Estudio De Los Factores Que Condicionan, a Escala Local, La Presencia De Epidendrum Rhopalostele (Orchidaceae): Implicaciones Para Su Conservación

UNIVERSIDAD POLITÉCNICA DE MADRID DEPARTAMENTO DE BIOTECNOLOGÍA Y BIOLOGÍA VEGETAL ESCUELA TÉCNICA SUPERIOR DE INGENIERÍA AGRONÓMICA, ALIMENTARIA Y DE BIOSISTEMAS

Estudio de los factores que condicionan, a escala local, la presencia de Epidendrum rhopalostele (Orchidaceae): Implicaciones para su conservación

Autor: María Lorena Riofrío Guamán

Directores de tesis: Dra. Elena Torres Dr. José María Iriondo

2015

María Elena Torres Lamas, Profesor Contratado Doctor del Departamento de Biotecnología-Biología Vegetal de la Universidad Politécnica de Madrid y José María Iriondo Alegría, Catedrático de Universidad del Departamento de Biología y Geología de la Universidad Rey Juan Carlos

CERTIFICAN:

Que los trabajos de investigación desarrollados en la memoria de tesis doctoral: “Estudios de los factores que condicionan, a escala local, la presencia de Epidendrum rhopalostele (Orchidaceae): Implicaciones para su conservación”, son aptos para ser presentados por María Lorena Riofrío Guamán ante el Tribunal que en su día se consigne, para aspirar al Grado de Doctor por la Universidad Politécnica de Madrid.

Vo.Bo. Director de Tesis Vo.Bo. Director de Tesis

Dra. María Elena Torres Lamas Dr. José María Iriondo Alegría

Agradecimientos

A mi familia, a ti mi amado esposo Carlos, por estar siempre junto a mi, a mis adorados hijos Carlos Axxel, María José y Santiago Daniel, gracias por compartir su vida y llenar la nuestra, por esperarme, Ustedes son mi motivo. A mis padres, a ti mami por mostrarme el camino, a ti papi por recordarme que la vida sigue. A Ustedes mis hermanos por estar allí. Gracias, infinitas gracias Dra. María Elena Torres y Dr. José María Iriondo, mis directores de tesis. Ustedes marcaron el camino a seguir, fueron mis guías en este proyecto, su tenacidad y gran paciencia, sus enseñanzas constantes en todos los aspectos, su gran visión me permitieron llegar hasta aquí, nuevamente gracias por creer, por brindarnos sus amistad. Gracias Dr. Marcelino de la Cruz, sus códigos fueron imprescindibles, sus enseñanzas, su persistencia nos permitieron avanzar. En el Real Jardín Botánico de Madrid gracias al Dr. Pablo Vargas por el espacio y disposición de ayuda y a Emilio Cano por su paciencia en la enseñanza. A Usted Dr. Juan Pedro Martín gracias por su apoyo y consejos en el laboratorio. Dr. César Pérez usted fue el conductor del proyecto de doctorado su ayuda permitió que estemos hoy aquí, a Usted y al Dr. Luis Miguel Romero nuestro rector (UTPL) gracias por soñar y abrirnos las puertas, vuestra gestión fue decisiva. En la UTPL a todos quienes creyeron y participaron de este crecimiento, el apoyo institucional fue fundamental. Juan Pablo, Paulo y Angelito gracias por su ayuda. Estimado Beto, aprecio mucho su apoyo y agradezco la amistad forjada. A mi compañero de campo, de laboratorio, de análisis, con quien superando todas las adversidades que los bosques de niebla nos brindaron (lluvias, frío y calor intensos, terreno con heterogeneidad de pendientes, escalados difíciles, mosquitos, fiebres, obstrucciones de carreteras, discusiones intensas) y compartiendo los buenos y gratos momentos conseguimos finalizar el trabajo, sin ti no lo habría logrado, muchas gracias Carlos eres el pilar fundamental en esta construcción. El trabajo de esta tesis doctoral fue financiada parcialmente por la SENESCYT- Gobierno de la República del Ecuador.

iii

Índice:

Resumen ...... 1 Abstract ...... 3 INTRODUCCIÓN GENERAL ...... 5 Conservación de orquídeas epífitas en los bosques montanos del Sur del Ecuador ...... 5 Estudios necesarios para sustentar la generación de planes de recuperación de orquídeas epífitas ...... 7 Descripción de la especie de estudio ...... 11 Área de estudio: ...... 13 Objetivos y planteamiento de la tesis ...... 14 Bibliografía ...... 15 1. Spatial structure of Pleurothallis, Masdevallia, Lepanthes and Epidendrum epiphytic orchids in a fragment of montane cloud forest in south Ecuador ...... 21 INTRODUCTION ...... 21 MATERIAL AND METHODS ...... 22 RESULTS AND DISCUSSION ...... 23 Bibliography ...... 26 2. Analysis of the factors that condition the presence of Epidendrum rhopalostele ...... 27 INTRODUCTION ...... 27 MATERIAL AND METHODS ...... 30 Study site ...... 30 Study species ...... 31 Spatial georeferencing of trees and E. rhopalostele ...... 31 Characterization of trees and E. rhopalostele ...... 33 Data analysis ...... 34 RESULTS ...... 37 DISCUSSION ...... 47 Bibliography ...... 51 3. Fine-scale genetic structure of Epidendrum rhopalostele ...... 57 INTRODUCTION ...... 57 MATERIAL AND METHODS ...... 59 Study species ...... 59 Study site and mapping ...... 59 Sampling for genetic analyses and DNA extraction ...... 60 Data analysis ...... 61 RESULTS ...... 63 DISCUSSION ...... 68 Bibliography ...... 70 4. Mycorrhizal preferences of epiphytic orchid Epidendrum rhopalostele (Orchidaceae) ...... 75 INTRODUCTION ...... 75 MATERIAL AND METHODS ...... 77 Study species and site ...... 77 Sampling ...... 77

Light microscopy ...... 78 DNA isolation, PCR, cloning, and sequencing ...... 78 Phylogenetic analyses of Epidendrum rhopalostele mycobionts ...... 78 Spatial and association analyses of mycorrhizal fungi ...... 79 RESULTS ...... 80 DISCUSSION ...... 85 Bibliography ...... 87 Supplementary information ...... 92 CONCLUSIONES ...... 93

Resumen

El análisis de los factores que determinan el establecimiento y supervivencia de orquídeas epífitas, incluyen: a) las condiciones microambientales de los bosques que las mantienen, b) preferencias por las características de los hospederos donde crecen, c) limitación en la dispersión de semillas, d) interacciones planta-planta, y e) asociaciones micorrízicas para la germinación y resultan esenciales para el desarrollo de estrategias para la conservación y manejo de este grupo de plantas. Este trabajo ha evaluado la importancia de estos factores en Epidendrum rhopalostele, orquídea epífita del bosque de niebla montano, a través de los análisis de los patrones espaciales de los árboles que la portan y de la propia orquídea, a escala de población, estudios de asociación y métodos moleculares. Estos últimos han consistido en el uso de marcadores AFLP para el análisis de la estructura genética de la orquídea y en la secuenciación-clonación de la región ITS para la identificación de los hongos micorrízicos asociados. El objetivo de esta tesis es, por tanto, una mejor comprensión de los factores que condicionan la presencia de orquídeas epífitas en los remanentes de bosque de niebla montano y una evaluación de las implicaciones para la conservación y mantenimiento de sus hábitats y la permanencia de sus poblaciones. El estudio fue realizado en un fragmento de bosque de niebla montano de sucesión secundaria situado al este de la Cordillera Real, en los Andes del sur de Ecuador, a 2250 m.s.n.m y caracterizado por una pendiente marcada, temperatura media anual de 20.8°C y precipitación anual de 2193 mm. En este fragmento se mapearon, identificaron y caracterizaron todos los árboles presentes con DBH > 1 cm y todos los individuos de Epidendrum rhopalostele. Así mismo se tomaron muestras de hoja para obtener ADN de todas las orquídeas registradas y muestras de raíces de individuos con flor de E. rhopalostele, uno por cada forófito, para el análisis filogenético de micorrizas. Análisis espaciales de patrones de puntos basados en la K de Ripley y la distancia al vecino más cercano fueron usados para los árboles, forófitos y la población de E. rhopalostele. Se observó que la distribución espacial de árboles y forófitos de E. rhopalostele no es aleatoria, ya que se ajusta a un proceso agregado de Poisson. De ahí se infiere una limitación en la dispersión de las semillas en el fragmento estudiado y en el establecimiento de la orquídea. El patrón de distribución de la población de E. rhopalostele en el fragmento muestra un agrupamiento a pequeña escala sugiriendo una preferencia por micro-sitios para el establecimiento de la orquídea con un kernel de dispersión de las semillas estimado de 0.4 m. Las características preferentes del micro-sitio como tipos de árboles (Clusia alata y árboles muertos), tolerancia a la sombra, corteza rugosa, distribución en los dos primeros metros sugieren una tendencia a distribuirse en el sotobosque. La existencia de una segregación espacial entre adultos y juveniles sugiere una competencia por recursos limitados condicionada por la preferencia de micro-sitio. La estructura genética de la población de E. rhopalostele analizada a través de Structure y PCoA evidencia la presencia de dos grupos genéticos coexistiendo en el fragmento y en los mismos forófitos, posiblemente por eventos de hibridización entre especies de Epidendrum simpátricas. Los resultados del análisis de autocorrelación espacial efectuados en GenAlex confirman una estructura genético-espacial a pequeña

escala que es compatible con un mecanismo de dispersión de semillas a corta distancia ocasionada por gravedad o pequeñas escorrentías, frente a la dispersión a larga distancia promovida por el viento generalmente atribuida a las orquídeas. Para la identificación de los micobiontes se amplificó la región ITS1-5.8S-ITS2, y 47 secuencias fueron usadas para el análisis filogenético basado en neighbor- joining, análisis bayesiano y máximum-likelihood que determinó que Epidendrum rhopalostele establece asociaciones micorrízicas con al menos dos especies diferentes de Tulasnella. Se registraron plantas que estaban asociadas con los dos clados de hongos encontrados, sugiriendo ausencia de limitación en la distribución del hongo. Con relación a las implicaciones para la conservación in situ resultado de este trabajo se recomienda la preservación de todo el fragmento de bosque así como de las interacciones existentes (polinizadores, micorrizas) a fin de conservar la diversidad genética de esta orquídea epífita. Si fuere necesaria una reintroducción se deben contemplar distancias entre los individuos en cada forófito dentro de un rango de 0.4 m. Para promover el reclutamiento y regeneración de E. rhopalostele, se recomienda que los forófitos correspondan preferentemente a árboles muertos o caídos y a especies, como Clusia alata, que posean además corteza rugosa, sean tolerantes a la sombra, y en el área del sotobosque con menor luminosidad. Además es conveniente que las orquídeas en su distribución vertical estén ubicadas en los primeros metros. En conclusión, la limitación en la dispersión, las características del micro-sitio, las interacciones intraespecíficas y con especies congenéricas simpátricas y las preferencias micorrízicas condicionan la presencia de esta orquídea epífita en este tipo de bosque.

Abstract

The analysis of factors that determine the establishment and survival of epiphytic depends on factors such as a) microenvironmental conditions of forest, b) preference for host characteristics where orchids grow, c) seed dispersal limitation, d) plant-plant interaction, e) priority mycorrhizal associations for germination, are essential for the development of strategies for management and conservation. This work evaluated the importance of these factors in Epidendrum rhopalostele, an epiphytic orchid of montane cloud forest through the analysis of spatial patterns of host trees and the orchid, in a more specific scale, with association studies and molecular methods, including AFLPs for orchid population genetic structure and the sequencing of the ITS region for associated mycorrhizal fungi. The aim of this thesis is to understand the factors that condition the presence of epiphytic orchids in the remnants of montane cloud forest and to assess the implications for the conservation and preservation of their habitats and the persistence of the orchid populations. The study was carried out in a fragment of montane cloud forest of secondary succession on the eastern slope of Cordillera Real in the Andes of southern Ecuador, located at 2250 m a.s.l. characterized by a steep slope, mean annual temperature of 20.8°C and annual precipitation of 2193 mm. All trees with DBH > 1 cm were mapped, characterized and identified. All E. rhopalostele individuals present were counted, marked, characterized and mapped. Leaf samples of all orchid individuals were collected for DNA analysis. Root samples of flowering E. rhopalostele individuals were collected for phylogenetic analysis of mycorrhizae, one per phorophyte. Spatial point pattern analysis based on Ripley`s K function and nearest neighbor function was used for trees, phorophytes and orchid population. We observed that spatial distribution of trees and phorophytes is not random, as it adjusts to a Poisson cluster process. This suggests a limitation for seed dispersal in the study fragment that is affecting orchid establishment. Furthermore, the small-scale spatial pattern of E. rhopalostele evidences a clustering that suggests a microsite preference for orchid establishment with a dispersal kernel of 0.4 m. Microsite features such as types of trees (dead trees or Clusia alata), shade tolerance trees, rough bark, distribution in the first meters suggest a tendency to prefer the understory for their establishment. Regarding plant-plant interaction a spatial segregation between adults and juveniles was present suggesting competition for limited resources conditioned for a microsite preference. Analysis of genetic structure of E. rhopalostele population through Structure and PCoA shows two genetic groups coexisting in this fragment and in the same phorophyte, possibly as a result of hybridization between sympatric species of Epidendrum. Our results of spatial autocorrelation analysis develop in GenAlex confirm a small-scale spatial-genetic structure within the genetic groups that is compatible with a short-distance dispersal mechanism caused by gravity or water run-off, instead of the long-distance seed dispersal promoted by wind generally attributed to orchids. For mycobionts identification ITS1-5.8S-ITS2 rDNA region was amplified. Phylogenetic analysis was performed with neighbor- joining, Bayesian likelihood and maximum-likelihood for 47 sequences yielded two Tulasnella clades. This orchid establishes mycorrhizal associations with at

least two different Tulasnella species. In some cases both fungi clades were present in same root, suggesting no limitation in fungal distribution. Concerning the implications for in situ conservation resulting from this work, the preservation of all forest fragment and their interactions (pollinators, mycorrhiza) is recommended to conserve the genetic diversity of this species. If a reintroduction were necessary, distances between individuals in each phorophyte within a range of 0.4 m, are recommended. To promote recruitment and regeneration of E. rhopalostele it is recommended that phorophytes correspond to dead or fallen trees or species, such as Clusia alata. Trees that have rough bark and are shade tolerant are also recommended. Furthermore, regarding vertical distribution, it is also convenient that orchids are located in the first meter (in understory, area with less light). In conclusion, limitation on seed dispersal, microsite characteristics, plant-plant interactions or interaction with cogeneric sympatric species and mycorrhizal preferences conditioned the presence of this epiphytic orchid in this fragment forest.

INTRODUCCIÓN GENERAL

Conservación de orquídeas epífitas en los bosques montanos del Sur del Ecuador

Diversidad de orquídeas en Ecuador Orchidaceae es una de las familias más diversa de plantas con flor, tal vez la más grande de todas ellas (Dressler, 1981). Se han reconocido aproximadamente 800 géneros en esta familia, con una estimación de 35,000 especies y en Ecuador se han descrito 228 géneros y 3920 especies que representan el 20% de la flora nativa ecuatoriana y el 45% de la flora endémica amenazada (Dodson et al., 2004; Endara, 2011). Las orquídeas pueden ser plantas perennes, hierbas, similares a arbustos o enredaderas, excepcionalmente variables en hábitat encontrándose desde los 0 m.s.n.m. hasta los 3000 m.s.n.m, en bosques secos, lluviosos o de niebla, páramos y otros. Son igualmente variables en cuanto al medio sobre el que se desarrollan: terrestre, litofítico, epifítico, raramente semiacuático o subterráneo (Garay, 1978). La mayoría de géneros de orquídeas se distribuyen a través del neotrópico en la región de los Andes al noroeste de Sudamérica, y un 70% de todas ellas son epífitas (esto es, germinan y se desarrollan sobre los árboles y lianas, denominándose a la planta hospedera: forófito), constituyendo el grupo más diverso y mejor representado de epífitos vasculares (Benzing, 1987). Las epífitas son un componente característico de bosques lluviosos tropicales y bosques de niebla, tanto en diversidad de especies como en biomasa (Gravendeel et al., 2004), probablemente por la diferenciación de microambientes típicos de esta región montañosa (Gentry & Dodson, 1987). En Ecuador más del 60% de las orquídeas son epífitas (Dodson, 1994-2003; Dodson et al., 2004). Su alta diversidad se incrementa en algunos hábitats dentro de los bosques tropicales de América por condiciones como humedad, elevación media y suelos ricos que permiten una gran diferenciación en micro-sitios (Gentry & Dodson, 1987), están adaptadas a condiciones variadas de crecimiento, pertenecen también a los grupos de plantas altamente evolucionadas y con mayor tolerancia al estrés (Benzing, 2012), utilizan adaptaciones particulares como sus raíces aéreas que permiten la fijación al sustrato además de la absorción de nutrientes y agua (Gentry & Dodson, 1987).

Estructura de los bosques montanos del Sur del Ecuador Las regiones con gran diversidad de condiciones ambientales son consideradas centros de alta biodiversidad. Este es el caso de las zonas montañosas que se encuentran al Sur de los Andes ecuatorianos en la depresión de los Andes, con variedad de pendientes topográficas y gradientes climáticos (Robbert et al., 2008; Homeier et al., 2008; Richter et al., 2009). Las floras montanas tienen niveles más altos de endemismo que las zonas tropicales bajas, el gran endemismo combinado con la devastación antropogénica hace que la preservación de los remanentes boscosos sea una prioridad de conservación mundial (Gentry, 1993). Los bosques

5 INTRODUCCIÓN GENERAL

de niebla están entre los tipos de bosques tropicales más amenazados y menos estudiados. La mayoría son remanentes de bosques fragmentados, con una gran variación natural en especies, estructura y composición debido a la heterogeneidad de hábitats (Gentry, 1993; Williams-Linera, 2002). En Ecuador los bosques montanos son únicos, con ecosistemas con alta biodiversidad, elevado número de especies de árboles, de epífitos, insectos entre otras especies (Valencia et al., 2000; Dodson et al., 2004). Las epífitas por tanto son un componente característico de los bosques lluviosos, son elementos claves en el funcionamiento de los ecosistemas, incrementan la diversidad, e intervienen en la interacción con otras plantas y animales y en el flujo de agua y nutrientes (Gentry & Dodson, 1987).

