also revealed examples of a strong fungal the trunks of Fraxinus caroliniana Mill., while occurring less 
specificity of epiphytic orchids associated with a single OMF species, frequently (36%) on Annona glabra L. These trees are typically located 
Ceratobasidium or Tulasnellaceae species (Rammitsu et  al., 2019, in the lower canopy under Taxodium distichum (L.) Rich. (Brown, 
2020) despite inconclusive early studies (Gowland et al., 2013; Wang 2002; Stewart and Richardson, 2008). Although D. lindenii grows in a 
et al., 2017). moist habitat, it experiences dry periods during the region’s dry season 
Amplicon sequencing, a type of environmental sequencing, is a which lacks any standing water (Mújica et al., 2018).
cost-effective advancement for investigating fungal communities Like all orchids, D. lindenii requires OMF for germination (Hoang 
compared to traditional culture-based methods (McCormick and et al., 2017). Early seedling stages of D. lindenii have a rudimentary 
Jacquemyn, 2014; McCormick et al., 2018). Ectomycorrhizal (ECM) ephemeral leaf. As an adult, the orchid lacks leaves and shoots and 
fungi as well as some OMF are known to be recalcitrant to being photosynthesizes predominantly via its roots (Benzing and Ott, 1981; 
cultured and recent studies utilizing amplicon sequencing have Benzing et al., 1983; Hoang et al., 2017). Benzing and Ott (1981), have 
detected a diversity of ECM fungi in the roots of both epiphytic and shown that the mature roots of D. lindenii utilizes CAM 
terrestrial orchids (Selosse et  al., 2022). Additionally, amplicon photosynthesis, and Chomicki et  al. (2014) using microscopy 
sequencing can increase the detection of potential OMF in epiphytic hypothesized that it forms a mutualism with an OMF 
orchid roots compared to Sanger Sequencing as Sanger Sequencing is (Ceratobasidiaceae) to obtain carbon to supplement its photosynthesis. 
often limited by the need for first culturing the fungi (Waud et al., Furthermore, seed germination experiments by Hoang et al. (2017) 
2016; Jacquemyn et al., 2017; Novotná et al., 2018; Johnson et al., 2021). and Mújica et al. (2018) have confirmed that D. lindenii associates with 
We chose the rare leafless epiphytic orchid Dendrophylax lindenii a Ceratobasidium and that this fungus is present in mature roots.
(Lindl.) Bentham ex Rolfe (Figure 1) as our study taxon to further Our primary aim for this study was to identify the OMF associated 
document OMF communities of rare tropical epiphytic orchids and with D. lindenii and to investigate the potential role of OMF in 
to examine the potential role of OMF as drivers of their phorophyte influencing its fine-scale distribution within naturally occurring 
specificity. In addition to sampling the roots of D. lindenii we sampled populations (i.e., why it was found on some potential phorophytes and 
co-occurring epiphytic orchids and the bark of potential phorophytes not on others). We tested two hypotheses: (1) D. lindenii has a specific 
with and without D. lindenii to uncover evidence of OMF specificity community of OMF compared to co-occurring epiphytic orchids; and 
Frontiers in Ecology and Evolution 02 frontiersin.org
152
Johnson et al. 10.3389/fevo.2023.1057940
(2) the OMF colonizing D. lindenii are found in the bark of D. lindenii Additionally, we conducted a pilot study to assess the success of 
phorophytes in higher abundances than in the bark of trees without amplicon sequencing of root tips for revealing the OMF community 
D. lindenii. Given that D. lindenii is currently state-listed as of D. lindenii. We obtained 50 mm root samples collected from three 
endangered, we  primarily restricted our sampling to root tips to mature individuals of D. lindenii at the FPNWR. A root of a D. lindenii 
minimize damage to the plant. To investigate if additional OMF were that was home cultivated from Redlands, Florida was also sampled. 
missed with this sampling method we also investigated the fungal For this pilot study, roots were cut into 5 mm long segments starting 
community of four whole roots. from the tip and labelled alphabetically (i.e., A, B, C, etc. see 
Supplementary Figure S4).
Materials and methods
DNA extraction, PCR amplification, and 
Study sites, tree bark and orchid root amplicon sequencing
sampling
Approximately 5 mm of root tip and bark tissue was collected and 
During 2016 and 2018 we collected >100 root and bark samples stored in cetyltrimethylammonium bromide (CTAB) buffer. Root and 
from five sites at the Florida Panther National Wildlife Refuge bark samples were surface sterilized with 70% ethanol, and 50% 
(FPNWR) a 10,684 ha area (Supplementary Table S1). Four of the sites Clorox® (2.6% sodium hypochlorite) using the method outlined in 
were natural habitats for D. lindenii. The fifth site lacked naturally Bayman et al. (1997). Next, genomic DNA was extracted from root 
occurring plants but had D. lindenii explants that were samples using the Qiagen DNeasy extraction kits (Qiagen, Valencia, 
micropropagated under axenic conditions in the lab and subsequently CA, United States) following the manufacturer’s instructions. DNA 
transplanted (attached) on appropriate species of trees. The site with from bark samples was extracted with the modified CTAB method of 
explants we identified as Site 4 in our study. Most of the sites were Murray and Thompson (1980), and for difficult to extract samples the 
dominated (over 90%) by F. caroliniana as the main phorophyte. Sites MOBIO Power Soil DNA Extraction kit (MOBIO Laboratories, 
were either sloughs or strand swamps and were separated by about Carlsbad, CA, United  States) was used following the 
1 km from each other. When we collected samples in 2016 (March), manufacturer’s instruction.
FPNWR sites all had standing water in sloughs and swamps The extracted genomic DNA from the 2016 root samples was 
(Supplementary Figure S1), but all sites were dry (not flooded) when amplified using the primers: ITS86f (5′- GTGAATCATCGAA 
we sampled in 2018 (April) (Supplementary Figure S2). This sampling TCTTTGAA-3′; Turenne et al., 1999) and ITS4 (5′- TCCTCCGCT 
period in 2018 was unusually dry. The precise sites at the FPNWR are TATTGATATGC-3′; White et  al., 1990). These fungal primers 
not disclosed herein because D. lindenii and several co-occurring (ITS86F/ITS4) amplify the internal transcribed region ITS, the 
orchids are state-listed as endangered and remain highly vulnerable to standard fungal barcode, for ITS subregion 2 which is shown to 
poaching. For each site, Special Use collecting permits were obtained be effective for delimiting OMF such as those in the Cantharellales.
(USFWS, OMB Control # 1018–0102), and permission to access and Next, amplicons from the PCR products were produced using a 
sample D. lindenii populations was subsequently granted. three step PCR sequencing protocol (see Johnson et al., 2021 materials 
In March 2016, root samples were collected from four sites at the and methods). This included PCR steps that used modified primers 
FPNWR Sites 1–4 (Supplementary Table S1). Root samples were with indices from the Nextera XT kit for 96 indices to sequence 
collected from the leafless epiphytic orchid species D. lindenii (n = 9) 2 × 250 bp. The final amplicon libraries generated for root and bark 
and several co-occurring epiphytic orchids: Campylocentrum samples were quantified using a Qubit dsDNA HS kit (Invitrogen) and 
pachyrrhizum (Rchb.f.) Rolfe (n = 3), Dendrophylax porrectus (Rchb.f.) a Bioanalyzer-Agilent 2100 (Agilent Technologies, Santa Clara, CA, 
Carlsward & Whitten (n = 6), Epidendrum amphistomum A. Rich. United States). Final amplicon libraries for root and bark samples were 
(n = 4), Epidendrum nocturnum Jacq. (n = 1) and Prosthechea cochleata pooled together in equimolar concentrations and the final pool was 
(L.) W. E. Higgins (n = 3). Simultaneous with the collection of root then sequenced on an Illumina MiSeq at the Pritzker Lab at the Field 
samples, bark samples were collected from phorophytes adjacent of all Museum (Chicago, IL).
epiphytic orchids (Supplementary Table S1). The root sections from the pilot study, root tips, and bark samples 
In April 2018, sampling of the roots of an additional 27 D. lindenii from 2018 were PCR amplified using modified fungal primers ITS86F/
plants was carried out at the original four sites plus one additional site ITS4 with barcodes supplied from Novogene Bioinformatics Institute 
(Site 5). Concurrent with the root tissue collection, bark samples (Beijing, China) following the protocol applied in 2016. The generated 
(n = 57) were collected from phorophytes of D. lindenii and trees final amplicon libraries were pooled to equimolar concentrations then 
without D. lindenii (Supplementary Table S1). Five trees with and five shipped to Novogene and sequenced on an Illumina HiSeq. Sequences 
trees without D. lindenii individuals were sampled at each of the five generated from this study were submitted to NCBI’s Sequence Read 
sites. The sampling design considered the position of D. lindenii on the Archive under the BioProject PRJNA948888.
tree and bark samples were collected from (1) the base of the tree 
trunk, (2) above D. lindenii, (3) the side of roots of D. lindenii root; and 
(4) the opposite side of the tree trunk (Supplementary Figure S3). In Bioinformatics and statistical analyses
instances where D. lindenii was not present bark samples were 
collected from the base of the tree and three additional samples were Initially, bioinformatics analyses were performed on roots 
taken at a height of at breast height (1.5 M) from base, where and bark collected in 2016 separately. Subsequently, the sequences 
D. lindenii would typically grow. obtained from the bark, root, and root sections of the 2018 
Frontiers in Ecology and Evolution 03 frontiersin.org
153
Johnson et al. 10.3389/fevo.2023.1057940
dataset were integrated, and bioinformatics analyses were Results
conducted on these samples collectively to determine patterns of 
similar Operational Taxonomic Units (OTUs) between Sequence analyses of root and bark 
sample types. samples
To conduct bioinformatic analyses, the sequences were first 
quality filtered, followed by OTU clustering utilizing the PIPITS Fungal sequence data were obtained from roots of D. lindenii and 
pipeline (version 1.4.0) default settings as described by Gweon et al. other co-occurring epiphytic orchids collected in the field in 2016 and 
(2015). Briefly, PIPITS joined reads and quality filtered short reads 2018. Root samples in 2016 and 2018 yielded 537,371 (n = 26) and 
(<50 bp), extracted non ITS fungal reads with the script ITSx 1,691,086 reads (n = 30), respectively, resulting in the identification of 
(Bengtsson-Palme et al., 2013), then clustered OTUs at 95% sequence 526 and 1,077 Operational Taxonomic Units (OTUs) at the 95% 
similarity. Additional PIPITS scripts assigned taxonomy to OTUs with sequence identity level. Sequences generated from sections of whole 
the Ribosomal Database Project Classifier [a Naïve Bayesian Classifier roots yielded 3,205,959 reads (n = 37 root section samples) and 
(Wang et al., 2007)] and the UNITE database (Nilsson et al., 2019). resulted in the identification of 1,372 OTUs. In 2016, we collected 30 
Sequences for the HiSeq dataset was analyzed separately from the bark samples and sequencing yielded 693,482 reads with a total of 550 
2016 MiSeq dataset. The single difference between analysis of the OTUs. Most phorophytes sampled in 2016 were from F. caroliniana 
MiSeq data analyses and HiSeq data analyses was omitting the ITSx (over 90%) with a small proportion of A. glabra also being sampled 
step for the HiSeq data. (Supplementary Table S1). Additionally, a bark sample was collected 
To further investigate differences between fungal communities from a Taxodium distichum that had an explant affixed to it 
we filtered rare OTUs that were less than 1,000 sequences, and the raw (Supplementary Table S1). In 2018 we successfully sequenced 57 bark 
read abundances were then normalized with Cumulative Sum Scaling samples mostly from F. caroliniana, from trees with D. lindenii (n = 43) 
in the R package metagenomeSeq (Paulson et al., 2013). All statistical and trees without D. lindenii (n = 14), yielding 7,245,995 reads with a 
analyses were conducted within R (R Core Team, 2022). The total of 1,141 OTUs resolved.