Situación de las orquídeas en Ecuador El 20% de la diversidad de plantas en Ecuador corresponde a orquídeas y el 40% de las orquídeas son endémicas, a pesar de esta alta diversidad (Dodson, 1994- 2003; Jørgensen & León-Yánez, 2001; Dodson et al., 2004; Endara, 2011), este grupo de plantas enfrenta amenazas a su supervivencia, pues se ve afectada por la deforestación al ser completamente dependiente de las condiciones ambientales de los bosques donde se encuentran sus hospederos. La alta tasa de deforestación que se ha producido en este país desde los años 1900 – 1950 en los bosques costeros de tierras bajas y 1970 en los bosques amazónicos, intensificándose en los últimos años en el sur del Ecuador, consecuencia de los aclareos para extensión de la frontera agrícola, minería y la implementación de nuevas infraestructuras (Mosandl et al., 2008), está ocasionando la pérdida y fragmentación del hábitat natural de numerosas especies (Santos & Tellería, 2006), y sigue siendo la principal amenaza para su persistencia (Richter et al., 2009). El impacto humano directo es pues la principal amenaza en los bosques de la Depresión de los Andes, así como factores indirectos de acción antropogénica, como el cambio global, que afecta principalmente a los bosques tropicales, también tienen su incidencia en la conservación de la diversidad (Achard et al., 2002; Richter et al., 2009). A estas amenazas hay que sumar la extracción de orquídeas destinada al comercio ilegal, que en algunos casos ha llevado a la extinción de las poblaciones naturales (Endara, 2011). Las autoridades ambientales en Ecuador han desarrollado medidas de control ante esta realidad, con la creación de leyes y políticas para la protección de la biodiversidad, como planes de reforestación, uso comercial sostenible, incentivos a comunidades, educación ambiental, políticas en investigaciones, capacitaciones, y desarrollo tecnológico. Por tanto, la generación de información que lleve al conocimiento de la ecología, biología y distribución de las especies que se pretende conservar, y que permitan el desarrollo de planes o estrategias de conservación centrados en la situación de las mismas, son necesarias y urgentes para contribuir a las políticas planteadas.

Conservación de orquídeas epífitas Una zona importante dentro de los ecosistemas del Ecuador son los bosques montañosos relictos del sur de los Andes, dentro de los cuales los bosques de niebla montanos son reconocidos como uno de los más importantes centros de

6 INTRODUCCIÓN GENERAL

diversidad florística del mundo y como un hotspot en prioridad de conservación (Richter & Moreira-Muñoz, 2005). Estos bosques presentan varias amenazas, una de las principales es la pérdida de hábitat por deforestación, constituyéndose en bosques de sucesión secundaria dada por el cambio de uso del suelo por acción antropogénica. Todas estas características priorizan la conservación de las especies y poblaciones que allí se encuentran. Las orquídeas epífitas se desarrollan en estos tipos de bosques y dependen estrechamente de ellos para su persistencia. Una de las implicaciones para la conservación de las orquídeas epífitas es el mantenimiento de sus hábitats. Esto solo es posible si se conocen los aspectos o factores que condicionan su presencia en los remanentes de bosque de niebla montano. De cara a la conservación in situ de orquídeas epífitas surgen varias inquietudes, el poder responder a cuestiones como ¿dónde se colocarían las orquídeas que han caído de sus forófitos? ¿deben ser colocadas en los mismos forófitos?¿podrían usarse para reintroducciones?¿árboles similares pueden servir como hospederos?¿pueden ser llevadas estas orquídeas a otros sitios similares?¿para estas orquídeas la especie de forófito es indiferente para su establecimiento? Por tanto, para poder responder a estas inquietudes y plantear estrategias de conservación viables para las orquídeas epífitas es necesario conocer las preferencias en microhábitat y la heterogeneidad ambiental como posibles limitantes para la dispersión efectiva de las orquídeas, las relaciones planta- planta, todos factores que podrían condicionar la presencia de una orquídea epífita.

Estudios necesarios para sustentar la generación de planes de recuperación de orquídeas epífitas

Dispersión de semillas y colonización de orquídeas epífitas. La dispersión es un proceso clave en la dinámica y evolución de las poblaciones de plantas. Los rangos de dispersión posibilitan la colonización de nuevos sitos, y determinan la estructura metapoblacional (Ouborg et al., 1999). El establecimiento exitoso de las plantas que está dado por el patrón de dispersión de las semillas y que determinará su distribución espacial, dependerá tanto de la producción de las semillas y su dispersión, como de la disponibilidad de recursos en el micrositio e influencia además la estructura genética de las poblaciones naturales de plantas (Williams, 1994; Jacquemyn et al., 2007b; Winkler et al., 2009). Por tanto, la dispersión de semillas y el reclutamiento de nuevos individuos son factores importantes para la dinámica y la estructura genética y espacial de las poblaciones (Jacquemyn et al., 2007b). Los patrones de dispersión de semillas estarán determinados por los patrones espaciales de los adultos y la lluvia de semillas, mientras que la distribución de las plántulas va a determinarla, además, la disponibilidad de un micrositio adecuado (Nathan & Muller-Landau, 2000; Jersáková & Malinová, 2007). Las poblaciones de orquídeas epífitas presentan una distribución en parches por su hábito epífito (Tremblay et al., 2006; Winkler et al., 2009), y están condicionadas además por la disponibilidad y distribución del forófito (Flores-

7 INTRODUCCIÓN GENERAL

Palacios & Ortiz-Pulido, 2005), la presencia de hongos micorrízicos adecuados, la heterogeneidad de las condiciones ambientales (McCormick & Jacquemyn, 2014), o las interacciones planta-planta (Wiegand et al., 2007a; Jacquemyn et al., 2007b; Jacquemyn et al., 2009; Raventós et al., 2011; Jacquemyn et al., 2014) que puedan suscitarse. La mayor parte de estos procesos tienen un componente espacial que puede explicar la distribución y coexistencia de las especies siendo la respuesta a muchos procesos ecológicos presentes en varias escalas (Wiegand et al., 2007a). Las orquídeas epífitas, con semillas diminutas y embrión no diferenciado con mínimas reservas, son dependientes del hongo micorrízico para su germinación (Rasmussen & Rasmussen, 2007; Dearnaley, 2007; Jersáková & Malinová, 2007). Sumado a la producción y dispersión de semillas (McCormick & Jacquemyn, 2014), los factores ambientales condicionan mucho su establecimiento y la disponibilidad de un sitio adecuado (Riofrío et al., 2013). Además, eventos como deforestación por expansión humana, conversión de tierra para la agricultura, tráfico de especies, han ocasionado procesos de fragmentación en las poblaciones naturales de plantas y han colocado a las orquídeas entre los grupos con una alta categoría de amenaza (Endara, 2011).

Importancia de los estudios genéticos Al tratar de conservar in situ a las orquídeas epífitas, nos enfrentamos a la escasez de información acerca de los procesos de dispersión de semillas y los factores que condicionan el establecimiento de estas orquídeas en los árboles hospederos. Dado el tamaño diminuto de las semillas de las orquídeas se presume que su mecanismo de dispersión sea esencialmente por el viento y por tanto pueda alcanzar grandes distancias (Arditti & Ghani, 2000). Sin embargo, para algunas especies de orquídeas terrestres se ha demostrado que la dispersión de las semillas no va más allá de los 10 metros y es limitada (Machon et al., 2003; Chung et al., 2007; Jacquemyn et al., 2007a). También se ha propuesto que las semillas puedan transportarse por el agua a través de la lluvia y solo alcanzar distancias cortas, limitando la distribución de los nuevos individuos, como propone Machon et al. (2003), en la orquídea terrestre Spiranthes spiralis, cuyas semillas se dispersaron 15 cm alrededor de la planta madre. El proceso de dispersión de semillas es importante pues determina la dinámica de las poblaciones así como su estructura genética y la distribución espacial de los alelos y genotipos dentro de las poblaciones (Jacquemyn et al., 2007b). Los estudios genéticos basados en marcadores moleculares de ADN han demostrado ser una alternativa eficaz para responder cuestiones ecológicas y asistir al momento de tomar decisiones de manejo (Allendorf & Luikart, 2007). Dentro de estos marcadores encontramos los AFLPs (amplified fragment length polymorphism) que permiten análisis de la variabilidad intra e inter poblacional, sin previo conocimiento de las secuencias del ADN, proveen una gran cantidad de datos con alta reproducibilidad de los resultados y son altamente polimórficos (Smith et al., 2004; Vuylsteke et al., 2007). Así mismo el análisis de la diversidad genética dentro y entre poblaciones permite estimar el flujo génico en numerosas especies e inferir los mecanismos de polinización y dispersión de semillas, y el estudio del patrón genético espacial contribuye a conocer los procesos

8 INTRODUCCIÓN GENERAL

subyacentes que los generan (Escudero et al., 2002; Wallace, 2006; Jacquemyn et al., 2007b). Esta información ha servido para apoyar medidas de conservación in situ, tales como la reintroducción o el refuerzo de poblaciones existentes, de numerosas especies. En lo que respecta a estudios genéticos en orquídeas usando marcadores moleculares basados en ADN, se ha dado énfasis a estudios en orquídeas terrestres de Norte América, Europa, Asia y Australia encontrando variaciones en la diversidad genética desde muy poca variación hasta muy alta, así mismo fluctuaciones entre elevada diversidad genética intrapoblacional y bajo flujo génico versus baja diferenciación intrapoblacional y elevado flujo génico (Machon et al., 2003; Chung et al., 2004; Smith et al., 2004; Forrest et al., 2004; Jacquemyn et al., 2006; Wallace, 2006; Jacquemyn et al., 2007b). Estas variaciones podrían deberse a la distancia de las poblaciones estudiadas y al tipo de muestreo efectuado (Chung & Chung, 2007). En algunos casos se han encontrado también patrones significativos de SGS (Machon et al., 2003) que podrían ser explicados por una limitada dispersión de semillas (Jersáková & Malinová, 2007) asumiendo los rangos de dispersión de 4 a 7 metros como en el caso de Orchis purpurea (Jacquemyn et al., 2007b). Los estudios que se han efectuado para orquídeas epífitas tropicales son escasos, a pesar de su importancia ecológica y su gran diversidad, situación debida posiblemente por las dificultades de accesibilidad a los árboles hospederos, o a la estructura y situación de los bosques donde se alojan (Riofrío et al., 2007). En algunos casos, los marcadores isoenzimáticos han sido los más utilizados encontrando bajos niveles de diferenciación genética (Ackerman & Ward, 1999; Alcántara et al., 2006; Vargas et al., 2006; Ávila-Díaz & Oyama, 2007), mientras que para otros estudios se evidencia una alta diversidad genética y una baja estructura de población que pudiera deberse a la adaptabilidad que han logrado las plantas epífitas para poder responder a los constantes cambios en su habitat (Ávila-Díaz & Oyama, 2007). Los factores que condicionan la colonización de nuevos árboles y el establecimiento de estas orquídeas en los hospederos pueden estar ligados a la disponibilidad de recursos de micrositios donde posiblemente se desarrollen (Otero et al., 2007a) y directamente relacionados con las condiciones de humedad. También pueden depender de la disposición de hongos micorrízicos, como lo plantea Otero et al. (2004), en dos especies de orquídeas epífitas, donde descarta las condiciones del hábitat, y propone la especificidad micorrízica considerando que la distribución de orquídeas epífitas está directamente relacionada con el carácter generalista o específico en su relación con los hongos micorrízicos, y que su distribución va a estar limitada por esta relación y la distribución de los hongos micorrízicos. Las micorrizas de orquídeas condicionan la presencia de orquídeas desde su germinación, desarrollo como plántulas y en muchos casos a través de su vida. Además de un sustrato inadecuado y condiciones físicas cambiantes, las semillas de orquídeas se enfrentan a otro problema que es la localización del micobionte compatible (Rasmussen, 2002). La especificidad de las orquídeas por los hongos micorrízicos ha sido un tema controvertido desde mucho tiempo atrás en parte por

9 INTRODUCCIÓN GENERAL

el problema de identificación de los hongos (Warcup, 1981; Rasmussen, 2002; Dearnaley, 2007; Porras-Alfaro & Bayman, 2007). Los procesos de identificación morfológica de micorrizas de orquídeas son complejos. La taxonomía fúngica está basada en la morfología de sus estructuras sexuales (estados telomorfos), pero en hongos del grupo Rhizoctonia raramente se consigue la fructificación en cultivo (Zelmer et al., 1996; Suárez et al., 2006; Waterman & Bidartondo, 2008; Taylor & McCormick, 2008). No obstante, el desarrollo de nuevas técnicas en biología molecular ha permitido la identificación de los hongos micorrízicos tanto de orquídeas terrestres como epífitas, aclarando la taxonomía de este grupo (Otero et al., 2002; Otero et al., 2004; Taylor et al., 2004; Otero et al., 2005; Shefferson et al., 2005; Suárez et al., 2006; Shefferson et al., 2007; Otero et al., 2007b; Bidartondo & Read, 2008; Kottke et al., 2008; Shefferson et al., 2008; Suárez et al., 2008; Smith & Read, 2008; Taylor & McCormick, 2008; Waterman & Bidartondo, 2008). El locus más utilizado para la identificación de especies e inferencias filogenéticas en las investigaciones micológicas basadas en secuencias es la región del espaciador interno transcrito (ITS) de la unidad repetitiva ribosomal nuclear (Nilsson et al., 2008). Este segmento de aproximadamente 550 pares de bases (bp) combina las ventajas de resolución a varias escalas: ITS1, evolución rápida, 5.8S muy conservada e ITS2 moderadamente rápida a rápida (Nilsson et al., 2008). Otro aspecto a considerar es que es un segmento multi-copias, de tal modo que se puede amplificar aún si las cantidades de ADN son moderadas, que es el caso de la mayoría de muestras de campo (Ryberg et al., 2008). Los métodos moleculares basados en amplificaciones de PCR hongo-específicas de ITS han revolucionado la caracterización de ecto micorrizas, micorrizas ericoideas, arbusculares y de orquídeas (Taylor & McCormick, 2008). Con los estudios moleculares, se ha podido esclarecer las hipótesis sobre especificidad de las interacciones orquídea-micorriza (OM), determinándose que si una especie de orquídea interactúa únicamente con un taxón fúngico contenido dentro de un clado filogenético único y restrictivo existe especificidad (McCormick et al., 2004; Otero et al., 2007b; McCormick et al., 2012). También debe considerarse que el grado de especificidad en las asociaciones OM va a depender de las coincidencias ecológicas de la orquídea y el hongo (especificidad ecológica), de la compatibilidad fisiológica entre el hongo y orquídea y de la relativa funcionalidad de cada interacción (fitness), ocasionando que la especificidad ecológica esté marcada por el rango de disponibilidad del hongo asociado (McCormick et al., 2004; Otero et al., 2007b). La presencia de las interacciones micorrízicas de orquídeas (OM) es muy importante para la ecología y conservación, pudiéndose esperar que la distribución de las orquídeas sea generalista en sus preferencias por los hongos micorrízicos o específica asociada a una amplia distribución del hongo, pues una reducida especificidad OM podría ser una razón para la rareza y vulnerabilidad de las especies (Otero et al., 2007b). La distribución y diversidad de las especies de orquídeas podría depender de la especificidad de las interacciones orquídea- micorriza (OM) y de la distribución del hongo (Otero et al., 2004; Otero et al., 2007a). Al estar las poblaciones de orquídeas fuertemente amenazadas por la destrucción de sus hábitats, resulta necesario entender el papel que juegan los

10 INTRODUCCIÓN GENERAL

hongos micorrízicos en los procesos de germinación y sustento de estas plantas en todos sus estados de desarrollo. A pesar del enorme interés que han suscitado estas plantas entre científicos y aficionados, el conocimiento que se tiene acerca de la dinámica de colonización en bosques regenerados y las interacciones orquídea-hospedero es prácticamente inexistente. Los estudios de especificidad por el hospedero tampoco arrojan mucha luz a esta cuestión, dado que no tienen en cuenta los factores físico- químicos que determinan la presencia y abundancia de orquídeas (Migenis & Ackerman, 1993). Sin embargo Otero et al. (2007a) encuentra alta preferencia en Psychilis monensis (orquídea epífita) por un grupo de forófitos, que poseen en su mayoría corteza rugosa, y concluye además que las características del micrositio donde se desarrollan y la presencia de micorrizas sería determinante para la distribución de este tipo de orquídeas.

Descripción de la especie de estudio Epidendrum se encuentra entre los géneros más numerosos de orquídeas con 431 especies nativas para Ecuador de las cuales 205 son endémicas (Endara, 2011) y están entre las aproximadamente 1500 especies reportadas; este género es nativo para las regiones tropicales y subtropicales de América, desde Carolina del Sur hasta el Norte de Argentina (Dodson, 1994-2003; Hágsater et al., 2001; Jørgensen & León-Yánez, 2001; Hágsater & Soto-Arenas, 2005). Se trata de especies de hábitos epífitos en su mayoría, pudiendo encontrarse también unas pocas especies terrestres o litófitas. La mayoría se encuentran en los Andes, en altitudes comprendidas entre los 1000 y 3000 m.s.n.m. Su hábitat varía desde las selvas húmedas hasta los bosques secos, desde los terrenos soleados y despejados hasta los bosques de niebla (Hágsater et al., 1999). Epidendrum rhopalostele, descrita por Hágsater & Dodson (2001) es una orquídea fotosintética, epífita, nativa de Ecuador, distribuida en los bosques de niebla montanos a través de la Cordillera de los Andes desde el centro de Ecuador hasta la zona limítrofe con Perú (Hágsater et al., 2001), si bien también se ha reportado en Bolivia (Jiménez-Pérez, 2011). Pertenece a la subtribu Laeliinae, grupo Alpicolum, subgrupo Dialychilum. Esta orquídea presenta un crecimiento cespitoso, su tamaño puede variar de 20 a 40 cm de alto; con raíces filiformes. Con una sola inflorescencia apical, racimosa, con 10 a 30 flores verde claro. El labio es libre, se inserta en la base de la columna, ligeramente lanceolado y acuminado, con 3 venaciones, su columna es delgada, ensanchada hacia el ápice y la cavidad estigmática ocupa el tercio apical, lo que la diferencia de E. dialychilum Hágsater & Dodson, cuya cavidad estigmática ocupa casi toda la columna (Fig. 1). Su fruto o cápsula es ovoidea de aproximadamente 1 cm de largo x 0.8 cm de ancho, sus semillas son diminutas y ligeras de 250 ± 350 µm de largo por 40 ± 55 µm de ancho (Fig. 2). Los periodos de floración observados en la población comprenden los meses de enero a marzo y junio–agosto, mientras que las cápsulas alcanzan la madurez aproximadamente en un mes (observaciones personales).

11 INTRODUCCIÓN GENERAL

Fig. 1. Lámina de Epidendrum rhopalostele indicando a) La estructura completa de la planta y su hábito epífito, b) La flor desde un ángulo frontal, c) La forma de los sépalos y pétalos, d) La forma de la columna y el labio, e) y f) Una imagen frontal y lateral de la columna y el labio, g) La forma de los polinios y la capucha de la antera que cubren los polinios. (Elaboración: Naranjo C., Riofrío M. y Mendoza A.)

12 INTRODUCCIÓN GENERAL

Fig. 2. Microfotografía de las semillas de Epidendrum rhopalostele (Elaboración: Naranjo C. y Riofrío M.)

Área de estudio: El sitio de estudio está localizado en el km. 24 de la vía Loja-Zamora al sur del Ecuador, coordenadas 9558545 N, 17710586 E y 2280 m.s.n.m. (79°6ʹ4ʹʹW, 3°59ʹ17ʹʹS), en los límites del Parque Nacional Podocarpus, en la vertiente este de la Cordillera Real de los Andes. El fragmento está limitado por la carretera Loja- Zamora al norte, mientras que al sur se encuentra la quebrada San Francisco y en las orientaciones este y oeste se localizan pastizales. Es un fragmento de bosque de sucesión secundaria de aproximadamente 1.3 hectáreas; la edad estimada del fragmento es de 35 años. El tipo de bosque según Sierra (1999) es un bosque de niebla montano alto y corresponde según su tipo de vegetación a bosque montano superior siempre verde (Beck et al., 2008b). Con árboles con una altura promedio de 5 a 8 m y lianas con longitudes que superan los 10 m. La flora predominante del lugar son Orchidaceae, Melastomataceae, con el género Miconia, Ericaceae con los géneros Psamisia, Cavendishia y Gaultheria, Clusaceae con el género Clusia, Tovomita, Rubiaceae con los géneros, Psycotria, Lauraceae, con Nectandra, Persea, Ocotea. El fragmento tiene una pendiente del 51% y una altitud comprendida entre los 2200 y 2250 m.s.n.m. con una precipitación media anual de 2700 mm, y una temperatura media anual de 20.8 oC (4.7 – 25.5oC).