visualization of abundance of sequences was first accomplished using The increase in read and OTU counts in 2018 can be attributed to 
Krona charts, which were generated using Krona-2.8.1 within R the use of the HiSeq platform instead of the MiSeq platform, as well 
(Ondov et al., 2011). Bar graphs showing relative and read abundances as the greater sampling intensity of root samples. The average OTU 
were produced with the R package ggplot2. To better visualize richness observed in root and bark samples for 2016 was 67 and 55 
differences between read abundances, the y-axis was truncated using respectively, whereas the OTU richness for 2018 samples for roots was 
the R package ggbreak (Xu et al., 2021). 233 and 304 for bark (Supplementary Table S1). Unfortunately, no 
Principal coordinate analysis (PCoA) was generated using amplicon libraries were generated for root samples collected at Site 
Bray-Curtis distances with the R package vegan and visualized 2 in 2016 as the library preps were unsuccessful resulting in sequence 
with ggplot2 (Wickham, 2016). Significance between fungal data that was unsuitable for data analysis. The analysis of the pilot 
communities of D. lindenii and epiphytic orchid roots, phorophyte study examining potential differences in fungal communities in 
and the bark of trees without D. lindenii present, and location of different sections of entire roots of both cultivated and wild collected 
sites were determined with “permutational manova” (Anderson, D. lindenii documented that fungal communities were similar across 
2001) in R package vegan (adonis2 function) by first permuting the all sections but were different between cultivated vs. wild collected 
raw data with 9,999 permutations (Oksanen et al., 2022). Prior to plants (Supplementary Figure S5).
executing adonis2 for the permutational multivariate analysis of 
variance (PERMANOVA) we also investigated the dispersion for 
groups using another vegan function betadisper. In addition, Ceratobasidiaceae is the dominant OMF 
pairwise comparisons were completed for the PERMANOVA using associated with Dendrophylax lindenii
the pairwiseAdonis R package with Bonferroni corrections 
(Martinez Arbizu, 2017). p-values that are < 0.05 were The fungal communities of naturally occurring D. lindenii roots 
considered significant. across all sites were observed to be similar and dominated by several 
Phylogenetic analyses were undertaken to investigate Ceratobasidiaceae, even during the flooded (2016) (Figure  2; 
relationships among the community of recovered Ceratobasidiaceae Supplementary Figure S6A) and not flooded (2018) periods 
sequences from 2016 and 2018 root and bark fungal samples. The (Supplementary Figure S6B). The dominant Ceratobasidiaceae OTUs 
phylogenetic tree incorporated Ceratobasidium sequences from associated with D. lindenii were OTU 11 (recovered from bark 
NCBI GenBank. A sequence of Tulasnella from the UNITE fungal samples), OTU 14 (recovered from 2016 root samples); and OTU 76 
database was used as an outgroup. We used MUSCLE (Edgar, 2004) (recovered from both 2018 root and bark samples). Phylogenetic 
in AliView version 1.27 (Larsson, 2014) for multiple sequence analysis resolved each of these dominant Ceratobasidium OTUs as 
alignments and also used AliView to generate a Maximum part of a well-supported monophyletic clade, Clade 2, and are 
Likelihood tree using the default settings of the program FastTree considered conspecific (Figure  3). A visual analysis of sequence 
version 2.1.10 (Price et  al., 2009). The final tree was rooted and alignments further supports this finding, as Clade 2 OTUs exhibit a 
visualized using FigTree version 1.4.4.1 sequence similarity of more than 98%. Ceratobasidium Clade 2 
includes other individuals that were previously recovered from mature 
roots of D. lindenii. Included in Clade 2 is Dlin-394, which was derived 
from cultures isolated from mature roots of D. lindenii and has been 
1 http://tree.bio.ed.ac.uk/software/figtree/ used to germinate seeds of D. lindenii (Hoang et al., 2017).
Frontiers in Ecology and Evolution 04 frontiersin.org
154
Johnson et al. 10.3389/fevo.2023.1057940
FIGURE 2
Relative abundance of fungal OTUs (putative OMF) obtained from root samples of D. lindenii and co-occurring epiphytic orchids in 2016. 
Dendrophylax lindenii explants at site 4 are bolded and are labeled with an asterisk.
The pilot study undertaken to assess if the OMF community Diversity of OMF in bark of Dendrophylax 
recovered from root tips of D. lindenii provided was reflective of the lindenii phorophytes and trees without 
full root OMF community provided further evidence of the Dendrophylax lindenii
dominance of Ceratobasidium Clade 2  in naturally occurring 
D. lindenii. The majority of root sections from naturally occurring Unlike the root fungal community, the bark fungal community 
D. lindenii, including the root tips, were dominated by Ceratobasidium was not dominated by Ceratobasidium Clade 2 or other 
Clade 2 (Supplementary Figure S5). Of note, root section samples of Ceratobasidiaceae OTUs. Ceratobasidiaceae OTUs accounted for less 
the home-cultivated D. lindenii lacked Ceratobasidiaceae OTUs. than 5% of the total reads (Supplementary Figures S7A,B). 
Instead, an abundance of Ascomycota OTU reads (Lasiodiplodia OTU Nonetheless, Ceratobasidium Clade 2 was present in most bark 
1174 and Diaporthales OTU 627) were recovered samples (Figure 4). Other putative OMF detected in bark samples 
(Supplementary Figure S5). were also rare and included a few Serendipitaceae and Tulasnella 
Ceratobasidium Clade 2 was present in most samples of OTUs. Additional rare OTUs also included ECM fungi including 