13 INTRODUCCIÓN GENERAL

Objetivos y planteamiento de la tesis El principal objetivo de este trabajo consiste en inferir los factores que condicionan a escala local la presencia de la orquídea epífita Epidendrum rhopalostele, los mecanismos de dispersión y la dinámica de colonización, a partir del estudio de la estructura genética espacial y las relaciones micorrízicas. La consecución de este objetivo permitirá valorar las consecuencias demográficas y genéticas de la fragmentación de hábitat en la supervivencia de esta especie, y la generación de información base para el manejo y gestión de poblaciones naturales de ésta y otras orquídeas epífitas. Para alcanzar el fin propuesto, se plantearon los siguientes objetivos específicos: 1. Descripción del patrón de distribución espacial de los individuos de E. rhopalostele. Para cumplir con este objetivo se georeferenciaron todos los individuos presentes en el fragmento del bosque en las poblaciones estudiadas. Los resultados nos permiten inferir el patrón de dispersión que presenta esta especie. 2. Caracterización descriptiva y asociaciones de la presencia de E. rhopalostele con determinados factores ambientales. Para cumplir con este objetivo se tomaron datos de las características físicas de los árboles hospederos y no hospederos de esta orquídea. Los resultados nos permiten describir el microhábitat de la especie. 3. Estimación de la diversidad genética y descripción de la estructura genética espacial. Para alcanzar este objetivo, se analizaron mediante AFLPs todos los individuos presentes localizados en cada forófito, desde su base hasta la copa. Los resultados permiten determinar la estructuración genética de la población e inferir los mecanismos de dispersión que prevalecen en la especie. 4. Determinar especificidad en la relación orquídea-hongo. Para alcanzar este objetivo se analizaron las raíces de las orquídeas con el fin de determinar la infección del hongo micorrízico y se clasificó molecularmente el hongo encontrado así como también se evaluó su distribución espacial. Esto nos permite conocer la preferencia de la especie por un grupo de hongos micorrízicos que pudiera limitar su distribución. A partir de los resultados obtenidos se proponen medidas para la conservación de esta especie de hábito epífito localizada en uno de los ecosistemas de alta vulnerabilidad en los bosques montanos. El documento de tesis está estructurado por capítulos. Cada capítulo ha sido redactado en inglés para facilitar su publicación en revistas científicas (la información del capítulo 1 y 4 ya ha sido publicada). Consta además de una introducción general y un apartado final correspondiente a conclusiones principales.

14 INTRODUCCIÓN GENERAL

El capítulo 1 "Spatial structure of Pleurothallis, Masdevallia, Lepanthes and Epidendrum epiphytic orchids in a fragment of montane cloud forest in South Ecuador" analiza la distribución espacial de cuatro géneros de orquídeas epífitas presentes en un fragmento de bosque de niebla secundario al sur del Ecuador, con el fin de inferir los patrones de dispersión de semillas y colonización. Además evalúa los efectos de la identidad y tamaño del forófito en el establecimiento de las orquídeas. Corresponde a un estudio previo a centrar la investigación en los factores que condicionan a escala local la presencia de Epidendrum rhopalostele. El capítulo 2: “Analysis of the factors that condition the presence of E. rhopalostele” trata de resolver la disposición de los individuos de E. rhopalostele en el fragmento de bosque de niebla secundario, y el porqué de su ubicación, a través de una caracterización descriptiva de la ubicación de E. rhopalostele. Se realizan análisis espaciales de patrones de puntos para entender donde se encuentra y porqué, tomando en cuenta factores como una posible limitación a la dispersión de las semillas y la heterogeneidad de las condiciones ambientales donde se desarrolla. Toda esta información tiene el propósito de obtener conclusiones útiles para la gestión y conservación de esta especie. El capítulo 3: “Fine-scale genetic structure of Epidendrum rhopalostele” trata de inferir los mecanismos de dispersión predominantes en la especie (corta vs. larga distancia) a través de la evaluación de la estructura genético-espacial a pequeña escala de E. rhopalostele en el fragmento de bosque de niebla secundario usando AFLPs. El capítulo 4: “Mycorrhizal preference and fine spatial structure of the epiphytic orchid Epidendrum rhopalostele” expone las asociaciones micorrízicas de E. rhopalostele, explicando cuáles son los micobiontes asociados con esta orquídea a través de la identificación molecular, las preferencias micorrízicas por uno o varios hongos simultáneamente, y la compartición de los mismos hongos entre individuos cercanos. Un mejor conocimiento de la interacción hongo-orquídea permite conocer el grado de especificidad de la misma y comprobar si es uno de los factores determinantes en la distribución de la orquídea. La identificación de los hongos micorrízicos asociados a nivel poblacional es esencial para la conservación de orquídeas.

Bibliografía

Achard F, Eva H, Stibig H, Mayaux P, Gallego J, Richards T, Malingreau J. 2002. Determination of deforestation rates of the world´s humid tropical forest. Science 297: 999-1002. Ackerman JD, Ward S. 1999. Genetic variation in a widespread epiphytic orchid: where is the evolutionary potencial? Systematic Botany 24: 282-291. Alcántara S, Semir J, Nisaka V. 2006. Low genetic structure in an epiphytic Orchidaceae (Oncidium hookeri) in the atlantic rainforest of south-eastern Brazil. Annals of Botany 98: 6, 1207-1213.

15 Bibliografía

Allendorf FW, Luikart G. 2007. Conservation and the genetics of populations. 1st edn. Blacwell Publishing, Australia. Arditti J, Ghani AK. 2000. Numerical and physical properties of orchid seeds and their biological implications. New Phytologist 145: 3, 367-421. Ávila-Díaz I, Oyama K. 2007. Conservation genetics of an endemic and endangered epiphytic Laelia speciosa (Orchidaceae). American Journal of Botany 94: 2, 184. Beck E, Mosandl R, Richter M, Kottke I. 2008. The Investigated Gradients. In: Beck E, Bendix J, Kottke I, Makeschin F, Mosandl R (eds.), Gradients in a Tropical Mountain Ecosystem of Ecuador. Springer, Verlang Berlin, pp. 55-62. Benzing D. 2012. Air Plants: Epiphytes and Aerial Gardens, 1st edn. Cornell University Press, Unites States of America. Benzing D. 1987. Vascular epiphytism: taxonomy participation and adaptative diversity. Annals of the Missouri Botanical Garden 74: 183-204. Bidartondo MI, Read DJ. 2008. Fungal specificity bottlenecks during orchid germination and development. Molecular Ecology 17: 3707-3716. Chung M, Nason JD, Chung M. 2007. Effects of population succession on demographic and genetic processes: predictions and tests in the daylily Hemerocallis thunbergii (Liliaceae). Molecular Ecology 16: 13, 2816-2829. Chung MY, Chung MY. 2007. Extremely low levels of genetic diversity in the terrestrial orchid Epipactis thumbergii (Orchidaceae) in South Korea: implications for conservation. Botanical Journal of the Linnean Society 155: 161-169. Chung MY, Nason JD, Chung MG. 2004. Spatial genetic structure in populations of the terrestrial orchid Cephalanthera longibracteata (Orchidaceae). American Journal of Botany 91: 1, 52. Dearnaley JD. 2007. Further advances in orchid mycorrhizal research. Mycorrhiza 17: 6, 475-486. Dodson C, Luer C, Romero G. 2004. Orchids of Ecuador, CD-Room version edn. Not published, . Dodson C. 1994-2003. Native Ecuadorian Orchids, 4th edn. Colina, Quito. Dressler RL. 1981. The orchids. Natural history and Classification, Second edn. Harvard University Press., Harvard. Endara L. 2011. Orchidaceae. In: León-Yañez S, Valencia R, Pitman N, Endara L, Ulloa C, Navarrete H (eds.), Libro rojo de plantas endémicas del Ecuador. Publicaciones del Herbario QCA. Pontificia Universidad Católica del Ecuador, Quito, pp. 441- 702. Escudero A, Iriondo J, Torres ME. 2002. Spatial analysis of genetic diversity as a tool for plant conservation. Biological Conservation 113: 351-365. Flores-Palacios A, Ortiz-Pulido R. 2005. Epiphyte orchid establishment on termite carton trails. Biotropica 37: 3, 457. Forrest AD, Hollingsworth ML, Hollingsworth PM, Sydes C, Bateman RM. 2004. Population genetic structure in European populations of Spiranthes romanzoffiana set in the context of other genetic studies on orchids. Heredity 92: 3, 218-227. Garay LA. 1978. Orchidaceae. Cypripedioideae, Orchidoideae, Neottioideae. In: Harling G, Sparre B (eds.), Flora of Ecuador. 9. University of Göteborg, Stockholm, Sweden, pp. 1-305. Gentry AH. 1993. Patterns of diversity and floristic composition in Neotropical montane forests. In: Churchill ST, Balslev H, Forero E, Luteyn JL (eds.), Biodiversity and Conservation of Neotropical Montane Forests. Proceedings of the Neotropical Montane Forest Biodiversity and Conservation Symposium, The New York, Botanical Garden, 2126 June . New York Botanical Garden, New York, pp. 103- 126. Gentry AH, Dodson C. 1987. Diversity and biogeography of neotropical vascular epiphytes. Annals of the Missouri Botanical Garden 74: 205-233.

16 INTRODUCCIÓN GENERAL

Gravendeel B, Smithson A, Slik F, Schuiteman A. 2004. Epiphytism and pollinator specialization: drivers for orchid diversity? Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 359: 1450, 1523-1535. Hágsater E, Soto-Arenas MA. 2005. Epidendrum L. In: Pridgeon AM, Cribb PJ, Chase MW, Rasmussen FN (eds.), Genera Orchidacearum Volume 4, Epidendroideae (Part one). Oxford University Press, Oxford, pp. 236-251. Hágsater E, Dodson C, Sánchez L, Cervantes L, Dressler RL, Silverstone-Sopkin P. 2001. Icones Orchidacearum. The genus Epidendrum. A third century of new species in Epidendrum, fasc 4, part 3 edn. Herbario AMO, D.F. México, México. Hágsater E, Sánchez L, Garcia J. 1999. Icones Orchidacearum. The genus Epidendrum. A second century of new species in Epidendrum, fasc. 3, part 2 edn. Herbario AMO, D.F. México, México. Homeier J, Werner F, Gradstein S, Breckle S, Richter M. 2008. Potential Vegetation and Floristic Composition of Andean Forests in South Ecuador, with a Focus on the RBSF. In: Beck E, Bendix J, Kottke I, Makenschin F, Mosandl R (eds.), Gradients in a Tropical Mountain Ecosystem of Ecuador. Ecological Studies. Springer-Verlag Berlin Heidelberg, Germany, pp. 87-100. Jacquemyn H, Brys R, Merckx VS, Waud M, Lievens B, Wiegand T. 2014. Coexisting orchid species have distinct mycorrhizal communities and display strong spatial segregation. New Phytologist 202: 2, 616-627. Jacquemyn H, Wiegand T, Vandepitte K, Brys R, Roldán-Ruiz I, Honnay O. 2009. Multigenerational analysis of spatial structure in the terrestrial, food‐deceptive orchid Orchis mascula . Journal of Ecology 97: 2, 206-216. Jacquemyn H, Brys R, Hermy M, Willems JH. 2007a. Long-term dynamics and population viability in one of the last populations of the endangered Spiranthes spiralis (Orchidaceae) in the Netherlands. Biological Conservation 134: 1, 14-21. Jacquemyn H, Brys R, Vandepitte K, Honnay O, Roldán-Ruiz I, Wiegand T. 2007b. A spatially explicit analysis of seedling recruitment in the terrestrial orchid Orchis purpurea . New Phytologist 176: 2, 448-459. Jacquemyn H, Brys R, Vandepitte K, Honnay O, Roldán-Ruiz I. 2006. Fine-scale genetic structure of life history stages in the food-deceptive orchid Orchis purpurea . Molecular Ecology 15: 10, 2801-2808. Jersáková J, Malinová T. 2007. Spatial aspects of seed dispersal and seedling recruitment in orchids. New Phytologist 176: 2, 237-241. Jiménez-Pérez I. 2011. New records of orchids from the Bolivian montane forest. I part. Ecología En Bolivia 46: 1, 57-61. Jørgensen P, León-Yánez S. 2001. Catalogue of the vascular plants of Ecuador. Monographs in Systematic Botany from the Missouri Botanical Garden 75: i–viii, 1-1182. Kottke I, Haug I, Setaro S, Suárez JP, Weiß M, Preusing M, Nebel M, Oberwinkler F. 2008. Guilds of mycorrhizal fungi and their relation to trees, ericads, orchids and liverworts in a neotropical mountain rain forest. Basic and Applied Ecology 9: 1, 13-23. Machon N, Bardin P, Mazer SJ, Moret J, Godelle B, Austerlitz F. 2003. Relationship between genetic structure and seed and pollen dispersal in the endangered orchid Spiranthes spiralis . New Phytologist 157: 3, 677-687. McCormick MK, Jacquemyn H. 2014. What constrains the distribution of orchid populations? New Phytologist 202: 2, 392-400. McCormick MK, Lee Taylor D, Juhaszova K, Burnett RK, Whigham DF, O’Neill JP. 2012. Limitations on orchid recruitment: not a simple picture. Molecular Ecology 21: 6, 1511-1523. McCormick MK, Whigham DF, O'neill J. 2004. Mycorrhizal diversity in photosynthetic terrestrial orchids. New Phytologist 163: 2, 425-438.

17 Bibliografía

Migenis L, Ackerman JD. 1993. Orchids-phorophyte relationships in a forest watershed in Puerto Rico. Journal of Tropical Ecology 9: 2, 231-240. Mosandl R, Gunter S, Stimm B, Weber M. 2008. Ecuador suffers the highest deforestation rate in south America. In: Beck E, Bendix J, Kottke I, Makeschin F, Mosandl R (eds.), Gradients in a tropical mountain ecosystem of Ecuador. Springer-Verlag Berlin Heidelberg, Germany, pp. 37-40. Nathan R, Muller-Landau HC. 2000. Spatial patterns of seed dispersal, their determinants and consequences for recruitment. Tree 15: 7, 278-285. Nilsson R, Kristiansson E, Ryberg M, Hallenberg N, Larsson K. 2008. Intraspecific ITS variability in the kingdom Fungi as expressed in the international sequence databases and its implications for molecular species identification. Evolutionary Bioinformatics Online 4: 193. Otero J, Aragón S, Ackerman JD. 2007a. Site variation in spatial aggregation and phorophyte preference in Psychilis monensis (Orchidaceae). Biotropica 39: 2, 227- 231. Otero J, Flanagan N, Herre E, Ackerman JD, Bayman P. 2007b. Widespread mycorrhizal specificity correlates to mycorrhizal function in the neotropical, epiphytic orchid Ionopsis utricularioides (Orchidaceae). American Journal of Botany 94: 12, 1944- 1950. Otero J, Ackerman JD, Bayman P. 2004. Differences in mycorrhizal preferences between two tropical orchids. Molecular Ecology 13: 8, 2393-404. Otero JT, Bayman P, Ackerman JD. 2005. Variation in mycorrhizal performance in the epiphytic orchid Tolumnia variegata in vitro: the potential for natural selection. Evolutionary Ecology 19: 1, 29-43. Otero JT, Ackerman JD, Bayman P. 2002. Diversity and host specificity of endophytic Rhizoctonia-like fungi from tropical orchids. American Journal of Botany 89: 11, 1852-1858. Ouborg NJ, Piquot Y, van Groenendael JM. 1999. Population genetics, molecular markers and the study of dispersal in plants. Journal of Ecology 87: 551-568. Porras-Alfaro A, Bayman P. 2007. Mycorrhizal fungi of Vanilla: diversity, specificity and effects on seed germination and plant growth. Mycologia 99: 4, 510-525. Rasmussen HN, Rasmussen FN. 2007. Trophic relationships in orchid mycorrhiza- diversity and implications for conservation. Lankesteriana 7: 334-341. Rasmussen HN. 2002. Recent developments in the study of orchid mycorrhiza. Plant and Soil 244: 149-163. Raventós J, Mujica E, Wiegand T, Bonet A. 2011. Analyzing the spatial structure of Broughtonia cubensis (Orchidaceae) populations in the dry forests of Guanahacabibes, Cuba. Biotropica 43: 2, 173-182. Richter M, Diertl K, Emck P, Peters T, Beck E. 2009. Reasons for an outstanding plant diversity in the tropical Andes of Southern Ecuador. Landscape Online 12: 1-35. Richter M, Moreira-Muñoz A. 2005. Heterogeneidad climática y diversidad de la vegetación en el sur de Ecuador: un método de fitoindicación. Revista Peruana De Biología 12: 2, 217-238. Riofrío ML, Cruz D, Torres E, De la Cruz M, Iriondo JM, Suárez JP. 2013. Mycorrhizal preferences and fine spatial structure of the epiphytic orchid Epidendrum rhopalostele. American Journal of Botany 100: 12, 2339-2348. Riofrío ML, Naranjo CJ, Iriondo JM, Torres ME. 2007. Spatial structure of Pleurothallis, Masdevallia, Lepanthes and Epidendrum epiphytic orchids in a fragment of montane cloud forest in south Ecuador. Lankesteriana 7: 1-2, 102-106. Robbert G, Jürgen H, Dirk G (eds.), 2008. The Tropical Mountain Forest. Patterns and Processes in a Biodiversity Hotspot. Biodiversity and Ecology Series: Volume 2 edn. Göttingen Centre for Biodiversity and Ecology, Georg-August-Universität Göttingen.