D. lindenii (Supplementary Figure S6C); however, it was absent or Mycena, Russula, Thelephoraceae, and Tomentella.
only had very low read numbers from the roots of co-occurring The principal coordinate analysis (PCoA) and subsequent 
epiphytic orchids (Figure 2). Other taxa of Ceratobasidiaceae were PERMANOVA tests on root samples collected in both 2016 and 
recovered from these orchids, e.g., Ceratobasidiaceae OTUs 19 and 2018 showed significant differences between sites and orchid 
22 were abundant in root samples of D. porrectus (Figure 2). These species (Supplementary Figures S8A,B). Specifically, the 2016 root 
OTUs belonged to different clades (Clade 1 and 3, Figure 3). Other samples reveal potentially significant differences in both orchid 
OMF taxa that were recovered from co-occurring epiphytic orchid species (PERMANOVA: F5, 25 = 2.11, R2 = 0.29, p < 0.05, betadisper: 
roots collected in 2016 (Figure 2; Supplementary Figures S6A,B) F =  2.08, p = 0.11) and location (PERMANOVA: F2, 25 = 3.20, 
included taxa of putative OMF Serendipitaceae and ECM fungi such R2 = 0.18, p < 0.05, betadisper: F = 5.09, p =  0.01). Furthermore, 
as Inocybaceae, Russulaceae, Scleroderma, Thelephoraceae, pairwise comparisons showed significant differences between Site 
Tomentella, and Tuber species. While present in lower proportions 4 (site with only introduced D. lindenii lab grown explants) and 
(<1%) we did not detect a widespread presence of Tulasnellaceae the two other sites, Site 1 (adjusted p = 0.009) and Site 3 (adjusted 
OTUs, a traditional OMF. p = 0.003). In addition, Site 4 also differed from Site 3 (adjusted 
Frontiers in Ecology and Evolution 05 frontiersin.org
155
Johnson et al. 10.3389/fevo.2023.1057940
FIGURE 3
Maximum likelihood phylogeny of putative Ceratobasidiaceae species based on ITS data set of 95 taxa using FastTree constructed with default 
parameters. A cultured Tulasnella was used as an outgroup taxon. Bootstrap support values above 70% are reported. Sequences generated during this 
study are indicated in red.
p = 0.03). Similarly, the D. lindenii root samples collected in 2018 data (corresponding PCoA is Supplementary Figure S9B) revealed 
revealed significant differences by site (PERMANOVA: F4,24 = 1.26, differences between Site 1 and Site 3 (adjusted p = 0.01); differences 
R2 = 0.20, p = 0.036, betadisper: F =  1.24, p > 0.5). However, between Site 2 and Site 5 (adjusted p = 0.03); and differences between 
pairwise comparisons revealed no significant differences between Site 2 and Site 4 (adjusted p = 0.05).
sites when adjusted p values were generated.
PCoA of bark data collected for 2016 (Supplementary Figure S9A) 
revealed significant differences for both location (PERMANOVA: Discussion
F4, 34 = 1.82, R2 = 0.16, p < 0.05, betadisper: F = 1.47, p > 0.05) and the 
presence of D. lindenii (PERMANOVA: F4, 34 = 1.50, R2 = 0.03, p = 0.028, Our study provides strong evidence that D. lindenii may have 
betadisper: F = 0.69, p = 0.4). Although sites were not different during a high specificity for a single Ceratobasidiaceae OTU 
the flooded period of 2016, pairwise comparisons of the 2018 bark (Ceratobasidium Clade 2) in its natural habitat at the Florida 
Frontiers in Ecology and Evolution 06 frontiersin.org
156
Johnson et al. 10.3389/fevo.2023.1057940
FIGURE 4
Read abundance of Ceratobasidium clade 2 obtained from bark collected in 2018 from trees with and without naturally occurring D. lindenii. Bar graph 
is condensed to better visualize the sample with the highest read abundance. Trees without naturally occurring D. lindenii are represented in bold and 
have an asterisk. All bark samples shown in graph are from F. caroliniana obtained at various positions on the trunk except the base (see 
Supplementary Table S1 for bark sample positions).
Panther National Wildlife Refuge (FPNWR). This OTU was found be  tested. Nevertheless, these orchids associated with different 
to be abundant in D. lindenii roots, and rare (<1% of total reads) Ceratobasidium. For example, Ceratobasidium OTU 19 and 22 were 
in other co-occurring epiphytic orchids at the detected primarily in D. porrectus, another leafless epiphytic orchid. 
FPNWR. Ceratobasidium Clade 2 was also widespread at all sites We hypothesize, with a caveat of small sample size, that mature roots 
in the bark of phorophytes with D. lindenii and potential of leafless epiphytic orchids are dominated by a single OMF unique to 
phorophyte trees without D. lindenii during both flooded (2016) that species.
and not flooded periods (2018). In addition to traditional OMF, we detected low read abundances 
This apparent extreme fungal specificity for one OMF, of ECM fungi in the roots of the epiphytic orchids examined. This is 
Ceratobasidium Clade 2, is similar to that reported for in contrast to aerial roots of V. planifolia which were heavily colonized 
mycoheterotrophic orchids (McKendrick et al., 2002; Selosse et al., by ECM fungi (Johnson et  al., 2021). Vanilla planifolia is a 
2002), terrestrial orchids (Thixton et al., 2020), and some epiphytic hemiepiphytic orchid and it is possible that the ECM fungi in the 
orchids (Otero et  al., 2002, 2004; Graham and Dearnaley, 2012; aerial roots are from systemic colonization emanating from the 
Rammitsu et  al., 2019, 2021a,b). Our findings of potential high terrestrial roots. ECM fungi have been commonly reported from 
specificity with Ceratobasidium Clade 2 aligns with previous studies terrestrial orchids, but except for those detected by Johnson et al. 
demonstrating the importance of Ceratobasidium taxa supporting (2021) an abundance of ECM fungi has not been reported colonizing 
healthy populations of other epiphytic orchids. For instance, Qin et al. arial/epiphytic orchid roots.
(2021) and Rammitsu et al. (2019) reported on other leafless epiphytic Foliar orchids exhibited lower read abundances relative to 
orchids that have a high specificity for single Ceratobasidium species. D. lindenii and other leafless epiphytic orchids (data not shown) 
Furthermore, Ceratobasidium Clade 2 is conspecific (>99% similar) was observed in our study. We  hypothesize that the greater 
with Ceratobasidium (Dlin-394) that was isolated and brought into photosynthetic capacity of foliar orchids provided by their leaves 
culture from roots of D. lindenii that was used to germinate D. lindenii reduces their dependence on OMF for supplemental 
seeds (Hoang et al., 2017). fungal carbon.