18 INTRODUCCIÓN GENERAL

Ryberg M, Nilsson R, Kristiansson E, Töpel M, Jacobsson S, Larsson E. 2008. Mining metadata from unidentified ITS sequences in GenBank: A case study in Inocybe (Basidiomycota). BMC Evolutionary Biology 8: 1, 50. Santos T, Tellería J. 2006. Pérdida y fragmentación del hábitat: efecto sobre la conservación de las especies. Ecosistemas 15: 2, 3-12. Shefferson RP, Kull T, Tali K. 2008. Mycorrhizal interactions of orchids colonizing Estonian mine tailings hills. American Journal of Botany 95: 2, 156-164. Shefferson RP, Taylor DL, Weiß M, Garnica S, McCormick MK, Adams S, Gray H, McFarland J, Kull T, Tali K, and others. 2007. The evolutionary history of mycorrhizal specificity among lady's slipper orchids. Evolution 61: 6, 1380-1390. Shefferson RP, Weiss M, Kull T, Taylor D. 2005. High specificity generally characterizes mycorrhizal association in rare lady's slipper orchids, genus Cypripedium. Molecular Ecology 14: 2, 613-626. Sierra R. 1999. Propuesta preliminar de un sistema de clasificación de vegetación para el Ecuador continental. Proyecto INEFAN/GEF-BIRF y EcoCiencia edn. Rimana, Quito, Ecuador. Smith S, Cowan R, Gregg K, Chase MW, Maxted N, Fay MF. 2004. Genetic discontinuities among populations of Cleistes (Orchidaceae, Vanilloideae) in North America. Botanical Journal of the Linnean Society 145: 87-95. Smith SE, Read DJ. 2008. Mycorrhizal symbiosis, 3rd edn. Academic Press, New York, New York, USA. Suárez JP, Weiss M, Abele A, Oberwinkler F, Kottke I. 2008. Members of Sebacinales subgroup B form mycorrhizae with epiphytic orchids in a neotropical mountain rain forest. Mycological Progress 7: 75-85. Suárez JP, Weiß M, Abele A, Garnica S, Oberwinkler F, Kottke I. 2006. Diverse tulasnelloid fungi form mycorrhizas with epiphytic orchids in an Andean cloud forest. Mycological Research 110: 11, 1257-1270. Taylor DL, McCormick MK. 2008. Internal transcribed spacer primers and sequences for improved characterization of basidiomycetous orchid mycorrhizas. New Phytologist 177: 4, 1020-1033. Taylor DL, Bruns TD, Hodges SA. 2004. Evidence for mycorrhizal races in a cheating orchid. Proceedings of the Royal Society of London Biological Science Series B 271: 35-43. Tremblay RL, Meléndez-Ackerman E, Kapan D. 2006. Do epiphytic orchids behave as metapopulations? Evidence from colonization, extinction rates and asynchronous population dynamics. Biological Conservation 129: 1, 70-81. Valencia R, Pitman N, León-Yañez S JP. 2000. Libro Rojo de las plantas endêmicas del Ecuador, 1st edn. Publicaciones del Herbario QCA. Pontifícia Universidad Católica del Ecuador, Quito-Ecuador. Vargas CF, Parra-Tabla V, Feinsinger P, Leirana-Alcocer J. 2006. Genetic Diversity and Structure in Fragmented Populations of the Tropical Orchid Myrmecophila christinae var christinae. Biotropica 38: 6, 754-763. Vuylsteke M, Peleman JD, Eijk Mv. 2007. AFLP technology for DNA fingerprinting. Nature Protocols 2: 6, 1387-1398. Wallace LE. 2006. Spatial genetic structure and frequency of interspecific hybridization in Platanthera aquilonis and P. dilatata (Orchidaceae) occurring in sympathy. American Journal of Botany 93: 7, 1001-1009. Warcup JH. 1981. The mycorrhizal relationships of Australian orchids. New Phytologist 87: 371-381. Waterman RJ, Bidartondo MI. 2008. Deception above, deception below: linking pollination and mycorrhizal biology of orchids. Journal of Experimental Botany 59: 5, 1085-1096.

19 Bibliografía

Wiegand T, Gunatilleke S, Gunatilleke N. 2007. Species associations in a heterogeneous Sri Lankan Dipterocarp forest. The American Naturalist 170: E77-E95. Williams CF. 1994. Genetic Consequences of Seed Dispersal in Three Sympatric Forest Herbs. II. Microspatial Genetic Structure within Populations. Evolution 48: 6, 1959- 1972. Williams-Linera G. 2002. Tree species richness complementarity, disturbance and fragmentation in a Mexican tropical montane cloud forest. Biodiversity and Conservation 11: 10, 1825-1843. Winkler M, Hülber K, Hietz P. 2009. Population dynamics of epiphytic orchids in a metapopulation context. Annals of Botany 104: 5, 995-1004. Zelmer CD, Cuthbertson L, Currah RS. 1996. Fungi associated with terrestrial orchid mycorrhizas, seeds and protocorms. Mycoscience 37: 439-448. Zettler LW, Hofer CJ. 1998. Propagation of the little club-spur orchid (Platanthera clavellata) by symbiotic seed germination and its ecological implications. Environmental and Experimental Botany 39: 189-195.

Chapter 1

1. Spatial structure of Pleurothallis, Masdevallia, Lepanthes and Epidendrum epiphytic orchids in a fragment of montane cloud forest in south Ecuador

“El contenido de este capítulo sirvió para la publicación del artículo: Riofrío, L., Naranjo, C., Iriondo, J.M., Torres, E. Spatial structure of Pleurothallis, Masdevallia, Lepanthes and Epidendrum epiphytic orchids in a fragment of montane cloud forest in south Ecuador. Lankesteriana (2007), 7(1-2): 102– 106”.

INTRODUCTION

Orchids are the most diverse family of vascular plants in Ecuador with 228 genera and nearly 4000 species. More than 60% of these species are epiphytes,

21 Chapter 1. MATERIAL AND METHODS

being Pleurothallis R.Br., Epidendrum L., Lepanthes Sw. and Masdevallia Ruiz y Pav., with 472, 358, 314 and 226 species respectively, some of the genera with greater number of epiphytic orchids (Dodson, 1994-2003). Although Ecuador is among those countries with the highest orchid biodiversity in the world, it also has one of the highest rates of deforestation: 1.2% of the country’s forests are lost each year (FAO, 2005). Extensive deforestation practices currently taking place pose a major threat for the survival of these orchids as they are greatly dependent on the environmental conditions of the forest that sustain them, and the host trees (phorophytes) on which they grow. Thus, understanding of orchid-phorophyte interactions, as well as the patterns of spatial distribution and colonization in secondary succession forests regenerated after deforestation, is essential for the in situ conservation. Nevertheless, few studies have been conducted in this field, and scientific basis supporting population reinforcement or reintroduction actions is scarce. The purpose of this study is to assess the spatial distribution of epiphytic orchids of the above-mentioned genera in an Ecuadorian fragment of secondary montane cloud forest to infer patterns of seed dispersal and colonization. In addition, the effects of phorophyte identity and size on orchid establishment are analyzed. Specifically the questions posed are: Do the distributions of Pleurothallis, Epidendrum, Lepanthes and Masdevallia plants vary in the altitudinal range of fragment studied? Are there specific patterns in their spatial distribution resulting from seed dispersal characteristics? Do plants of these orchids exhibit any preference over the trees where they grow? Does phorophyte trunk diameter affect the establishment of these orchids? The results presented, although preliminary, provide useful information for orchid management plans.

MATERIAL AND METHODS

The study site was carried out in a fragment of regenerated forest located on the Loja-Zamora Chinchipe road, on the border of Podocarpus National Park (southern Ecuador). The age of the forest is about 30 years old, and it is characterized by a steep slope (51%), with trees 5-8 m high and lianas that are over 10 m long. Mean annual precipitation is 2700 mm, and annual mean temperature is 15.5 °C (14.4 – 17.5°C). A total of nine 10 x 10 m plots were established at 2200, 2230 and 2250 m a.s.l. (Three plots in each altitude). All trees (including fern trees), shrubs and lianas of diameter at breast height (DBH) over 1 cm were determined at the genus level, measured and mapped. The census included vascular plants belonging to more than 70 different genera. Miconia Ruiz & Pav. (148 trees), Nectandra Rol. Ex Rottb. (65 trees), Clusia L. (59 trees), Elaeagia Weed. (59 trees) and Psammisia Klotzsch (56) were the most frequent genera. Presence and abundance of all orchids occurring in the first 3 m height were also recorded. In this zone, which corresponds to zone 1 of Johansson’s scheme, the microclimatic conditions are relatively constant (Johansson, 1974).

22 Chapter 1. Spatial structure of epiphytic orchids

RESULTS AND DISCUSSION

In total 2798 orchids belonging to 12 genera were identified. Although it is difficult to make comparisons between different researches (mainly because the degree of forest disturbance varies), the high number of orchids that we found on the base of the tree trunks contrasts with other studies, which have reported no orchids or low abundance on this zone (Mehltreter et al., 2005). The most abundant orchid genus was Stelis Sw. (73.8%), followed by Epidendrum (8.5%), Pleurothallis (6.4%), Lepanthes (3.7%), Hexisea Lindl. (2.5%) and Masdevallia (1.9 %). Orchids were not uniformly distributed in the altitudinal range studies. Epidendrum and Lepanthes plants were observed at 2200 m. On the other hand, the presence of Pleurothallis and Masdevallia was similar in all the altitudinal range, although their abundance was greater in the higher zone. Thus, the 66.5% of Pleurothallis plants and the 48.1% of Masdevallia plants were found at 2250m. Altitude-related microclimatic factors may be partially responsible for this occurrence pattern, although other environmental factors independent of altitude may also play a role. Epiphyte orchids were found on 325 of the 1025 recorded trees, shrubs and lianas. The most frequent trees in the fragment were also the ones that had the greatest richness and number of orchids. Of the four genera studies, Pleurothallis occupied the greatest number of trees (57), while Masdevallia was present in only 29 (Table 1.1). The average number of individuals per phorophyte was small in all of them (ranging from 4.8 in Epidendrum to 1.8 in Masdevallia), but the variance was large especially in Epidendrum (±34.8) and Pleurothallis (±16.3). In order to know how the individuals are distributed among phorophytes, Morisita’s index (IM) (Hurlbert, 1990) was calculated considering the phorophyte as sampling unit. According to this aggregation index (Table 1.2), two different patterns were detected: Epidendrum and Pleurothallis plants tended to be clumped (IM values were significantly different from 1), while Lepanthes and Masdevallia plants were randomly distributed. Differences in seed dispersal process may explain this result. Thus, the aggregated pattern observed in Epidendrum and Pleurothallis may due to limited dispersal ability of their seeds. If this were the case, there would be a higher probability of finding a plant of its same genus in a near-by tree than in a more distant tree. To test this hypothesis, a bivariate point pattern analysis was performed in those plots where the number of phorophytes was greater than eight (Diggle, 1983; Wiegand & Moloney, 2004). For Pleurothallis, the values of K21(r) – K22(r) were inside the confidence interval of the null hypothesis of random labeling for the range of distances 0-5 m (Fig. 1.1), which means that presence of these orchids is at random. Similar results were obtained for Epidendrum. Thus, since there is no contagious distribution between one phorophyte and nearby trees, seed dispersal in Epidendrum and Pleurothallis is not limited to short distances. This conclusion is also supported by mean distance between phorophytes (4.0 ± 5.6 m for Epidendrum and 4.5 ± 6.8 m for Pleurothallis), which is not significantly different than the mean distance between all trees in the plots (4.7 ± 2.5 m), and maximum distance to the

23 Chapter 1. RESULTS AND DISCUSSION

nearest neighbor (13.2 m for Epidendrum and for Pleurothallis). Other reasons, such as differences in life cycle or reproductive biology could explain the presence of these two distribution patterns.

Table 1.1. Distribution of Epidendrum, Pleurothallis, Lepanthes or Masdevallia orchids on their perspectives host trees in a secondary montane cloud forest in South Ecuador. For each orchid genus, frequency of host trees (first column) and frequency of orchid individuals on the different host tree genera (second column) are shown.

Host genus Epidendrum Pleurothallis Lepanthes Masdevallia nº of trees nº of orchids nº of trees nº of orchids nº of trees nº of orchids nº of trees nº of orchids Albuta Aubl. 1 3 ------Alchornea Sw. - - 1 18 - - - - Alzatea Ruiz y Pav. - - 1 1 - - - - Aniba Aubl. - - 1 2 - - - - Anthurium Schott 1 7 ------Ardisia Gaertn 2 10 1 5 - - - - Axinaea Ruiz & Pav. - - 2 2 - - 1 2 Calyptranthes Sw. 1 1 3 16 - - - - Cinchona L. 1 7 ------Clethra L. - - 1 1 - - - - Clusia L. 8 28 2 2 3 4 1 2 Cyathea Sm. 2 2 1 21 - - - - Endlichera C. Presl 1 18 ------Elaeagia Weed. 4 12 5 10 4 4 3 8 Eugenia L. ------1 1 Fagamea Aubl. 1 3 ------Ficus L. - - - - 1 1 - - Graffenrieda DC: 2 2 2 2 2 8 - - Guarea F. Allam. Ex L. - - 1 1 - - - - Hedyosmun Sw. 1 36 ------Helicostylis Trécul 2 14 6 28 3 8 2 3 Hydrangea L. 1 3 ------Mabea Aubl. 1 5 ------Markea Rich. - - 1 2 - - - - Maytenus Molina - - 1 3 2 5 - - Miconia Ruiz & Pav. 3 8 7 7 12 25 3 4 Mikania Willd. 1 2 - - 1 5 - - Myrsine L. 2 16 1 6 3 8 2 3 Nectandra Rol. Ex 1 5 3 7 3 8 3 4 Rottb. Ocotea Aubl. - - 1 1 - - 1 1 Palicourea Aubl. 2 19 2 2 - - 2 8 Persea Mill. - - 1 1 - - - - Piper L. - - - - 1 3 - - Psammisia Klotzsch 2 4 5 22 3 4 6 7 Psychotria L. 2 3 - - 3 4 - - Ruagea H. Karst. - - 1 1 - - - - Turpinia Vent. - - 1 1 - - - - Fallen tree 3 6 3 13 4 8 1 4 Unidentified liana 1 4 2 5 2 5 - - Unidentified tree 1 13 - - - - 2 3 Total 50 239 57 179 51 104 29 52

24 Chapter 1. Spatial structure of epiphytic orchids

Table 1.2. Average number of individuals per host tree (mean ± variance) and Morisita’s index IM for Epidendrum, Pleurothallis, Lepanthes or Masdevallia orchids in a secondary cloud forest in South Ecuador. Values in parentheses are minimum and maximum ***. P <0.001.

Orchid genus Individuals per host tree IM Epidendrum 4.8 ± 34.8 2.32*** (1-36) Pleurothallis 3.1 ± 16.3 2.34*** (1-21) Lepanthes 2.0 ± 2.4 1.10 (1-7) Masdevallia 1.8 ± 1.3 1.09 (1-5)

Fig. 1.1 A) Spatial distribution of trees in plot 1 located at 2250 m. Triangles indicates trees with Pleurothallis orchids. B) Bivariate point pattern analysis plotting the L12 function across distance. Dotted lines represent the confidence interval of random labeling null hypothesis. No phorophyte specificity was observed for any of the epiphytic orchids included in the study. Epidendrum and Pleurothallis grew o more than 20 different genera, and Lepanthes and Masdevallia on more than 10 (Table 1.2). Nevertheless, Epidendrum was more frequent on Clusia, Pleurothallis and Lepanthes on Miconia, and Masdevallia on Psammisia. Preference patterns in orchids have also

25 Chapter 1. Bibliography

been reported by other authors (Migenis & Ackerman, 1993; Díaz-Santos, 2000; Trapnell & Hamrick, 2006) although the reasons why orchids occur on particular species remain unclear. The possible effect of phorophyte size on orchid establishment was explored calculating the Spearman’s correlation coefficient (rS) between DBH and orchid abundance for each of these four genera. No relationship was found in any of them (rS=-0.12, P=0.42 for Epidendrum, rS=0.15, P=0.27, for Pleurothallis, rS=0.09, P=0.52 for Lepanthes, rS=0.17, P=0.37, for Masdevallia), which means phorophyte trunk diameter does not seem to be a crucial factor for orchid colonization. At present, other phorophyte physical characteristics such as bark stability and roughness, and substrate moisture conditions are being investigated. In conclusion, this study shows the existence of different patterns of presence and abundance depending on each orchid genus. Anyhow, colonization of new trees does not seem to be constrained by limited seed dispersal. Light conditions may be a more important factor for epiphytic orchid establishment than phorophyte identity and size. The abundance of orchids in the lower section of the tree trunks in this regenerating forest is in clear contrast with previous reports made on primary or non-disturbed forest. This outlines the importance of taking into account the different succession states of the forest in which the orchids occur. Finally, although pattern analysis can be helpful in identifying the causes of present spatial structure, additional experimental studies are needed to determine the underlying processes originating these distributions. Bibliography

Díaz-Santos F. 2000. Orchid preference for host tree genera in a Nicaragua tropical rain forest. Selbyana 21: 1, 25-29. Diggle PJ. 1983. Statistical analysis of spatial point patterns. Academic Press, London. Dodson C. 1994-2003. Native Ecuadorian Orchids, 4th edn. Colina, Quito. FAO. 2005. State of the world’s forests, 1st edn. Food and Agriculture Organization of the United Nations, Rome. Hurlbert S. 1990. Spatial distribution of the montane unicorn. Oikos 58: 257-271. Johansson D. 1974. Ecology of vascular epiphytes in West African rain forest. Acta Phytogeographica Suecica 59: 1-136. Mehltreter K, Flores-Palacios A, García-Franco J. 2005. Host preference of low-trunk vascular epiphytes in a cloud forest of Veracruz, Mexico. Journal of Tropical Ecology 21: 6, 651-660. Migenis L, Ackerman JD. 1993. Orchids-phorophyte relationships in a forest watershed in Puerto Rico. Journal of Tropical Ecology 9: 2, 231-240. Trapnell DW, Hamrick JL. 2006. Variety of phorophyte species colonized by the neotropical epiphyte, Laelia rubescens (Orchidaceae). Selbyana 27: 1, 60-64. Wiegand T, Moloney KA. 2004. Rings, circles, and null-models for point pattern analysis in ecology. Oikos 104: 2, 209-229.

Chapter 2

2. Analysis of the factors that condition the presence of Epidendrum rhopalostele

INTRODUCTION

Epiphytic orchids are the most diverse group of vascular epiphytes and their diversity increases in some habitats of neotropical forest by environmental conditions such as humidity, elevation and soil properties that allow great differentiation in microsites (Gentry & Dodson, 1987). They are adapted to different environmental conditions, are highly evolved and have high stress tolerance (Benzing, 2012). They have particular adaptations, such as aerial roots that allow fixing to the substrate, besides absorption of nutrients (Garay, 1978; Zotz & Winkler, 2013) and the velamen that lets rapid water absortion during rain (Gradstein, 2008). The epiphytic orchids are a characteristic component of tropical rain forest in diversity and biomass, mainly in the mountain cloud forest were their diversity is higher (Gravendeel et al., 2004; Gradstein, 2008). In Ecuador more than 60% of orchids are epiphytes (Dodson, 1994-2003) and are present in the mountain cloud forest.