We also observed evidence of possible specificity in some of the Amplicon sequencing enabled us to document the presence of 
other co-occurring epiphytic orchids, but the sample size was small ECM fungi that are resistant to culturing in the orchid root 
for many of these epiphytic orchids and clear hypotheses could not communities. However, we were not successful in recovering species 
Frontiers in Ecology and Evolution 07 frontiersin.org
157
Johnson et al. 10.3389/fevo.2023.1057940
of Tulasnellaceae. Whether this was actually due to very low This study establishes the usefulness of amplicon sequencing as 
abundance of these species is not clear. Some primer bias of the primer a method to examine fungal communities in the roots of endangered 
pair ITS86F/ITS4 for Tulasnella species has been reported and this orchids such as D. lindenii. Sampling the actively growing root tips 
primer pair is likely poor for detecting Tulasnella spp. (Tedersoo et al., provides a non-destructive sampling method for future studies of this 
2015; Vogt-Schilb et al., 2020; Johnson et al., 2021; Rammitsu et al., and other threatened and endangered orchids. Naturally occurring 
2021a). Thus, future work using primers that are not biased towards D. lindenii appears to partner with a specific undescribed species of 
Tulasnellaceae is needed. Ceratobasidium. However, lab-grown explants of D. lindenii have low 
While bark is not a carbon source for orchids (Eskov et al., 2020), abundance so long-term survival and successful reintroduction to 
it is the likely source of the OMF that epiphytic orchids need for natural habitats should account for potential phorophytes with 
establishment including seed germination and seedling growth abundant Ceratobasidium present for a viable conservation method. 
(Rasmussen et al., 2015). Pellitier et al. (2019) documented that tree Additionally, while the presence of the required fungus is necessary 
bark can serve as an environmental filter for the fungal communities for establishment of the orchid on a particular tree, it is likely that 
available to epiphytic orchids. Thus, the distribution of OMF in tree other factors which impact its fine-scale distribution, are also 
bark throughout an orchid’s range could influence its fine-scale involved. Understanding how the difference in OMF abundance 
distribution. Ceratobasidium OTU Clade 2 was recovered from all between naturally occurring plants and explants and the factors 
trees with D. lindenii, the fungus was also recovered in low abundance influencing successful establishment on phorophytes are needed to 
from many potential phorophytes without the orchid, indicating that enhance the success of efforts to augment the population of the Ghost 
additional studies are necessary to comprehend the factors that Orchid and refine conservation actions.
contribute to the fine-scale distribution of D. lindenii beyond the 
presence of the required OMF. Although several A. glabra trees, the 
other phorophyte of D. lindenii in Florida, were sampled, we were not Data availability statement
successful in obtaining sequences from those samples. Thus, attempts 
should be made to sample sufficient numbers of A. glabra to better The datasets presented in this study can be found in online 
understand the situation in Florida. Additionally, D. lindenii in Cuba repositories. The name of the repository and accession number can be 
is found on several phorophyte species in comparison to the two found at: NCBI; PRJNA948888.
primary phorophyte species associated with D. lindenii in Florida. 
Therefore, a fuller understanding of factors influencing the fine-scale 
distribution of D. lindenii needs to include an analysis of Author contributions
Cuban phorophytes.
When present, Ceratobasidium Clade 2 in bark was recovered LJ and MC contributed to the study conception, 
at low read abundances, i.e., <5% relative abundance even from performed sample preparation, and data collection. LJ and GM 
bark samples collected adjacent to actively growing root tips of wrote the first draft of the manuscript with an initial review by 
D. lindenii. If Ceratobasidium Clade 2 is functioning as a saprobe LZ. All authors contributed with comments on the later 
in bark, then it is likely an inefficient saprobe and being versions of the manuscript and approved of the final 
outcompeted by more efficient saprotrophic fungi in the bark manuscript.
fungal community.
Although some F. caroliniana in Site 4 had Ceratobasidium Clade 
2 it may be below the threshold of abundance to facilitate establishment Funding
and support the growth of naturally occurring plants (McCormick 
et al., 2016). Understanding site differences in terms of the presence/ Financial support for this research was secured by LJ with a Fred 
abundance of Ceratobasidium Clade 2 and other factors influencing Case Grant from the Native Orchid Conference. Additional funds 
establishment is crucial to sustaining populations of D. lindenii and were provided by the Plant Biology and Conservation program, a joint 
preventing ‘senile’ populations’, an ageing orchid population that lacks graduate degree training program between Northwestern University 
seedling recruitment (Rasmussen et al., 2015). and the Chicago Botanic Garden.
The findings of this study indicate that D. lindenii, has high 
specificity for a specific taxon of Ceratobasidium, Clade 2. While 
this study provides data that suggest that the presence/absence (or Acknowledgments
very low abundance) of the required OMF influences which tree 
D. lindenii is likely to establish and persist, the fungus is probably We gratefully acknowledge Mark Danaher (USFWS) and Larry 
not the sole factor driving fine-scale distribution. Future studies Richardson (Richardson Nature) for their assistance with field work 
should focus on the role of abiotic factors, such as bark in the Florida Panther NWR, and other Refuge staff including 
characteristics like pH and phenolics, on Ceratobasidium growth, Mitchell Barazowski, Kevin Godsea, and Ben Nottingham. Thanks 
as well as the ability of the orchid to establish on the tree surface. are extended to Ernesto Mújica (Orquideo Soroa, Cuba) for 
Pellitier et al. (2019) demonstrated that fungal communities are providing helpful insight and suggestions, and Adam R. Herdman 
likely affected by pH and total phenolic content, therefore (Southern Illinois University-Edwardsville). We also extend thanks 
experiments to test this hypothesis should consider other to Renata Șerban for assisting with molecular methods for the 
abiotic factors. laboratory work.
Frontiers in Ecology and Evolution 08 frontiersin.org
158
Johnson et al. 10.3389/fevo.2023.1057940
Conflict of interest or those of the publisher, the editors and the reviewers. 
Any product that may be evaluated in this article, or claim that may 
The authors declare that the research was conducted in the be made by its manufacturer, is not guaranteed or endorsed by the 
absence of any commercial or financial relationships that could publisher.
be construed as a potential conflict of interest.
Supplementary material
Publisher’s note
The Supplementary material for this article can be found online 
All claims expressed in this article are solely those of the authors at: https://www.frontiersin.org/articles/10.3389/fevo.2023.1057940/
and do not necessarily represent those of their affiliated organizations, full#supplementary-material
References
Anderson, M. J. (2001). A new method for non-parametric multivariate analysis of Johnson, L. J. A. N., del Gónzalez-Chávez, M., Carrillo-González, R., Porras-Alfaro, A., 
variance. Austral Ecol. 26, 32–46. doi: 10.1111/j.1442-9993.2001.01070.pp.x and Mueller, G. M. (2021). Vanilla aerial and terrestrial roots host rich communities of 
Bayman, P., Lebron, L. L., Tremblay, R. L., and Lodge, D. J. (1997). Variation in orchid mycorrhizal and ectomycorrhizal fungi. Plants People Planet 3, 541–552. doi: 
endophytic fungi from roots and leaves of Lepanthes (Orchidaceae). New Phytol. 135, 10.1002/ppp3.10171
143–149. doi: 10.1046/j.1469-8137.1997.00618.x Larsson, A. (2014). AliView: a fast and lightweight alignment viewer and editor for 
Bengtsson-Palme, J., Ryberg, M., Hartmann, M., Branco, S., Wang, Z., Godhe, A., et al. large datasets. Bioinformatics 30, 3276–3278. doi: 10.1093/bioinformatics/btu531
(2013). Improved software detection and extraction of ITS1 and ITS2 from ribosomal Li, T., Wu, S., Yang, W., Selosse, M.-A., and Gao, J. (2021). How mycorrhizal 
ITS sequences of fungi and other eukaryotes for analysis of environmental sequencing associations influence orchid distribution and population dynamics. Front. Plant Sci. 
data. Methods Ecol. Evol. 4, 914–919. doi: 10.1111/2041-210X.12073 12:647114. doi: 10.3389/fpls.2021.647114
Benzing, D. H., Friedman, W. E., Peterson, G., and Renfrow, A. (1983). Shootlessness, Martinez Arbizu, P. (2017). PairwiseAdonis: Pairwise Multilevel Comparison Using 
velamentous roots, and the pre-eminence of Orchidaceae in the epiphytic biotope. Am. Adonis. R Package Version 0.4.1.
J. Bot. 70, 121–133. doi: 10.2307/2443212 McCormick, M. K., and Jacquemyn, H. (2014). What constrains the distribution of 
Benzing, D. H., and Ott, D. W. (1981). Vegetative reduction in epiphytic Bromeliaceae orchid populations? New Phytol. 202, 392–400. doi: 10.1111/nph. 
and Orchidaceae: its origin and significance. Biotropica 13:131. doi: 10.2307/2387715 12639
Brown, P. M. (2002). Wild Orchids of Florida, with References to the Atlantic and Gulf McCormick, M. K., Taylor, D. L., Whigham, D. F., and Burnett, R. K. (2016). 
Coastal Plains. Gainsville, Florida: University Press of Florida. Germination patterns in three terrestrial orchids relate to abundance of mycorrhizal 
Chomicki, G., Bidel, L. P. R., and Jay-Allemand, C. (2014). Exodermis structure fungi. J. Ecol. 104, 744–754. doi: 10.1111/1365-2745.12556
controls fungal invasion in the leafless epiphytic orchid Dendrophylax lindenii (Lindl.) McCormick, M. K., Whigham, D. F., and Canchani-Viruet, A. (2018). Mycorrhizal 
Benth. Ex Rolfe. Flora-Morphol. Distrib. Funct. Ecol. Plants 209, 88–94. doi: 10.1016/j. fungi affect orchid distribution and population dynamics. New Phytol. 219, 1207–1215. 
flora.2014.01.001 doi: 10.1111/nph.15223
Dearnaley, J. D. W., Martos, F., and Selosse, M.-A. (2012). “12 orchid Mycorrhizas: McKendrick, S. L., Leake, J. R., Lee Taylor, D., and Read, D. J. (2002). Symbiotic 
molecular ecology, physiology, evolution and conservation aspects” in Fungal germination and development of the myco-heterotrophic orchid Neottia nidus-avis in 
Associations. ed. B. Hock (Berlin, Heidelberg: Springer Berlin Heidelberg), 207–230. nature and its requirement for locally distributed Sebacina spp. New Phytol. 154, 
Edgar, R. C. (2004). MUSCLE: multiple sequence alignment with high accuracy and 233–247. doi: 10.1046/j.1469-8137.2002.00372.x
high throughput. Nucleic Acids Res. 32, 1792–1797. doi: 10.1093/nar/gkh340 Mújica, E. B., Herdman, A. R., Danaher, M. W., González, E. H., and Zettler, L. W. 
Eskov, A. K., Voronina, E. Y., Tedersoo, L., Tiunov, A. V., Manh, V., Prilepsky, N. G., (2021). Projected status of the ghost orchid (Dendrophylax lindenii) in Florida during 
et al. (2020). Orchid epiphytes do not receive organic substances from living trees the next decade based on temporal dynamic studies spanning six years. Plants. 10:1579. 
through fungi. Mycorrhiza 30, 697–704. doi: 10.1007/s00572-020-00980-w doi: 10.3390/plants10081579
Gebauer, G., Preiss, K., and Gebauer, A. C. (2016). Partial mycoheterotrophy is more Mújica, E. B., Mably, J. J., Skarha, S. M., Corey, L. L., Richardson, L. W., Danaher, M. W., 
widespread among orchids than previously assumed. New Phytol. 211, 11–15. doi: et al. (2018). A comparision of ghost orchid (Dendrophylax lindenii) habitats in Florida 
10.1111/nph.13865 and Cuba, with particular reference to seedling recruitment and mycorrhizal fungi. Bot. J. Linn. Soc. 186, 572–586. doi: 10.1093/botlinnean/box106
Gowland, K. M., van der Merwe, M. M., Linde, C. C., Clements, M. A., and 
Nicotra, A. B. (2013). The host bias of three epiphytic Aeridinae orchid species is Murray, M. G., and Thompson, W. F. (1980). Rapid isolation of high molecular weight 
reflected, but not explained, by mycorrhizal fungal associations. Am. J. Bot. 100, plant DNA. Nucleic Acids Res. 8, 4321–4326. doi: 10.1093/nar/8.19.4321
764–777. doi: 10.3732/ajb.1200411 Nilsson, R. H., Larsson, K.-H., Taylor, A. F. S., Bengtsson-Palme, J., Jeppesen, T. S., 
Graham, R. R., and Dearnaley, J. D. W. (2012). The rare Australian epiphytic orchid Schigel, D., et al. (2019). The UNITE database for molecular identification of fungi: 
Sarcochilus weinthalii associates with a single species of Ceratobasidium. Fungal Divers. handling dark taxa and parallel taxonomic classifications. Nucleic Acids Res. 47, D259–
54, 31–37. doi: 10.1007/s13225-011-0106-0 D264. doi: 10.1093/nar/gky1022
Gweon, H. S., Oliver, A., Taylor, J., Booth, T., Gibbs, M., Read, D. S., et al. (2015). Novotná, A., Benítez, Á., Herrera, P., Cruz, D., Filipczyková, E., and Suárez, J. P. 