27 Chapter 2. INTRODUCTION

Epiphytic orchids are dependent on environmental conditions of forests and host trees (phorophytes) where they grow. Loss of habitat is the main threat to the survival of epiphytic orchids (Jacquemyn et al., 2005; Gradstein, 2008). A high rate of deforestation is present in Ecuador, where 1.2% of forest is lost every year (FAO, 2005). In addition to habitat loss this process also causes habitat fragmentation (Hágsater & Dumond, 1996; Mosandl et al., 2008). Policies for biodiversity protection have been created, but more research is needed to generate knowledge on the ecology of species and to allow the generation of suitable conservation strategies to contribute to the policies proposed. Thereby, an important aspect of orchid conservation is habitat protection and maintenance of orchid populations through the understanding of epiphytic orchid population dynamics (Winkler & Hietz, 2001; Raventós et al., 2011). In this context, one key element is the regeneration of orchid epiphyte populations in the secondary tropical montane forests, which normally takes place at a very slow rate (Nöske et al., 2008; Gradstein, 2008). For this reason, the knowledge of the factors that determine the colonization of a secondary mountain cloud forest by epiphytic orchids and their persistence is necessary for the implementation of an adequate conservation strategy. Epiphytic habit predisposes orchids to populations distributed in patches (Tremblay et al., 2006). Availability of suitable phorophyte (Flores-Palacios & Ortiz-Pulido, 2005), heterogeneous environmental conditions, mycorrhizal associations (McCormick & Jacquemyn, 2014) and plant-plant interactions (Wiegand et al., 2007a; Jacquemyn et al., 2007b; Jacquemyn et al., 2009) can condition epiphytic orchid population abundance and distribution (McCormick & Jacquemyn, 2014). All these factors have a spatial component that explains species coexistence and distribution (Wiegand et al., 2007a). In this sense, spatial pattern analysis can provide insights into the many biological and ecological processes present at various scales. This analysis can be helpful in understanding seed dispersal processes, species coexistence or environmental heterogeneity (Wiegand & Moloney, 2004; Wiegand et al., 2007a; Wiegand et al., 2007b; Wiegand et al., 2009). Seed dispersal is the principal mechanism by which plants move in space (Nathan & Muller-Landau, 2000) and can be estimated through the study of spatial patterns of plant populations (Wiegand & Moloney, 2004; Epperson, 2005). Short-distance seed dispersal next to parent plants can produce an aggregated spatial pattern that increases the probability of success for microsite availability and pollinator selection or micorrhizal preference (Shimamura et al., 2007; Cousens et al., 2008; Raventós et al., 2011; Riofrío et al., 2013), while long- distance seed dispersal allows colonization of new habitats (Wright et al., 2008), reduces within-species competition by resources (Cousens et al., 2008) and maintains genetic variability in populations (Nathan et al., 2002; Bialozyt et al., 2006). The ability of plants to disperse their seeds is critical when there is habitat fragmentation (Schurr et al., 2008) and environmental heterogeneity (Wiegand et al., 2007a; Getzin et al., 2008; Belinchón et al., 2011). Environmental

28 Chapter 2. Factors than condion the presence of E. rhopalostele

heterogeneity, with the presence of patches with conditions that are suitable for the species in a matrix where the species cannot thrive, can lead to aggregated spatial patterns in the distribution of the species (Wiegand & Moloney, 2004; Wiegand et al., 2007a; Getzin et al., 2008). In orchids, reduced seed sizes are efficient for long-distance seed dispersal by wind. This mechanism is considered to be the predominant one for seed dispersal (Healey et al., 1980; Arditti & Ghani, 2000). The absence of endosperm allows reduction in weight and size facilitating colonization of host trees and new areas (Yoder et al., 2010). However, in montane cloud forest or rain forest the rain could be another mechanism for orchid seed dispersal (Vittoz & Engler, 2007). Drops of rain could transport seeds along branches of trees or the small ravines could move the tiny seeds across nearby trees downslope. Since the distribution of epiphytic orchids could be limited by the availability of host trees, several studies have been conducted to evaluate factors affecting their distribution focusing on characteristics like host tree identity (i.e., degree of preference for a particular species or taxonomic group) (Migenis & Ackerman, 1993; Tremblay et al., 1998; Gowland et al., 2011; Adhikari et al., 2012b), host traits, like size, age, architecture, bark roughness, bark, pH (Bergstrom & Carter, 2008; Adhikari et al., 2012b), dead versus living trees, and shade tolerant versus pioneer trees. The potential preference for a particular microhabitat within a tree (e.g., presence or absence of moss, more or less bark roughness, distribution in altitude within a tree) should also be considered. On the other hand, plant-plant interactions, could also be acting and conditioning epiphyte orchid distribution. Mother plant – seedling facilitative interactions can generate spatial aggregation of adults and recruits (Seidler & Plotkin, 2006; Jacquemyn et al., 2007b; Jacquemyn et al., 2009; McCormick & Jacquemyn, 2014). However, intraspecific competitive interactions at different stages of the life cycle will promote spatial dispersion (Antonovics & Levin, 1980; Silander & Pacala, 1985; De La Cruz et al., 2008). Epidendrum rhopalostele Hágsater & Dodson is a photosynthetic epiphytic orchid without pseudobulbs (Dodson, 1994-2003; Hágsater et al., 2001), adapted to grow in montane secondary cloud forest in Andean forest, where high topographic diversity and abiotic heterogeneity generate distinct vegetation types and variety of microhabitats (Homeier et al., 2008). Due to deforestation and illegal trade, population sizes have been reduced drastically in this and similar forests. In this context, E. rhopalostele constitutes a relevant species to study the factors that condition the colonization and permanence of epiphytic orchids in montane forests. Although the number of studies in vascular epiphytes is growing (Zotz & Hietz, 2001; Laube & Zotz, 2003; Zotz & Bader, 2009; Zotz et al., 2014; Einzmann et al., 2015), little is known about the factors that influence the establishment of orchids in montane forests (Kromer et al., 2005; Werner & Gradstein, 2008; Gradstein, 2008; Robbert et al., 2008) and far less in montane cloud forests of secondary succession. In the present work, we analyzed some factors that could condition the presence of E. rhopalostele, such as environmental heterogeneity,

29 Chapter 2. MATERIAL AND METHODS

range of seed dispersal, and plant-plant interactions. We used this knowledge to discuss the implications for the conservation and management of this epiphytic species. Specifically, we asked the following questions: 1) How are E. rhopalostele individuals arranged in space? 2) How are environmental conditions and microsite availability affecting E. rhopalostele distribution? 3) Is the spatial distribution of the orchid conditioned by limited seed dispersal? 4) What are some other factors that condition their presence? To answer these questions association analysis (Agresti, 2002) and spatial point pattern techniques were used (Ripley, 1976; Dixon, 2002; Wiegand & Moloney, 2004; Baddeley & Turner, 2005; De La Cruz et al., 2008). The spatial analysis was carried out at three levels: 1) spatial distribution of trees in the study fragment, 2) spatial pattern of trees hosting E. rhopalostele (hereafter, phorophytes) and trees that do not host E. rhopalostele (hereafter, non-phorophyte trees) in the study fragment, 3) spatial pattern of E. rhopalostele plants in the forest fragment including spatial relationships between adults and juveniles (i.e., aggregation or segregation between adults and juveniles). Taking into account these questions we posed the following hypotheses: 1) The trees of the forest fragment will be spatially aggregated as a response to the environmental heterogeneity in the forest fragment. Variations in the topography and associated factors in terms of wind exposure, nutrient availability, humidity, etc. could generate this pattern. 2) The presence of E. rhopalostele in certain trees depends on additional environmental factors, which could be related with the host traits (species, dbh, type of bark, etc) or the conditions offered by the host (presence of mycorrhiza, presence of moss, etc.). 3) The colonization of new trees by E. rhopalostele inside the forest will be limited due to the difficulties for seed dispersal derived from the dense vegetation matrix of the montane cloud forest. In this context, we expect that phorophytes will show an additional aggregated pattern. 4) Individuals of E. rhopalostele within phorophytes will be clustered as a result of a facilitation process of adults to juveniles or limited seed dispersal. Differences in the environment at a microscale may also generate this pattern.

MATERIAL AND METHODS

Study site The study site (~ 1 ha) is located on the eastern slope of Cordillera Real in the Andes of southern Ecuador on the border of Podocarpus National Park near the Loja-Zamora road in Zamora-Chinchipe province, at around 2250 m a.s.l. This regenerated forest, approximately 35 years old, is a secondary succession forest. It is characterized by a steep slope (51%) (Fig. 2.1) and classified as evergreen,

30 Chapter 2. Factors than condion the presence of E. rhopalostele

upper montane forest (Beck et al., 2008a). It has 5-8 m high trees and lianas that are over 10 m long. The most diverse and abundant seed plant families are Orchidaceae, Melastomataceae, Ericaceae, and Rubiaceae. Mean annual temperature is 20.8°C (4.7–25.5°C) and annual precipitation is 2193 mm. A moderate rainy season typically extends from April to July. On average, there is 1 month with less than 100 mm of rainfall during the driest part of the year, from October to January. Shorter dry spells of 1–2 wk. are frequent. Fog is common at this elevation and provides an additional water surplus of 9.6% (Bendix et al., 2008).

Fig. 2.1. The study fragment of regenerated montane cloud forest has a steep slope of ~ 51% .

Study species Epidendrum rhopalostele is a photosynthetic, epiphytic species native to Ecuador, Peru and Bolivia, present in evergreen montane forests. It belongs to the alpicolum group, with a single racemose apical inflorescence of 10–25 light green flowers (Hágsater et al., 2001). It is very similar to Epidendrum dialychilum Hágsater & Dodson, both having the lip free from the column, but differs from the latter by a) its linear-lanceolate and acuminate lip; b) the long, filiform, acuminate petals; and c) the stigmatic cavity only in the apical third of the column.

Spatial georeferencing of trees and E. rhopalostele All trees (dead and alive) with dbh > 1 cm were georeferenced. Their x, y and z coordinates were obtained using a total station theodolite (Zeiss Elta 6). Twelve base points where established to cover the whole study area. The determination of

Chapter 2. Factors than condion the presence of E. rhopalostele

Characterization of trees and E. rhopalostele

We taxonomically determined each georeferenced tree as far as possible according to Flora of Ecuador (Harling & Sparre, 1980; Harling & Andersson, 1986-1998), Catalogue of the Vascular Plants of Ecuador (Jørgensen & León- Yánez, 2001) and others (Renner et al., 1990; Jørgensen & Ulloa, 1994; Nishida, 1999; Neill & Ulloa, 2011) and collected information of diameter at breast height (dbh), height and bark rugosity. For bark rugosity two classes (smooth and rough) were primarily identified through visual observation. Furthermore, all tree species were assigned to one of two functional categories (shade-tolerant or pioneer) according to literature (Finegan, 1992; Poorter et al., 2006). We also recorded whether each tree was dead or alive. Trees without branches or leaves, many of which showed clear signs of bark decomposition, and large branches fallen on the ground were classified as dead. The characteristics of E. rhopalostele individuals, including plant size, leaf number (Fig. 2.4), flower and fruit number, location on tree (trunk or branch) and distance to soil were recorded. The life stage of the orchid (juvenile or adult) and the presence or absence of moss on its base were also recorded.

Fig. 2.4. E. rhopalostele on a phorophyte A) Bottom of orchid. B) Orchid ramets. C) Inflorescence. Orchid size was taken considering distance from point A to C as shown in the picture (gray line).

33 Chapter 2. MATERIAL AND METHODS

Data analysis

Association studies for E. rhopalostele We employed contingency table analysis (Agresti, 2002) to test the possible association of E. rhopalostele with a particular type of tree. The following contrasts were tested: a) dead vs. living trees, b) pioneer vs. shade-tolerant trees, and c) association to particular phylogenetic clades (i.e., orchid associations with particular families of plants). We also tested the preference for a particular microhabitat within the host tree including: a) bark type (smooth or rough bark), b) presence or absence of moss, c) vertical distribution on the tree (measured as distance to the ground: h ≤ 1 m, 1 < h ≤ 2 m, h > 2 m). In addition, the vertical distribution of the orchid on phorophytes was contrasted between adult and juvenile plants. Pearson´s Chi-square tests with Yates´s continuity correction were applied. Because Clusia alata Planch. & Triana is the living species that had the greatest number of trees with E. rhopalostele, a randomization test was carried to determine if the orchid was positively associated to this phorophyte. Taking into account the abundance of the different tree species in the fragment and that there are 25 phorophytes of E. rhopalostele, we performed 10000 random extractions of 25 trees from the total of trees in the fragment and recorded for each extraction the number of trees belonging to Clusia alata. From the frequency distribution obtained, we calculated the probability of obtaining a number of trees equal or above the number of Clusia alata individuals recorded in our study. All these analyses were carried out using the package ecespa (De La Cruz et al., 2008) in the R environment for statistical computing (R Development Core Team., 2011).

Spatial point pattern analyses Spatial point pattern analyses were used to describe the spatial distribution of trees, phorophytes and orchids. We employed Ripley’s K function K(r) and nearest-neighbor function G(r). Univariate functions: Ripley’s K function (Ripley, 1976) is a measure of the average number of points found within a set distance r, from each point, divided by the mean intensity of the pattern. The mean intensity is simply the number of points per area, λ = n / πr2. Usually K is compared with true values of K for a completely random (Poisson) distribution where K(r) = πr2. Deviation between empirical and theoretical K curves may suggest spatial clustering (i.e., K(r) > πr2) or spatial regularity (i.e., K(r) < πr2) (Wiegand & Moloney, 2014). Nearest- neighbor function G(r) is the cumulative distribution of distance r to the nearest neighbor, that is, the probability that average points of the pattern have its nearest neighbor within distance r (Diggle, 2003; Illian et al., 2008). The probability of 2 having no points within an area A= πr is given by a Poisson distribution P(k=0,A)=exp(-λ A)=exp(λπr2). Then, G(r)=1-exp(-λπr2). For a clustered pattern, G(r)>1-exp(-λπr2) and, for a hiperdispersed pattern, G(r)<1-exp(-λπr2) (Wiegand & Moloney, 2014).

34 Chapter 2. Factors than condion the presence of E. rhopalostele

To facilitate the interpretation of the results, in the univariate analyses, K(r) was transformed into the frequently used linearized version L(r), where L(r) = [K(r) /π]1/2.

Bivariate associations: The bivariate K-function K12 (r) (also called K-cross or Kij) is defined as the expected number of points of type 2 within distance r of an arbitrary point of type 1, where λ2 is the intensity of points of type 2 (i.e. the expected number of points 2 per unit area) (Dixon, 2002; Wiegand & Moloney, 2004). The difference between functions K1(r) – K2(r) evaluates the differences in the intensity of aggregation of the two point patterns and K1(r) – K12(r) or K2(r) – K12(r) evaluate the degree of segregation of each individual pattern (Diggle, 2003; De La Cruz et al., 2008). Bivariate function for G12(r) indicates the cumulative distribution of distance from a random point of type 1 to the nearest neighbor of type 2, where λ2 is the intensity of points of type 2 (Baddeley & Turner, 2005). Deviation between empirical and therorical G12 functions suggests dependence between points of type 1 and 2. To evaluate which model describes better the empirical spatial structure of trees, phorophytes and orchids, we generated confidence envelopes based on the simulation of different models. The goodness-of-fit (GoF) test (Loosmore & Ford, 2006) was computed to evaluate the overall fit of spatial models, using µ statistic. The spatial analyses were carried out with the Kest and Gest functions of the package spatstat (Baddeley & Turner, 2005) and K1K2 of the package ecespa (De La Cruz et al., 2008) in the R environment (R Development Core Team., 2011). The specific analyses conducted at the different spatial scales to test the hypotheses presented in the introduction, and the corresponding null models used are specified below. In all spatial analyses a mask (window) delimiting the polygon coordinates of forest fragment was used to prevent a virtual aggregation.

Analysis 1. Spatial distribution of trees As a reference point we described the spatial pattern of all trees in the forest fragment with the L function. Next, the results were compared to a complete spatial randomness (CSR) null model. This model, also referred as homogeneous Poisson process, assumes that 1) intensity λ of the pattern is constant over the study plot, and the number of points in the study plot of area A follows a Poisson distribution with an expected mean of λA, and 2) the points of the process are independently distributed. Thus, this means that any point of the pattern has an equal probability of occurring at any position in the study plot, and that there is no interaction between the points of the pattern (Diggle, 2003; Illian et al., 2008; Wiegand & Moloney, 2014). Thus, L(r)=0 indicates CSR, L(r) >0 indicates tree aggregation, and L(r)<0 indicates tree regularity. As the pattern revealed that the distribution did not follow a CSR null model, a homogeneous Poisson cluster process (HPC) null model was tested. This consists of randomly located cluster centers. Around each cluster center, points (in this case, trees) are positioned according to a radially symmetric Gaussian distribution defined by the parameters ρ (density of cluster center), and σ (standard deviation of clump size). The mean

35 Chapter 2. MATERIAL AND METHODS

squared distance of a tree from the center of its cluster corresponds to 2σ2. A HPC process incorporates a clustering mechanism (Diggle, 2003). Thus, the aggregation can be caused by local seed dispersal or gap recruitment (Plotkin et al., 2000; Seidler & Plotkin, 2006). Confidence envelopes associated to a given null model were generated by calculating for each distance r the 5th-lowest and 5th-highest K (r) and G(r) values from 199 Monte Carlos simulations of each null model.

Analysis 2. Spatial distribution of phorophytes As a first approach, we tested the spatial distribution of phorophytes against a null model of random labeling (Wiegand & Moloney, 2004). As this test discarded the random labeling hypothesis (GoF test: p-value <0.05), we tested the spatial pattern of phorophytes against a new null model, that we call "Poisson cluster labeling" (PCL). The PCL model assumes that 1) there exist some parent events that follow a Poisson process with intensity ρ; 2) each parent event produces a random number of offspring according to a Poisson distribution with mean µ = λ/ρ (where λ is intensity of offspring) and 3) the offspring disperse to neighbor trees around each parent event follows a bivariate Gaussian distribution η(r, σ), with standard deviation σ. The PCL model is fitted, as any other Poisson cluster model, by minimum contrast (Wiegand & Moloney, 2004) from the pattern of observed phorophytes. To simulate it, first a Poisson pattern of parent events with intensity ρ is simulated and then, around each parent event, a random number of trees (from the Poisson distribution with mean µ = λ/ρ) are selected (i.e., "labeled") among the whole set of trees (both phorophytes and non-phorophytes) with probability equal to a Gaussian probability density with mean 0 and standard deviation σ. The PCL test involves comparing the values of the K(r) function of observed phorophytes with the 5th-lowest and 5th-highest value of K functions from 199 Monte Carlo simulations of the PCL null model.

Analysis 3. Relative spatial distribution of phorophytes and non-phorophytes. We conducted bivariate analyses to test the relative spatial distribution of phorophytes and non-phorophytes of E. rhopalostele within the fragment by means of differences between univariate and bivariate K functions (Dixon, 2002; De La Cruz et al., 2008). To contrast their patterns Poisson clustering labeling null models were used. Deviations from the null hypothesis of Poisson cluster labeling was evaluated by the difference between pairs of functions of K and G (Wiegand & Moloney, 2004; De La Cruz et al., 2008). K1 (r) -K2 (r) and G1 (r) - G2 (r) evaluate the difference in the intensity of aggregation of the two point patterns (K1/G1: phorophytes and K2/G2: non-phorophytes), K1(r)-K12(r) and G1(r)-G12(r) evaluate segregation between processes. Thus, K(r)1-K(r)12 > 0 indicates that phorophytes are more surrounded by phorophytes than expected (Dixon, 2002; De La Cruz et al., 2008) and G1(r)-G12(r) > 0 indicates that phorophytes are closer to other phorophytes than expected (Diggle, 2003). Confidence limits were generated by 199 simulations of the PCL model and using the 5th-lowest and 5th-highest value of K/G functions from 199 Monte Carlo simulations of the PCL null model. The analyses were made with a sequential r between 0 – 10 m, and a spatial resolution of 0.25 m.

36 Chapter 2. Factors than condion the presence of E. rhopalostele

Analysis 4. Spatial distribution of the orchid population To describe the basic characteristics of the spatial pattern of the E. rhopalostele population a univariate analysis was used. As the pattern revealed that the distribution of orchid population did not follow a CSR null model, a homogeneous Poisson cluster process (HPC) null model was also tested. Confidence envelopes associated to a given null model were generated by calculating for each distance r the 5th-lowest and 5th-highest K (r) values from 199 Monte Carlos simulations of each null model.