PIPITS: an automated pipeline for analyses of fungal internal transcribed spacer (2018). High diversity of root-associated fungi isolated from three epiphytic orchids in 
sequences from the Illumina sequencing platform. Methods Ecol. Evol. 6, 973–980. doi: southern Ecuador. Mycoscience 59, 24–32. doi: 10.1016/j.myc.2017. 
10.1111/2041-210X.12399 07.007
Haaland, D., Estenoz, S., Williams, M., Director, P. D., and Williams, L. (2022). Oksanen, J., Simpson, G. L., Blanchet, F. G., Kindt, R., Legendre, P., Minchin, P. R., 
Petition to List the Ghost Orchid (Dendrophylax lindenii), as Threatened or Endangered et al. (2022). Vegan: Community Ecology Package. R Package Version 2.6-4.
Under the Endangered Species Act, and to Designate Critical Habitat. Ondov, B. D., Bergman, N. H., and Phillippy, A. M. (2011). Interactive metagenomic 
Hoang, N. H., Kane, M. E., Radcliffe, E. N., Zettler, L. W., and Richardson, L. W. visualization in a web browser. BMC Bioinformatics 12:385. doi: 
(2017). Comparative seed germination and seedling development of the ghost orchid, 10.1186/1471-2105-12-385
Dendrophylax lindenii (Orchidaceae), and molecular identification of its mycorrhizal Otero, J. T., Ackerman, J. D., and Bayman, P. (2002). Diversity and host specificity of 
fungus from South Florida. Ann. Bot. 119, 379–393. doi: 10.1093/aob/mcw220 endophytic Rhizoctonia-like fungi from tropical orchids. Am. J. Bot. 89, 1852–1858. doi: 
Izuddin, M., Srivathsan, A., Lee, A. L., Yam, T. W., and Webb, E. L. (2019). Availability 10.3732/ajb.89.11.1852
of orchid mycorrhizal fungi on roadside trees in a tropical urban landscape. Sci. Rep. Otero, J. T., Ackerman, J. D., and Bayman, P. (2004). Differences in mycorrhizal 
9:19528. doi: 10.1038/s41598-019-56049-y preferences between two tropical orchids. Mol. Ecol. 13, 2393–2404. doi: 
Jacquemyn, H., Waud, M., Brys, R., Lallemand, F., Courty, P. E., Robionek, A., et al. 10.1111/j.1365-294x.2004.02223.x
(2017). Mycorrhizal associations and trophic modes in coexisting orchids: an ecological Paulson, J. N., Colin Stine, O., Bravo, H. C., and Pop, M. (2013). Differential 
continuum between auto-and mixotrophy. Front. Plant Sci. 8:1497. doi: 10.3389/ abundance analysis for microbial marker-gene surveys. Nat. Methods 10, 1200–1202. 
fpls.2017.01497 doi: 10.1038/nmeth.2658
Frontiers in Ecology and Evolution 09 frontiersin.org
159
Johnson et al. 10.3389/fevo.2023.1057940
Pecoraro, L., Rasmussen, H. N., Gomes, S. I. F., Wang, X., Merckx, V. S. F. T., Cai, L., et al. Neottia nidus-avis (L.) L.C.M. rich. And neighbouring tree ectomycorrhizae. Mol. Ecol. 
(2021). Fungal diversity driven by bark features affects phorophyte preference in epiphytic 11, 1831–1844. doi: 10.1046/j.1365-294x.2002.01553.x
orchids from southern China. Sci. Rep. 11:11287. doi: 10.1038/s41598-021-90877-1 Smith, S. E., and Read, D. J. (2010). Mycorrhizal Symbiosis. 3rd edn New York, NY: 
Pellitier, P. T., Zak, D. R., and Salley, S. O. (2019). Environmental filtering structures Academic Press.
fungal endophyte communities in tree bark. Mol. Ecol. 28, 5188–5198. doi: 10.1111/
mec.15237 Stewart, S., and Richardson, L. (2008). Orchid flora of the Florida Panther National Wildlife Refuge. North Am. Native Orchid J. 14, 70–104.
Petrolli, R., Augusto Vieira, C., Jakalski, M., Bocayuva, M. F., Vallé, C., Cruz, E. D. S., 
et al. (2021). A fine-scale spatial analysis of fungal communities on tropical tree bark Tedersoo, L., Anslan, S., Bahram, M., Põlme, S., Riit, T., Liiv, I., et al. (2015). Shotgun 
unveils the epiphytic rhizosphere in orchids. New Phytol. 231, 2002–2014. doi: 10.1111/ metagenomes and multiple primer pair-barcode combinations of amplicons reveal 
nph.17459 biases in metabarcoding analyses of fungi. MycoKeys 10, 1–43. doi: 10.3897/
mycokeys.10.4852
Petrolli, R., Zinger, L., Perez-Lamarque, B., Collobert, G., Griveau, C., Pailler, T., et al. 
(2022). Spatial turnover of fungi and partner choice shape mycorrhizal networks in Thixton, H. L., Esselman, E. J., Corey, L. L., and Zettler, L. W. (2020). Further evidence 
epiphytic orchids. J. Ecol. 110, 2568–2584. doi: 10.1111/1365-2745.13986 of Ceratobasidium D.P. Rogers (Basidiomycota) serving as the ubiquitous fungal 
associate of Platanthera leucophaea (Orchidaceae) in the north American tallgrass 
Price, M. N., Dehal, P. S., and Arkin, A. P. (2009). Fasttree: computing large minimum prairie. Bot. Stud. 61:12. doi: 10.1186/s40529-020-00289-z
evolution trees with profiles instead of a distance matrix. Mol. Biol. Evol. 26, 1641–1650. 
doi: 10.1093/molbev/msp077 Turenne, C. Y., Sanche, S. E., Hoban, D. J., Karlowsky, J. A., and Kabani, A. M. (1999). 
Rapid identification of fungi by using the ITS2 genetic region and an automated 
Qin, J., Zhang, W., Feng, J.-Q., and Zhang, S.-B. (2021). Leafless epiphytic orchids fluorescent capillary electrophoresis system. J. Clin. Microbiol. 37, 1846–1851. doi: 
share Ceratobasidiaceae mycorrhizal fungi. Mycorrhiza 31, 625–635. doi: 10.1007/ 10.1128/JCM.37.6.1846-1851.1999
s00572-021-01043-4
Vogt-Schilb, H., Těšitelová, T., Kotilínek, M., Sucháček, P., Kohout, P., and Jersáková, J. 
R Core Team (2022). R: A Language and Environment for Statistical Computing. (2020). Altered rhizoctonia assemblages in grasslands on ex-arable land support 
Available at: https://www.R-project.org/. germination of mycorrhizal generalist, not specialist orchids. New Phytol. 227, 
Rammitsu, K., Abe, S., Abe, T., Kotaka, N., Kudaka, M., Kudaka, N., et al. (2021a). The 1200–1212. doi: 10.1111/nph.16604
endangered epiphytic orchid Dendrobium okinawense has a highly specific mycorrhizal Wagner, K., Mendieta-Leiva, G., and Zotz, G. (2015). Host specificity in vascular 
association with a single Tulasnellaceae fungus. J. For. Res. 26, 215–221. doi: epiphytes: a review of methodology, empirical evidence and potential mechanisms. AoB 
10.1080/13416979.2021.1876587 Plants 7:plu092. doi: 10.1093/aobpla/plu092
Rammitsu, K., Kajita, T., Imai, R., and Ogura-Tsujita, Y. (2021b). Strong primer bias Wang, Q., Garrity, G. M., Tiedje, J. M., and Cole, J. R. (2007). Naïve bayesian classifier 
for Tulasnellaceae fungi in metabarcoding: specific primers improve the characterization for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl. Environ. 
of the mycorrhizal communities of epiphytic orchids. Mycoscience 62, 356–363. doi: Microbiol. 73, 5261–5267. doi: 10.1128/AEM.00062-07
10.47371/mycosci.2021.06.005
Wang, X., Li, Y., Song, X., Meng, Q., Zhu, J., Zhao, Y., et al. (2017). Influence of host 
Rammitsu, K., Yagame, T., Yamashita, Y., Yukawa, T., Isshiki, S., and Ogura-Tsujita, Y. tree species on isolation and communities of mycorrhizal and endophytic fungi from 
(2019). A leafless epiphytic orchid, Taeniophyllum glandulosum Blume (Orchidaceae), roots of a tropical epiphytic orchid, Dendrobium sinense (Orchidaceae). Mycorrhiza 27, 
is specifically associated with the Ceratobasidiaceae family of basidiomycetous fungi. 709–718. doi: 10.1007/s00572-017-0787-7
Mycorrhiza 29, 159–166. doi: 10.1007/s00572-019-00881-7
Waud, M., Busschaert, P., Lievens, B., and Jacquemyn, H. (2016). Specificity and 
Rammitsu, K., Yukawa, T., Yamashita, Y., Isshiki, S., and Ogura-Tsujita, Y. (2020). The localised distribution of mycorrhizal fungi in the soil may contribute to co-
mycorrhizal community of the epiphytic orchid Thrixspermum japonicum is strongly existence of orchid species. Fungal Ecol. 20, 155–165. doi: 10.1016/j.
biased toward a single Ceratobasidiaceae fungus, despite a wide range of fungal partners. funeco.2015.12.008
Am. J. Bot. 107, 1654–1662. doi: 10.1002/ajb2.1575
White, T. J., Bruns, T., Lee, S., and Taylor, J. W. (1990). “Amplification and direct 
Rasmussen, H. N., Dixon, K. W., Jersáková, J., and Těšitelová, T. (2015). Germination sequencing of fungal ribosomal RNA genes for phylogenetics” in PCR Protocols: A Guide 
and seedling establishment in orchids: a complex of requirements. Ann. Bot. 116, to Methods and Applications. eds. M. A. Innis, D. H. Gelfand, J. J. Sninsky and T. J. White 
391–402. doi: 10.1093/aob/mcv087 (New York: Academic Press, Inc), 315–322.
Schimper, A. F. W. (1888). Die Epiphytische Vegetation Amerikas. Jena, Germany: Wickham, H. (2016). ggplot2: Elegant Graphics for Data Analysis. New York: Springer-
Gustav Fischer Verlag. Verlag.
Selosse, M.-A., Petrolli, R., Mujica, M. I., Laurent, L., Perez-Lamarque, B., Figura, T., Xu, S., Chen, M., Feng, T., Zhan, L., Zhou, L., and Yu, G. (2021). Use ggbreak to 
et al. (2022). The waiting room hypothesis revisited by orchids: were orchid mycorrhizal effectively utilize plotting space to deal with large datasets and outliers. Front. Genet. 
fungi recruited among root endophytes? Ann. Bot. 129, 259–270. doi: 10.1093/aob/ 12:774846. doi: 10.3389/fgene.2021.774846
mcab134 Zotz, G. (2016). “The role of vascular epiphytes in the ecosystem” in Plants on Plants–
Selosse, M.-A., WEIss, M., Jany, J. L., and Tillier, A. (2002). Communities and the Biology of Vascular Epiphytes. (Springer Cham: Springer International Publishing), 
populations of sebacinoid basidiomycetes associated with the achlorophyllous orchid 229–243. doi: 10.1007/978-3-319-39237-0
Frontiers in Ecology and Evolution 10 frontiersin.org
160
Frontiers in
Ecology and Evolution
Ecological and evolutionary research into our 
natural and anthropogenic world
This multidisciplinary journal covers the spectrum 
of ecological and evolutionary inquiry. It provides 
insights into our natural and anthropogenic world, 
and how it can best be managed. 
Discover the latest 
Research Topics
See more 
Frontiers
Avenue du Tribunal-Fédéral 34
1005 Lausanne, Switzerland
frontiersin.org
Contact us
+41 (0)21 510 17 00 
frontiersin.org/about/contact