Analysis 5. Spatial association of adults and juveniles To test the fourth hypothesis (i.e., an expected aggregation pattern between adults and juveniles due to facilitation of the former over the latter) we used a bivariate K function (Dixon, 2002; De La Cruz et al., 2008). In this case, we tested the observed differences of K functions against the null model of random labeling. We evaluated the model comparing the observed difference K1-K2 (where K1 corresponds to adults and K2 corresponds to juveniles). K11(r)-K22(r) evaluates if the pattern of points of adults is more aggregated than that of juveniles, K11(r)- K12(r) evaluates segregation between the two processes. Thus, K11(r)-K12(r) > 0 would indicate than adults tend to be surrounded by adults and K2(r)-K12(r) > 0 that juveniles tend to be surrounded by more juveniles (Dixon, 2002; De La Cruz et al., 2008). Confidence envelopes of the K function differences were generated by 999 simulations of the null model, using the lowest and highest values of the test statistic. The analyses were carried out using a sequential radius range from 0 to 8 m by intervals of 0.01 m of distance.

RESULTS

Considering phorophytes and non-phorophytes of E. rhopalostele, 714 trees with dbh between 1 - 42 cm were found and mapped at the study site. The site contained 103 different tree species and three additional groups, consisting of lianas, dead trees and unidentified trees, in all cases without adequate characters for morphological identification (Table 2.1. The most abundant species were Palicourea angustifolia Kunth. with 39 individuals, Elaeagia karstenii Standl. with 37 individuals and Clusia alata with 35 trees. The lianas and dead trees conformed other important groups with 75 and 73 individuals, respectively.

Table 2.1. Species identification of trees present in the forest fragment. Family and frequency of the tree taxa found in the study fragment.

Taxa Family Frequency Abarema killipii Bareby & Grimes Mimosaceae 3 Ageratina dendroides R.M. King & Rob. Asteraceae 5 Alchornea grandiflora Müll. Arg. Euphorbiaceae 12 Alchornea triplinervia (Spreng.) Müll. Arg. Euphorbiaceae 3 Aniba sp. Aubl. Lauraceae 15 Beilschmiedia sp. 1 Lauraceae 2 Beilschmiedia tovarensis (Klotzch & Karst. Ex Meisn.) Lauraceae 5

37 Chapter 2. RESULTS

Cavendishia bracteata (Ruiz & Pav. Ex J. St.-Hil.) Hoerold Ericaceae 12 Cavendishia sp. Ericaceae 1 Cedrela sp. 1 Meliaceae 2 Cedrela sp. 2 Meliaceae 1 Cestrum schlechtendalii G. Don Solanaceae 2 Chusquea sp. Poaceae 1 Clethra revoluta (Ruiz & Pav.) Spreng. Clethraceae 6 Clusia alata Planch. & Triana Clusiaceae 35 Cyathea peladensis (Hierro.) Domin Cyatheaceae 17 Dead tree Dead tree 73 Elaeagia karstenii Standl. Rubiaceae 37 Endlicheria sp. 1 Lauraceae 3 Endlicheria sp. 3 Lauraceae 3 Escallonia paniculata (Ruiz & Pav.) Roem. & Schult Glossulariaceae 19 Eschweilera sessilis A.C. Sm. Lecythidaceae 12 Eugenia sp. Myrtaceae 10 Faramea anisocalyx Poepp. & Endl. Rubiaceae 7 Faramea coerulescens K. Schum. & K. Krause Rubiaceae 5 Geissanthus ecuadorensis Mez Myrsinaceae 10 Geonoma densa Linden & H. Wendl. Arecaceae 3 Graffenrieda emarginata (Ruiz & Pav.) Triana Melastomataceae 3 Graffenrieda harlingii Wurdack Melastomataceae 4 Graffenrieda sp. 1 Melastomataceae 3 Graffenrieda sp. 2 Melastomataceae 2 Guatteria sp. 2 Annonaceae 2 Hedyosmum spectabile Todzia Chlorantaceae 5 Helicostylis tovarensis (Klotzsch & H. Karst.) C.C. Moraceae 20 Hyeronima asperifolia Pax & K. Hoffm. Euphorbiaceae 6 Hyeronima moritziana Muller Arg. Euphorbiaceae 7 Ilex hippocrateoides Kunth Aquifoliaceae 3 Ilex sp. Aquifoliaceae 1 Liana Liana 75 Licaria subsessilis van der Werff Lauraceae 5 Maytenus sp. 1 Celastraceae 3 Maytenus sp. 2 Celastraceae 1 Meliosma herbertii Rolfe Sabiaceae 1 Meliosma sp2 Sabiaceae 1 Meriania rigida (Benth.) Triana Melastomataceae 5 Meriania sp. Melastomataceae 1 Meriania sp. 1 Melastomataceae 5 Meriania sp. 2 Melastomataceae 7 Miconia micropetala Cogn. Melastomataceae 1 Miconia punctata (Desr.) D. Don ex DC. Melastomataceae 1 Miconia sp. 1 Melastomataceae 2 Miconia sp. 2 Melastomataceae 1 Miconia sp. 3 Melastomataceae 1 Miconia sp. 4 Melastomataceae 19 Miconia sp. 5 Melastomataceae 4 Miconia sp. 6 Melastomataceae 2 Miconia sp. 7 Melastomataceae 3 Miconia sp. 8 Melastomataceae 11 Miconia sp. 9 Melastomataceae 5 Miconia sp. 10 Melastomataceae 2 Miconia tinifolia Naudin Melastomataceae 5 Mollinedia sp. Monimiaceae 4 Monnina sp. Polygalaceae 2 Myrcia sp. Myrtaceae 1 Myrcianthes sp. Myrtaceae 8 Myrica pubescens Humb. & Bonpl. ex Willd. Myricaceae 1 Myrsine andina (Mez) Pipoly Myrsinaceae 13 Nectandra laurel Klotzsch ex Nees Lauraceae 3 Nectandra reticulata (Ruiz & Pav.) Mez Lauraceae 4 Nectandra sp. 2 Lauraceae 1 Ocotea sp. Lauraceae 4 Palicourea jaramilloi C.M. Taylor Rubiaceae 4

38 Chapter 2. Factors than condion the presence of E. rhopalostele

Palicourea andrei Standl. Rubiaceae 2 Palicourea angustifolia Kunth Rubiaceae 39 Palicourea sp. 1 Rubiaceae 1 Palicourea sp. 2 Rubiaceae 1 Persea sp. 1 Lauraceae 1 Persea sp. 2 Lauraceae 3 Piper obliquum Ruiz & Pav. Piperaceae 1 Piper sp. 1 Piperaceae 21 Piptocoma discolor (Kunth) Pruski Asteraceae 2 Podocarpus oleifollius D. Don ex Lamb. Podocarpaceae 1 Prunus huantensis Pilg. Rosaceae 2 Prunus opaca (Benth.) Walp. Rosaceae 9 Psychotria sp. 1 Rubiaceae 2 Psychotria tinctoria (Aubl.) Raeusch. Rubiaceae 9 Ruagea glabra Triana & Planch. Meliaceae 4 Saurauia tomentosa Griseb. Actinidaceae 7 Schefflera acuminata (Ruiz & Pav.) Harms Araliaceae 3 Schefflera sp. 2 Araliaceae 4 Schefflera sp 3 Araliaceae 2 Siparuna cascada Monimiaceae 2 Sloanea sp. Elaeocarpaceae 2 Solanum asperolanatum Ruiz & Pav. Solanaceae 1 Stilpnophyllu oellgaardii L. Andersson Rubiaceae 1 Symplocos bogotensis Brand Symplocaceae 1 Symplocos fuscata B. Ståhl Symplocaceae 2 Tibouchina lepidota (Bonpl.) Baill. Melastomataceae 1 Unidentified trees Unidentified trees 6 Viburnum pichinchense Benth. Caprifoliaceae 9 Vismia tomentosa Ruiz & Pav. Clusiaceae 2 Weinmannia pubescens Kunth Cunoniaceae 1 Weinmannia spruceana Engl. Cunoniaceae 1 TOTAL 714

A total of 239 E. rhopalostele individuals were located in 25 of the 714 trees in the forest plot (Fig. 2.5). 151 of them were adult plants and 88 juveniles. Sizes of plants ranged from 1 to 45 cm (Fig. 2.6). Number of ramets per plant ranged from 1 to 29. 175 orchids were located on the trunk of the phorophyte while 64 were on the branches. 59 flowering plants were found with the number of flowers varing from 7 to 25 per inflorescence and a maximum of 115 flowers in one plant. E. rhopalostele plants were found in the first 3 m of height of the tree hosts. The phorophytes of E. rhopalostele belonged to seven different species and the groups assigned to lianas and dead trees (Table 2.2).

Chapter 2. RESULTS

Table 2.2. Category and frequency of E. rhopalostele phorophytes and number of E. rhopalostele plants present in each phorophyte species

Phorophyte Frequency Number of E. rhopalostele plants Clusia alata Planch. & Triana 7 64 Dead trees 11 118 Faramea anisocalyx Poepp. & Endl. 1 5 Geissanthus ecuadorensis Mez 1 6 Hedyosmum spectabile Todzia 1 18 Lianas 1 1 Miconia sp. 7 1 4 Myrsine andina (Mez) Pipoly 1 2 Psychotria tinctoria (Aubl.) Raeusch. 1 21

Association analyses of E. rhopalostele Epidendrum rhopalostele was marginally more likely to be found on shade- tolerant trees than expected (χ2 = 4.22, df = 1, p = 0.040) and clearly more likely to be found on dead trees than expected (χ2 = 28.4991, df = 1, p < 0.001). Among living trees, no preference for a particular phylogenetic clade of trees was found (χ2 = 1.12, df = 3, p = 0.777). However, the randomization test showed that there is a preference of E. rhopalostele for Clusia alata since the probability of getting the number of phorophytes of Clusia alata obtained by chance was < 0.0001. When testing for microhabitat preference (i.e. preference for particular host tree characteristics) a marginal association was shown for host trees with rough bark (χ2 = 3.89, df = 1, p = 0.048). The presence of moss in the trees was not significantly associated to the presence of E. rhopalostele (χ2 = 0.60, df = 1, p = 0.43). On the contrary, regarding the vertical distribution of orchids on the tree (measured as distance to the soil) a clear preference for the distance ranging from 0 to 1 meter was found (χ2 = 232.76, df = 2, p = < 0.001). Moreover, the positive association of juvenile orchids to the first distance class was greater than that found for the adults (χ2 = 10.152, df = 2, p = 0.006) (Fig. 2.7).

Chapter 2. Factors than condion the presence of E. rhopalostele

a) b) ) ) r r ( ( 2 2 1 2 K K ) r ) ( r 1 (

K 1 K -50 0 50 100 200 -100 0 100 200 300

0 2 4 6 8 0 2 4 6 8

Distance r (m) Distance r (m)

c) d) ) ) r r ( ( 2 1 2 G G ) r ) ( r ( 1 1 G G -0.6 -0.4 -0.2 0.0 -0.6 -0.4 -0.2 0.0

0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5

Distance r (m) Distance r (m)

Fig. 2.10. Spatial relationship between Epidendrum rhopalostele phorophytes (K1) and non- phorophytes (K2) in the studied forest fragment. a) Difference between univariate K function of phorophytes and non-phorophytes (K1 – K2); b) Difference between univariate K function of phorophytes and bivariate K function (K1-K12); c) Difference between univariate G functions (G1-G2); d) difference between univariate G function of phorophytes and bivariate G function (G1–G12).. Black continuous line represents the difference between functions; gray shaded area represents the confidence envelope of the Poisson cluster labeling model; dashed red line represents mean of the simulated Poisson cluster labeling process.

Spatial distribution of orchid population Spatial distribution of orchid population in this fragment is non-random. A strong aggregation pattern was present in E. rhopalostele at small scales, with a maximum value of L(r) at one meter of distance, which indicates the most frequent size of aggregates (Fig. 2.11a). The pattern of E. rhopalostele in the study fragment was described by a homogeneous Poisson cluster (HPC) process (GoF test: p-value > 0.05, Fig. 2.11b). The number of clusters (Aρ) for orchids in fragment plot was 9, and the number of clusters by m2 (i.e. ρ) was 0.013, with a mean number of orchids per cluster of 26.1 and a cluster size (i.e., rC ≈2σ, (Jacquemyn et al., 2007b) of 0.4 m (Table 2.3).

Chapter 2. DISCUSSION

distance seed dispersal in E. rhopalostele, as well as in phorophytes and trees in general. Seed dispersal limitation in E. rhopalostele is also backed by the lower nearest neighbor distance of phorophytes to phorophytes than to non- phorophytes. In the case of E. rhopalostele predominant seed dispersal may take place by water runoff through the branches. The spatial pattern is influenced by microhabitat preference of this orchid for certain types of trees (dead trees, Clusia alata, shade-tolerant trees) and certain characteristics like rough bark and location in the first meter from the soil in the vertical distribution over the host tree, which determines a preference to grow in areas with low light exposure and high humidity in the understory. The small-scale clustering of E. rhopalostele individuals located in each phorophyte also suggests a micro-site preference for orchid establishment. Juveniles were not more aggregated than adult plants but tended to be more surrounded by other juveniles than chance. On the contrary, adults did not tend to be more surrounded by juveniles than chance. Therefore, no evidence of competition, adult-juvenile facilitation or seed dispersal limitation at the micro-scale was found.

How are E. rhopalostele individuals arranged in space? A patchy distribution is evident in the E. rhopalostele population and a small- scale aggregation pattern is present. According to the fitted HPC model, individuals are distributed in 9 clusters in the fragment and a probability of 26 orchids per cluster. The epiphytic condition of orchid and the spatial distribution of potential host trees, which also fits a HPC model, limits the distribution of the orchid in the first instance. Additional factors such as microsite preference and short-distance seed dispersal further condition its presence as reflected by the lower cluster size of phorophytes versus trees in the fitted HPC models (2.1m vs. 5 m). The few existing studies of local spatial distribution in other orchids provide contrasting results. In Lepanthes, using nearest neighbor techniques, Tremblay (1997) found a regular dispersal pattern at distances of 5-10 m, but for large distances the pattern was aggregated, while in Broughtonia cubensis, using spatial point pattern analysis, an intraspecific spatial aggregation was observed (Raventós et al., 2011). The low intensity of colonization of E. rhopalostele in the study forest fragment, affecting only to 3.5% of the trees, could be related to strong environmental filtering or to the fact that we are now observing an early stage of the colonization process in a secondary forest, which is limited by seed availability and seed dispersal.

How are environmental conditions and microsite availability affecting E. rhopalostele distribution? Environmental conditions in this fragment located at the Andean depression in “Nudo de Loja” are highly diverse for abiotic and biotic factors. The montane forest of southern Ecuador has great diversity as a result of a broad matrix of environmental conditions (Homeier et al., 2008) with a climate gradient originated by differences in elevation (Bendix et al., 2008). Another important condition is the successional state of vegetation regeneration (i.e. secondary

48 Chapter 2. Factors than condion the presence of E. rhopalostele

montane cloud forest) with heterogeneity in species distribution located in patches and an open appearance with respect to primary forest (Martinez et al., 2008). Because the study forest fragment is variable in terms of elevation, orientation and slope we hypothesized that the trees of the forest fragment would be spatially aggregated as a result of this, concentrating in the most environmentally favorable sites. We expected that variations in the topography and associated factors in terms of wind exposure, nutrient availability, humidity, etc. could generate this pattern. However, the spatial distribution of trees in the fragment fitted a homogeneous Poisson cluster process. Although a homogeneous Poisson cluster process pattern may be partially generated by differential habitat suitability within the fragment, the fact is that it was not necessary to use heterogeneous models to explain the spatial distribution, which would more clearly evidence the limitation of the distribution by environmental conditions. Raventós (2011), who observed clustering in phorophytes and non-phorophytes of epiphytic orchids, suggests that this can be caused by small-scale habitat structure. Because E. rhopalostele was mainly found at the tree bases near the small ravines of the forest fragment, we suggest that high humidity could be a relevant environmental filter affecting the presence of this orchid species. In this sense, it is important to note that E. rhopalostele requires constant moisture because it lacks pseudobulbs and thick leaves (Riofrío et al., 2013). Because humidity is higher in the understory than in the canopy, this environmental difference might explain the fact that E. rhopalostele occurs only in the lower part of the trees. On the other hand, the small ravines could provide better conditions to E. rhopalostele; humidity in these areas is also greater, especially during the driest part of the year. Concerning microsite availability, previous studies about the factors affecting the epiphytic orchid distribution have focused on the identity of tree and its physical characteristics (e.g., size, age, architecture, bark roughness, cover of bark) (Bergstrom & Carter, 2008; Adhikari et al., 2012b). The small number of trees hosting E. rhopalostele in the forest and the high tree species richness made it statistically difficult to assess a preference for a particular species, as some authors have accomplished in other epiphytic orchid species (Tremblay et al., 1998; Gowland et al., 2011). However, our results show that E. rhopalostele is more likely to grow on dead trees than expected and, for living host trees, a preference is observed for Clusia alata. Epiphytic orchid species are relatively common on dead trees (Zimmerman, 1991; Ackerman et al., 1996; Gulledge et al., 2011), and preference for this type of substrate has been reported in some cases (Otero et al., 2007a; Cruz-Fernández et al., 2011). Mycorrhizal fungal like Tulasnella is mainly described as saprotrophic (Roberts, 1999), decomposing wood might offer better conditions for the growth and reproduction of these fungi. With a greater abundance of fungi, the probability of orchid seed infection in dead trees would also be greater. In this sense, it would be interesting to test whether the germination and establishment of E. rhopalostele can be enhanced by the presence of decomposing wood, as found in some terrestrial orchids (Rasmussen & Whigham, 1998). The rugosity of bark in host trees was another microsite characteristic positively affecting E. rhopalostele establishment. In the case of Clusia alata its characteristics, shade tolerance and distribution in the lower portion of the trunk may create favorable conditions for orchid

49 Chapter 2. DISCUSSION

establishment. A similar result was reported by Adhikari et al. (2012a) in 23 species of epiphytic orchids. E. rhopalostele was most frequent at the lower portion of the trunks (< 1 m height) and over shade tolerant trees suggesting an adaptation to low intensity of sunlight in the understory. In this sense, Adhikari et al. (2012a), in a study in an unmanaged forest observed that epiphytic orchids grew preferably in places with low light intensity. Furthermore, a study developed in coffee plantations showed that the orchid Oncidium polilostalix was also preferentially found in the trunk of shade tolerant trees (Inga micheliana) as well as in coffee branches (García-González et al., 2011). Therefore, our results support the previously posed hypothesis indicating that the presence of E. rhopalostele in certain trees depends on additional environmental factors, which could be related with the host traits (species, dbh, type of bark, etc) or the conditions offered by the host (presence of mycorrhiza, presence of moss, etc.).

Is the spatial distribution of the orchid determined by limited seed dispersal? Our analysis showed that the spatial pattern of E. rhopalostele is described by an homogeneous Poisson cluster process, that is, that the distribution of trees hosting E. rhopalostele is random within clusters and that a limited seed dispersal process may be taking place throughout the plot. Thus, in addition to the environmental factors limiting successful establishment of the orchid, the spatial distribution pattern of trees that had at least one adult plant of E. rhopalostele suggests limitation to seed dispersal at the scale of the clusters identified through the fitting to the Poisson cluster models, indicating that seed dispersal mostly occurs within trees. Spatial distribution of E. rhopalostele host trees seems to be conditioned by limited seed dispersal, with a low number of clusters of phorophytes in the study area and low number of trees per cluster with a cluster size of 2.1 m. The facts that, in the fitted HPC models, the cluster size of the phorophytes is notably lower than that of the forest fragment trees further supports this thesis. On the contrary, larger number of clusters and larger cluster sizes would indicate a great potential for long-distance seed transport (Seidler & Plotkin, 2006). It is also worthy to note in this sense that the distance to the nearest neighbor of phorophytes to phorophytes was lower than that of phorophytes to non-phorophytes. This is also in accordance to our hypothesis that stated that the colonization of new trees by E. rhopalostele inside the forest would be limited due to difficulties in seed dispersal. The limitation in seed dispersal may be originated by the dense vegetation matrix of the montane cloud forest that constrains seed dispersal by wind. Furthermore, it might be caused by seed dispersal preferentially mediated by rain water runoff through branches, instead of by wind.

What are some of the other factors that condition the presence of E. rhopalostele? We found no evidence supporting our hypothesis that stated that individuals of E. rhopalostele within phorophytes would be clustered as a result of a facilitation process of adults to juveniles. There was no pattern of adults being preferentially surrounded by juveniles that would be originated by facilitation. No segregation

50 Chapter 2. Factors than condion the presence of E. rhopalostele

patterns between adults and juveniles was found either, which would suggest competitive interactions by limited resources (i.e. suitable mycorrhizal fungi (Dearnaley, 2007; Otero et al., 2007b), pollinator (Ackerman et al., 1996; Tremblay & Ackerman, 2007; Waterman & Bidartondo, 2008) or nutrients (McCormick & Jacquemyn, 2014). In Orchis mascula, a terrestrial orchid, spatial segregation was present and a tight clustering of recruits around adults suggested a distance-dependent germination probably by mycorrhizal decline (Diez, 2007; Jacquemyn et al., 2009). In other epiphytic orchids in dry forest, a segregation between reproductive and non-reproductive individuals was observed, but with a small-scale aggregation in reproductive plants, suggesting a lack of competition by resources (Raventós et al., 2011). In juveniles, a significant trend to be surrounded by other juveniles than what is expected by chance is observed which would support the importance of availability of suitable micro-site during early stages of development of orchid. This aggregation of juveniles could change with time caused by a competition for resources with consequent death of some individuals resulting in a lower aggregation pattern for adult stages.

What are the implications for conservation? In the Andean depression, deforestation and anthropogenic impacts are the major threat for population persistence at the microscale (i.e., epiphytic micro-habitats), followed by global warming that will change flora structure (Richter et al., 2009). Information based on orchid population studies and the factors affecting presence at the local scale can generate strategies for the management and conservation of this epiphytic orchid. In this sense, it becomes essential to understand the spatial scale at which epiphyte orchid populations are viable in long term (Raventós et al., 2011). From a conservation perspective, the knowledge provided in this study may help detect recruitment sites and support management measures in secondary montane cloud forests, similar to the one where the study took place. The availability of a suitable site is determinant for orchid conservation (Tremblay et al., 2006). Thus, fallen and dead trees, Clusia alata trees and trees with rough bark are determinant for the regeneration of E. rhopalostele populations in secondary montane cloud forests. Concerning the vertical distribution of orchids over phorophytes, characteristics like low luminosity (shade tolerant trees) and high humidity should be considered in the establishment of these orchids. Management practices should favor these conditions to promote recruitment.

Bibliography

Ackerman JD, Sabat A, Zimmerman JK. 1996. Seedling establishment in an epiphytic orchid: an experimental study of seed limitation. Oecologia 106: 2, 192-198. Adhikari YP, Fischer A, Fischer HS. 2012a. Micro-site conditions of epiphytic orchids in a human impact gradient in Kathmandu valley, Nepal. Journal of Mountain Science 9: 3, 331-342.

51 Chapter 2. Bibliography

Adhikari YP, Fischer HS, Fischer A. 2012b. Host tree utilization by epiphytic orchids in different land-use intensities in Kathmandu Valley, Nepal. Plant Ecology 213: 9, 1393-1412. Agresti A. 2002. Categorical data analysis, 2nd edn. Wiley, Hoboken, New Jersey, USA. Antonovics J, Levin D. 1980. The ecological and genetic consequences of density- dependent regulation in plants. Annual Review of Ecology and Systematics 11: 411- 452. Arditti J, Ghani AK. 2000. Numerical and physical properties of orchid seeds and their biological implications. New Phytologist 145: 3, 367-421. Baddeley A, Turner R. 2005. Spatsat: An R package for analyzing spatial point patterns. Journal of Statistical Software 12: 1, 42. Beck E, Makeschin F, Haubrich F, Richter M, Bendix J, Valarezo C. 2008. The ecosystem (Reserva Biológica San Francisco). In: Beck E, Bendix J, Kottke I, Makenschin F, Mosandl R (eds.), Gradients in a tropical mountain ecosystem of Ecuador. Ecological studies. Springer-Verlag Berlin Heidelberg, Germany, pp. 1- 14. Belinchón R, Martínez I, Aragón G, Escudero A, De La Cruz M. 2011. Fine spatial pattern of an epiphytic lichen species is affected by habitat conditions in two forest types in the Iberian Mediterranean region. Fungal Biology 115: 12, 1270-1278. Bendix J, Rollenbeck R, Fabian P, Emck P. 2008. Climate. In: Beck E, Bendix J, Kottke I, Makeschin F, Mosandl R (eds.), Gradients in a Tropical Mountain Ecosystem of Ecuador. Springer-Verlag Berlin Heidelberg, Germany, pp. 281-290. Benzing D. 2012. Air Plants: Epiphytes and Aerial Gardens, 1st edn. Cornell University Press, Unites States of America. Bergstrom BJ, Carter RJ. 2008. Host-tree selection by an epiphytic orchid, Epidendrum magnoliae Muhl. (green fly orchid), in an inland hardwood hammock in Georgia. Southeastern Naturalist 7: 571-580. Bialozyt R, Ziegenhagen B, Petit R. 2006. Contrasting effects of long distance seed dispersal on genetic diversity during range expansion. Journal of Evolutionary Biology 19: 1, 12-20. Cousens R, Wiegand T, Taghizadeh M. 2008. Small-scale spatial structure within patterns of seed dispersal. Oecologia 158: 3, 437-448. Cruz-Fernández OT, Alquicira-Arteaga ML, Flores-Palacios A. 2011. Is orchid species richness and abundance related to the conservation status of oak forest? Plant Ecology 212: 1091-1099. De La Cruz M, Romao RL, Escudero A, Maestre FT. 2008. Where do seedlings go? A spatio-temporal analysis of seedling mortality in a semi-arid gypsophyte. Ecography 31: 6, 720-730. Dearnaley JD. 2007. Further advances in orchid mycorrhizal research. Mycorrhiza 17: 6, 475-486. Diez J. 2007. Hierarchical patterns of symbiotic orchid germination linked to adult proximity and environmental gradients. Journal of Ecology 95: 1, 159-170. Diggle PJ. 2003. Statistical analysis of point patterns, 2nd edn. Oxford University, Press Inc., Arnold, London. Dixon PM. 2002. Ripley's K Function. In: El-Shaarawi A, Piergorsch W (eds.), The encyclopedia of Environmetrics. Wiley, New York, New York, USA, pp. 1976- 1803. Dodson C. 1994-2003. Native Ecuadorian Orchids, 4th edn. Colina, Quito. Einzmann HJ, Beyschlag J, Hofhansl F, Wanek W, Zotz G. 2015. Host tree phenology affects vascular epiphytes at the physiological, demographic and community level. AoB Plants 7: 1-16.

52 Chapter 2. Factors than condion the presence of E. rhopalostele

Epperson BK. 2005. Estimating dispersal from short distance spatial autocorrelation. Heredity 95: 7-15. FAO. 2005. State of the world’s forests, 1st edn. Food and Agriculture Organization of the United Nations, Rome. Finegan B. 1992. Bases ecológicas para la silvicultura, Tema I. V. Curso internacional sobre silvicultura y manejo de bosques tropicales, Centro Agronómico Tropical de Investigación y Enseñanza, Turrialba, Costa Rica. Flores-Palacios A, Ortiz-Pulido R. 2005. Epiphyte orchid establishment on termite carton trails. Biotropica 37: 3, 457. Garay LA. 1978. Orchidaceae. Cypripedioideae, Orchidoideae, Neottioideae. In: Harling G, Sparre B (eds.), Flora of Ecuador. 9. University of Göteborg, Stockholm, Sweden, pp. 1-305. García-González A, Damon A, Esparza L, Valle-Mora J. 2011. Population structure of Oncidium poikilostalix (Orchidaceae), in coffee plantations in Soconusco Chiapas, México. Lankesteriana 11: 1, 23-32. Gentry AH, Dodson C. 1987. Diversity and biogeography of neotropical vascular epiphytes. Annals of the Missouri Botanical Garden 74: 205-233. Getzin S, Wiegand T, Wiegand K, He F. 2008. Heterogeneity influences spatial patterns and demographics in forest stands. Journal of Ecology 96: 4, 807-820. Gowland KM, Wood J, Clements MA, Nicotra AB. 2011. Significant phorophyte (substrate) bias is not explained by fitness benefits in three epiphytic orchid species. American Journal of Botany 98: 2, 197-206. Gradstein S. 2008. Epiphytes of tropical montane forests—impact of deforestation and climate change. Biodiversity and Ecology Series 2: 51-65. Gravendeel B, Smithson A, Slik F, Schuiteman A. 2004. Epiphytism and pollinator specialization: drivers for orchid diversity? Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 359: 1450, 1523-1535. Gulledge K, Johnson AF, Schultz G. 2011. Rare plant survey of Chassahowitzka wildlife management area, Herando County, Report to the Florida Fish and Wildlife Conservation Commission, Tallahassee, Florida, USA. Hágsater E, Dodson C, Sánchez L, Cervantes L, Dressler RL, Silverstone-Sopkin P. 2001. Icones Orchidacearum. The genus Epidendrum. A third century of new species in Epidendrum, fasc 4, part 3 edn. Herbario AMO, D.F. México, México. Hágsater E, Dumond V (eds.), 1996. Orchids: Status survey and conservation action plan. 1st edn. IUCN Gland, Switzerland and Cambridgw, IK, Cambridge, United Kingdom. Harling G, Andersson L. 1986-1998. Flora of Ecuador. 27-60, University of Götenborg; Riksmuseum, Göteborg; Stockholm. Harling G, Sparre B. 1980. Flora of Ecuador. 13, University of Götenborg, Göteborg. Healey P, Michaud J, Arditti J. 1980. Morphometry of orchid seeds. III. Native California and related species of Goodyera, Piperia, Platanthera and Spiranthes. American Journal of Botany 67: 508-518. Homeier J, Werner F, Gradstein S, Breckle S, Richter M. 2008. Potential Vegetation and Floristic Composition of Andean Forests in South Ecuador, with a Focus on the RBSF. In: Beck E, Bendix J, Kottke I, Makenschin F, Mosandl R (eds.), Gradients in a Tropical Mountain Ecosystem of Ecuador. Ecological Studies. Springer-Verlag Berlin Heidelberg, Germany, pp. 87-100. Illian J, Penttinen A, Stoyan H, Stoyan D. 2008. Statistical Analysis and Modelling of Spatial Point Patterns, 1st edn edn. John Wiley & Sons, Chichester. Jacquemyn H, Wiegand T, Vandepitte K, Brys R, Roldán-Ruiz I, Honnay O. 2009. Multigenerational analysis of spatial structure in the terrestrial, food‐deceptive orchid Orchis mascula . Journal of Ecology 97: 2, 206-216.

53 Chapter 2. Bibliography

Jacquemyn H, Brys R, Vandepitte K, Honnay O, Roldán-Ruiz I, Wiegand T. 2007. A spatially explicit analysis of seedling recruitment in the terrestrial orchid Orchis purpurea . New Phytologist 176: 2, 448-459. Jacquemyn H, Brys R, Hermy M, Willems JH. 2005. Does nectar reward affect rarity and extinction probabilities of orchid species? An assessment using historical records from Belgium and the Netherlands. Biological Conservation 121: 2, 257-263. Jørgensen P, León-Yánez S. 2001. Catalogue of the vascular plants of Ecuador. Monographs in Systematic Botany from the Missouri Botanical Garden 75: i–viii, 1-1182. Jørgensen P, Ulloa C. 1994. Seed plants of the high Andes of Ecuador - A checklist. AAU Rep 34: 1-443. Kromer T, Kessler M, Robbert G, Acebey A. 2005. Diversity patterns of vascular epiphytes along an elevational gradient in the Andes. Journal of Biogeography 32: 10, 1799-1809. Laube S, Zotz G. 2003. Which abiotic factors limit vegetative growth in a vascular epiphyte? Functional Ecology 17: 5, 598-604. Loosmore NB, Ford ED. 2006. Statistical inference using the g or K point pattern spatial statistics. Ecology 87: 8, 1925-1931. Martinez A, Mahecha MD, Lischeid G, Beck E. 2008. Succession Stages of Vegetation Regeneration: Secondary Tropical Mountain Forests. In: Beck E, Bendix J, Kottke I, Makeschin F, Mosandl R (eds.), Gradients in a tropical mountain ecosystem of Ecuador. Ecological studies. Springer-Verlag Berlin Heidelberg, Germany, pp. 409- 415. McCormick MK, Jacquemyn H. 2014. What constrains the distribution of orchid populations? New Phytologist 202: 2, 392-400. Migenis L, Ackerman JD. 1993. Orchids-phorophyte relationships in a forest watershed in Puerto Rico. Journal of Tropical Ecology 9: 2, 231-240. Mosandl R, Gunter S, Stimm B, Weber M. 2008. Ecuador suffers the highest deforestation rate in south America. In: Beck E, Bendix J, Kottke I, Makeschin F, Mosandl R (eds.), Gradients in a tropical mountain ecosystem of Ecuador. Springer-Verlag Berlin Heidelberg, Germany, pp. 37-40. Nathan R, Katul GG, Horn HS, Thomas SM, Oren R, Avissar R, Pacala SW, Levin SA. 2002. Mechanisms of long-distance dispersal of seeds by wind. Nature 418: 6896, 409-413. Nathan R, Muller-Landau HC. 2000. Spatial patterns of seed dispersal, their determinants and consequences for recruitment. Tree 15: 7, 278-285. Neill DA, Ulloa C. 2011. Adiciones a la Flora del Ecuador: Segundo Suplemento, 2005- 2010
, Fundación Jatún Sacha, Quito. Nishida S. 1999. Revision of Beilschmiedia (Lauraceae) in the Neotropics. Annals of the Missouri Botanical Garden 86: 3, 657-701. Nöske N, Hilt N, Werner F, Brehm G, Fiedler K, Sipman H, Gradstein S. 2008. Disturbance effects on diversity of epiphytes and moths in a montane forest in Ecuador. Basic and Applied Ecology 9: 1, 4-12. Otero J, Aragón S, Ackerman JD. 2007a. Site variation in spatial aggregation and phorophyte preference in Psychilis monensis (Orchidaceae). Biotropica 39: 2, 227- 231. Otero J, Flanagan N, Herre E, Ackerman JD, Bayman P. 2007b. Widespread mycorrhizal specificity correlates to mycorrhizal function in the neotropical, epiphytic orchid Ionopsis utricularioides (Orchidaceae). American Journal of Botany 94: 12, 1944- 1950.

54 Chapter 2. Factors than condion the presence of E. rhopalostele

Plotkin JB, Potts MD, Leslie N, Manokaran N, Lafrankie J, Ashton PS. 2000. Species- area curves, spatial aggregation, and habitat specialization in tropical forests. Journal of Theoretical Biology 207: 1, 81-99. Poorter L, Bongers L, Bongers F. 2006. Architecture of 54 moist-forest tree species: Traits, trade-offs, and functional groups. Ecology 87: 1289-1013. R Development Core Team. 2011. R: A language and environment for statistical computing, R Foundation for Statistical Computing, Vienna, Australia. Rasmussen HN, Whigham DF. 1998. Importance of woody debris in seed germination of Tipularia discolor (Orchidaceae). American Journal of Botany 85: 6, 829-834. Raventós J, Mujica E, Wiegand T, Bonet A. 2011. Analyzing the spatial structure of Broughtonia cubensis (Orchidaceae) populations in the dry forests of Guanahacabibes, Cuba. Biotropica 43: 2, 173-182. Renner S, Balslev H, Holm-Nielsen L. 1990. Flowering plants of Amazonian Ecuador— A checklist. . AAU Rep 4: 1-241. Richter M, Diertl K, Emck P, Peters T, Beck E. 2009. Reasons for an outstanding plant diversity in the tropical Andes of Southern Ecuador. Landscape Online 12: 1-35. Riofrío ML, Cruz D, Torres E, De la Cruz M, Iriondo JM, Suárez JP. 2013. Mycorrhizal preferences and fine spatial structure of the epiphytic orchid Epidendrum rhopalostele. American Journal of Botany 100: 12, 2339-2348. Ripley BD. 1976. The second order analysis of stationary point processes. Journal of Applied Probability 13: 255-266. Robbert G, Jürgen H, Dirk G (eds.), 2008. The Tropical Mountain Forest. Patterns and Processes in a Biodiversity Hotspot. Biodiversity and Ecology Series: Volume 2 edn. Göttingen Centre for Biodiversity and Ecology, Georg-August-Universität Göttingen. Roberts P. 1999. Rhizoctonia-forming Fungi: a taxonomic guide. 1st edn. Royal Botanical Garden, Kew, Kew. Schurr F, Steinitz O, Nathan R. 2008. Plant fecundity and seed dispersal in spatially heterogeneous environments: models, mechanisms and estimation. Journal of Ecology 96: 4, 628-641. Seidler TG, Plotkin JB. 2006. Seed Dispersal and Spatial Pattern in Tropical Trees. PLoS Biol 4: 11, e344. Shimamura R, Kachi N, Kudoh H, Whigham DF. 2007. Hydrochory as a determinant of genetic distribution of seeds within Hibiscus moscheutos (Malvaceae) population. American Journal of Botany 94: 7, 1137-1145. Silander JJ, Pacala SW. 1985. Neighborhood predictors of plant performance. Oecologia 66: 2, 256-263. Tremblay RL, Ackerman JD. 2007. Floral color patterns in a tropical orchid: Are they associated with reproductive success? Plant Species Biology 22: 2, 95-105. Tremblay RL, Meléndez-Ackerman E, Kapan D. 2006. Do epiphytic orchids behave as metapopulations? Evidence from colonization, extinction rates and asynchronous population dynamics. Biological Conservation 129: 1, 70-81. Tremblay RL, Zimmerman JK, Lebrón L, Bayman P, Sastre I, Axelrod F, Alers-García J. 1998. Host specificity and low reproductive success in the rare endemic Puerto Rican orchid Lepanthes caritensis. Biological Conservation 85: 3, 297-304. Tremblay RL. 1997. Distribution and dispersion patterns of individuals in nine species of Lepanthes (Orchidaceae). Biotropica 29: 1, 38-45. Vittoz P, Engler R. 2007. Seed dispersal distances: a typology based on dispersal modes & plant traits. Botanica Helvetica 117: 2, 109-124. Waterman RJ, Bidartondo MI. 2008. Deception above, deception below: linking pollination and mycorrhizal biology of orchids. Journal of Experimental Botany 59: 5, 1085-1096.

55 Chapter 2. Bibliography

Werner F, Gradstein S. 2008. Seedling establishment of vascular epiphytes on isolated and enclosed forest trees in an Andean landscape, Ecuador. Biodiversity and Conservation 17: 13, 3195-3207. Wiegand T, Moloney KA. 2014. Handbook of spatial point pattern analysis in Ecology, 1st edn. Chapman and Hall, . Wiegand T, Martínez I, Huth A. 2009. Recruitment in tropical tree species: Revealing complex spatial patterns. The American Naturalist 174: 4, E106-E140. Wiegand T, Gunatilleke S, Gunatilleke N. 2007a. Species associations in a heterogeneous Sri Lankan Dipterocarp forest. . The American Naturalist 170: E77- E95. Wiegand T, Gunatilleke S, Gunatilleke N, Okuda T. 2007b. Analyzing the spatial structure of a Sri Lankan tree species with multiple scales of clustering. Ecology 88: 12, 3088-3102. Wiegand T, Moloney KA. 2004. Rings, circles, and null-models for point pattern analysis in ecology. Oikos 104: 2, 209-229. Winkler M, Hietz P. 2001. Population structure of three epiphytic orchids (Lycaste aromatica, Jacquiniella leucomelana, and J. teretifolia) in a mexican humid montane forest. Selbyana 22: 1, 27-33. Wright J, Trakhtenbrot A, Bohrer G, Detto M, Katul G, Horvitz N, Muller-Landaua H, Jones F, Nathan R. 2008. Understanding strategies for seed dispersal by wind under contrasting atmospheric conditions. Proceedings of the National Academy of Sciences 105: 49, 19084-19089. Yoder JA, Imfeld S, Heydinger D, Hart C, Collier M, Gribbins K, Zettler LW. 2010. Comparative water balance profiles of Orchidaceae seeds for epiphytic and terrestrial taxa endemic to North America. Plant Ecology 211: 1, 7-17. Zimmerman JK. 1991. Ecological Correlates of Labile Sex Expression in the Orchid Catasetum Viridiflavum. Ecology 72: 597-608. Zotz G, Mendieta-Leiva G, Wagner K. 2014. Vascular epiphytes at the treeline – composition of species assemblages and population biology. Flora - Morphology, Distribution, Functional Ecology of Plants 209: 8, 385-390. Zotz G, Winkler U. 2013. Aerial roots of epiphytic orchids: the velamen radicum and its role in water and nutrient uptake. Oecologia 171: 3, 733-741. Zotz G, Bader MY. 2009. Epiphytic Plants in a Changing World-Global: Change Effects on Vascular and Non-Vascular Epiphytes. In: Lüttge U, Beyschlag W, Büdel B, Francis D (eds.), Springer Berlin Heidelberg, pp. 147-170. Zotz G, Hietz P. 2001. The physiological ecology of vascular epiphytes: current knowledge, open questions. Journal of Experimental Botany 52: 364, 2067-2078.

Chapter 3

3. Fine-scale genetic structure of Epidendrum rhopalostele

INTRODUCTION

Seed dispersal is a key process for plant population dynamics, as affect colonization of new sites and determine plant metapopulation structure (Ouborg et al., 1999). Seed dispersal patterns are determined by the spatial patterns of adults and the seed rain, while seedlings distribution is also conditioned by the availability of suitable microsite (Nathan & Muller-Landau, 2000; Jersáková & Malinová, 2007). Therefore, the successful establishment of plants is given by the seed dispersal pattern that determines their spatial distribution, seed production and availability of resources in the microsite (Williams & Guries, 1994; Jacquemyn et al., 2007b; Winkler et al., 2009). Because seed dispersal and recruitment of new individuals are essential factors in the generation of the spatial genetic structure (SGS) of populations (Jacquemyn et al., 2007b), the study of the spatial genetic structure of populations can provide relevant information on the seed dispersal patterns and suitable microsite availability that operate in a particular species and population.

57 Chapter 3. INTRODUCTION

Spatial genetic structure is the non-random spatial distribution of genotypes and is influenced by multiple factors such as historical events, selective pressure, pollen dispersal, seed dispersal, mating system, availability of sites for establishment among others (Epperson, 1993; Escudero et al., 2002; Vekemans & Hardy, 2004; Jacquemyn et al., 2009). Genetic diversity in a spatial context, that is Spatial Genetic Structure (SGS), contributes to interpretation of mechanisms maintaining population dynamics and provides valuable tools to understand genetic events like selective pressures, gene flow and genetic drift (Escudero et al., 2002). This information is also essential for the identification of conservation and management units (MUs) (Allendorf & Luikart, 2007; Colling et al., 2010). Spatial genetic structure can be a consequence of a limited dispersal of seeds, which can locate genetically related seedlings close to each other (Hamrick et al., 1993; Epperson, 2000). In this context, genetic similarity is greater among neighbors and isolation by distance theory predicts the expected pattern of SGS at drift-dispersal equilibrium (Vekemans & Hardy, 2004). In orchids, in addition to the ecological factors related to fruit and seed dispersal (McCormick & Jacquemyn, 2014), dust-like and tiny seeds with an undifferentiated embryo with minimal reserves and its dependence on mycorrhizal fungus are factors that condition seed germination and recruitment of seedlings (Rasmussen & Rasmussen, 2007; Dearnaley, 2007; Jersáková & Malinová, 2007); these particular features of orchid seed facilitating long-distance dispersal mechanisms where wind is the main dispersal vector (Arditti & Ghani, 2000; Winkler et al., 2009; Bullock et al., 2012; McCormick & Jacquemyn, 2014). In orchids, genetic studies have shown both high and low levels of genetic diversity (Chung et al., 2005a; Chung et al., 2005b; Jacquemyn et al., 2006; Ávila-Díaz & Oyama, 2007). Most of them have focused on the analysis of terrestrial orchids (Forrest et al., 2004; Jacquemyn et al., 2006; Jacquemyn et al., 2007b), and have found, in some cases, significant patterns of SGS (Machon et al., 2003). The presence of SGS has been explained by limited seed (Jersáková & Malinová, 2007), having obtained estimates of seed dispersal range from 4 to 7 meters in Orchis purpurea (Jacquemyn et al., 2007b). Despite their ecological importance and diversity, seed dispersal studies in epiphytic orchids are scarce (Kartzinel et al., 2013). The number of genetic studies is possibly limited by difficulties of accessibility to the forests where they occur (Riofrío et al., 2007). In any case, some studies already performed in epiphytic orchids indicate high levels of genetic diversity and low population structure that could be due to the adaptability achieved in response to the constant changes in their habitat (Ávila- Díaz & Oyama, 2007). To evaluate SGS it is necessary to analyze genetic population structure and spatial distribution (Escudero et al., 2002). To study genetic population structure DNA-based molecular markers are used. AFLPs (Vos et al., 1995) are a suitable alternative in genetic ecology and conservation analysis (Vekemans et al., 2002) due to the high level of polymorphic loci they contain and because not prior knowledge of the genome is required. This is an important point to consider because epiphytic orchids have normally no previous genetic studies.

58 Chapter 3. Fine scale genetic structure of E. rhopalostele

In this chapter we assessed the small-scale spatial genetic structure of E. rhopalostele Hágsater & Dodson, an epiphytic perennial orchid to infer dominant seed dispersal mechanisms (short or long distance), considering the following assumptions: 1) If E. rhopalostele seeds are essentially dispersed at a short distance, near to the mother plant by gravity or by water run-off a small-scale spatial-genetic structure would be expected where individuals would be more genetically related at short distances. 2) If E. rhopalostele seeds are mostly dispersed by wind, a long-distance seed rain would be expected and the genetic diversity of individuals would be randomly distributed in space. The aim of this work was to infer the predominant seed dispersal mechanism in this species through the assessment of the fine-scale genetic-spatial structure of E. rhopalostele in a montane cloud forest fragment using AFLPs. The main questions posed in this work are: 1) What is the predominant seed dispersal mechanism and the seed dispersal range (short or long distance) in this orchid? 2) Is there a Spatial Genetic Structure in this species than can help explain the seed dispersal and seedling recruitment processes?.

MATERIAL AND METHODS

Study species Epidendrum rhopalostele is a photosynthetic epiphytic perennial orchid native of Ecuador, Peru and Bolivia that usually occurs in light gaps or along the forest edge in the evergreen montane forest. It belongs to the alpicolum group and it is very similar to Epidendrum dialychilum Hágsater & Dodson, both having the lip free from the column, but differing from the latter by its linear-lanceolate and acuminate lip, the long filiform acuminate petals and the stigmatic cavity only in the apical third of the column (Hágsater et al., 2001). Epidendrum rhopalostele plants are 20 – 40 cm long and have an inflorescence with 10-30 bright green flowers. Flowering takes place between January – March and June - August and capsules ripen between the end of August and the beginning of November (personal observation). Seeds are small (250 ± 350 µm long and 40 ±55 µm wide). Its reproductive biology and seed dispersal mechanism have not been studied and are unknown.

Study site and mapping The study site (~1 ha) was located on the eastern slope of Cordillera Real in the Andes of southern Ecuador, at the border of Podocarpus National Park and near the Loja-Zamora road in Zamora-Chinchipe province. It is an approximately 35- year-old regenerated forest classified as evergreen upper montane forest (Beck et al., 2008a). The most diverse and abundant seed plant families in this type of forest are Orchidaceae, Melastomataceae, Ericaceae and Rubiaceae (Homeier et al., 2008). Annual precipitation is 2193 mm, and annual mean temperature is

59 Chapter 3. MATERIAL AND METHODS

20.8°C (4.7-25.5°C) (Bendix et al., 2008). All trees (dead and alive) with a diameter at breast height (DBH) > 1 cm was georeferenced. Their x, y and z coordinates were obtained using a total station theodolite (Zeiss Elta 6). Twelve base points were established to cover the whole study area. The coordinates of the trees were taken by referencing the base of the tree trunk. Each of the mapped trees was examined for the presence of orchids. All the individuals (seedlings, juveniles and adults) of E. rhopalostele present at the time of the study in 2008 were counted and marked, including individuals located as close as 1 cm from each other. Their x and y coordinates were obtained using a total station theodolite (Zeiss Elta 6), as done for the trees with the difference that to determine the coordinate point of each orchid a plumb line was laid down to the ground and this was the site read by the total station.

Sampling for genetic analyses and DNA extraction We collected young leaf samples of all individuals (adults, juveniles and seedlings) of E. rhopalostele present in phorophytes for DNA analysis. Leaf material was dried in silica gel and stored at room temperature until DNA extraction. Total DNA was extracted from 30 mg of dried leaf material using PureLink Plant Total DNA Purification Kit (Invitrogen, Carlsbad, California, USA). After extraction, DNA, concentration was estimated using a fluorospectrometer (NanoDrop 3300, Thermo Scientific, Wilmington, Delaware, USA). DNA samples were stored at -20 ºC until use.

AFLP procedure AFLP reactions were performed following the procedure of Vos et al. (Vos et al., 1995) with minor modifications. Genomic DNA (approximately 500 ng) was restricted with 0.1 units of MseI (New England BioLabs, Ipswich, Massachussetts, USA) and 0.5 units of EcoRI (Takara Bio Inc., Otsu, Japan) endonucleases, and ligated to MseI and EcoRI adapters with 6 units of T4 DNA- ligase (Takara Bio Inc.). Samples were incubated in a thermocycler for 3 h at 37º C and 1 h at 17º C. Preselective amplification was performed using the primers that complement the MseI and EcoRI adaptors plus one additional nucleotide i.e. MseI+C (5ʹ GAT GAG TCC TGA GTA AC 3') and EcoRI+A (5ʹ-GAC TGC GTA CCA ATT CA- 3ʹ). PCR reactions were conducted in 12.5 µl reaction volume containing 2.5 µl of 10-fold diluted restriction-ligation product, 1X buffer (GeneAmp 10X PCR Buffer II, Applied Biosystems), 1.5 mM MgCl2, 0.2 mM dNTPs, 0.2 µM of each primer, and 0.5 units of Taq polymerase (AmpliTaq DNA Polymerase, Applied Biosystems). The thermocycler program used for amplifications was as follows: 72 ºC for 2 min first, then 30 cycles at 94 ºC for 30 s, 56 ºC for 30 s, and 72 ºC for 2 min, with a final extension at 72 ºC for 10 min. The quality of undiluted

60 Chapter 3. Fine scale genetic structure of E. rhopalostele

preselective and restricted/ligation products were visualized on 1% (w/v) agarose gels. Selective amplifications were conducted in a reaction volume of 12.5 µl containing 2.5 µl of 10-fold diluted preselective product, 1X buffer (GeneAmp 10X PCR Buffer II, Applied Biosystems), 1.5 mM MgCl2, 0.8 mM dNTPs, 0.08 µM of EcoRI fluorescent primer (Applied Biosystems), 0.2 µM of MseI primer (Bonsai Advanced Technologies, Alcobendas, Madrid, Spain), and 0.5 U of Taq polymerase (AmpliTaq Gold DNA Polymerase, Applied Biosystems). PCR conditions were: 95 ºC for 2 min first, then 13 cycles at 94 ºC for 30 s, 65 ºC for 1 min (with a decrease gradient of 0.7 ºC every cycle), and 72 ºC for 2 min; and 24 cycles at 94 ºC for 30 s, 56 ºC for 1 min, and 72 ºC for 2 min, with a final extension at 72 ºC for 10 min. Four primer combinations producing clear bands were used for selective amplification: MseI-CTA/EcoRI-AGA(6FAM), MseI-CAC/EcoRI-AGA(6FAM), MseI-CAA/EcoRI-AAC(VIC), and MseI-CAT/EcoRI-AGG(VIC). The selection was based on a preliminary screening of 32 primer combinations in five individuals (three individuals from the different trees and two from the same tree). AFLP fragments were separated on an ABI3730 sequencer (Applied Biosystems). A gene scan 500 Liz-labelled sizes standard (Applied Biosystems) was injected with each AFLP sample to allow sizing of the DNA fragments. To assess reproducibility of the protocol, two independent DNA extractions were carried out for eight samples. The overall error rate-that is, the percentage of differently scored loci between repeated samples-was low (2%).

Data analysis AFLPs profiles were imported into GeneMarker v. 4.1. (Softgenetics) and scored manually. All bands between 80 and 500 bp were scored as present (1) or absent (0) excluding those that could not be decisively assigned. Monomorphic and polymorphic fragments were included in analysis.

Detection of genetic structure As a preliminary step a principal coordinates analysis (PCoA) was implemented to provide a visual representation of the genetic distance relationships among all sampled individuals of E. rhopalostele. Following the suggestions of Bonin et al. (2007), simple matching coefficient was used to calculate the distance matrix. Next, population structure was inferred by applying model-based clustering methods. Unlike distance-based clustering, these methods use a model to identify genetic groups from the full set of genotype data and to assign probabilistically individuals (or a fraction of their genomes) to that group representing the best fit for the variation patterns found. We used the approach implemented in Structure v. 2.3.4, which involves Bayesian inference and parameter estimation through

61 Chapter 3. MATERIAL AND METHODS

Markov Chain Monte Carlo (MCMC) in the modeling process (Pritchard et al., 2000; Falush et al., 2007). For analysis with Structure, we used the admixture model (which assumes individuals may have mixed ancestry) with correlated allele frequencies (Falush et al., 2003), and set the number of groups (K) from 1 to 10. For each value of K, 10 independent runs were carried out with a burn-in period of 10,000 followed by 300,000 MCMC iterations. The optimal number of groups was determined estimating the log-likelihood of the data for each K [ln(P(XǀK)] as suggested by Pritchard et al. (2000) and the ∆K statistic proposed by Evanno et al. (2005), which is based on the rate of change in the log-likelihood of data between successive K values. Afterwards, the software CLUMPP v. 1.1.2 (Jakobsson & Rosenberg, 2007) was used, employing the FullSearch algorithm, to find the optimal alignment of the 10 independent replicate cluster analyses, and to compute the mean membership coefficient matrix (Q-matrix).

Detection of spatial genetic structure The spatial genetic structure was assessed using the spatial autocorrelation method proposed by Smouse & Peakall (1999), which allows to simultaneously assess the signal generated by multiple loci. Briefly, this approach calculates an autocorrelation coefficient (r) between genetic and geographic distances for all pairs of individuals within user specified distance classes. Under isolation by distance, geographically close individuals are expected to be more genetically similar between them than to other individuals occurring at greater distance. Therefore, r > 0 values are expected for short-distance lags and r < 0 for long- distances lags. Genetic distances between individuals were calculated using the method of Huff et al. (1993). Distance classes were defined at 1-m intervals, and each one of them included at least 30 pairs of points (Waser & Mitchell, 1990). We followed a classical rule of thumb and only considered pairs of points separated by less than half the maximum distance observed (Le Corre et al., 1998), in our case 15 m. Each value of r was tested for significant deviations from the expected value under the null hypothesis of no spatial genetic structure by a permutation test in which the whole AFLP phenotypes were randomly shuffled among the spatial positions occupied and r-values recalculated each time (up to a maximum of 999 times) [see Smouse & Peakall (1999) and Smouse et al. (2008) for details]. Because we are interested in the detection of positive autocorrelation at the short distance classes, one-tailed probability values were calculated. The significance of r was also tested by generating bootstrap 95% confidence intervals around the mean value of r. Bootstrap values were obtained by sampling, with replacement, pairs of comparisons within a given distance class (Peakall et al., 2003). Bootstrap resampling was performed 999 times and the significance of r inferred when the 95% confidence interval did not contain the zero value. Significance of the entire correlogram was tested using the heterogeneity test proposed by Smouse et al. (2008). Because the previous clustering analyses identified two genetic groups, the spatial autocorrelation analysis was performed for all individuals and also separately for 62 Chapter 3. Fine scale genetic structure of E. rhopalostele

each group. Admixture proportions (Q) provided by Structure were used to assign each individual to one of the two groups. The applied threshold value was Q > 0.5. The patterns of autocorrelation generated for each group were compared (at the distance class and whole correlogram levels) using the heterogeneity tests proposed by Smouse et al. (2008). All analyses were implemented in GenAlEx v.6.5 (Peakall & Smouse, 2006; Peakall & Smouse, 2012).

RESULTS

714 trees were found in the studied forest fragment, but E. rhopalostele was found just on 25 (Fig. 3.1a). The number of orchids per phorophyte ranged from 1 to 32, with a total of 239 individuals (Fig. 3.1b). AFLP analysis was successful in 216 of 239 individuals, and provided a total of 621 bands between 80 and 500 bp. 100% of the studied loci were polymorphic.

Population genetic structure The scattered plot of the first and the second axis of the PCoA showed that the first axis separated the 216 orchids in two distinct groups (Fig. 3.2). Visualization of results considering the phorophytes showed that no relationship between the two genetic groups of orchids and the phorophytes on which the orchids are located. No correspondence was also found when the type of individual (adult or juvenile) was taken into account. The first two principal coordinates accounted for 22% of total variance (18% for PCoA1 and 4% for PCoA2). The Bayesian clustering analysis conducted by Structure also identified two genetic groups (hereafter called group 1 and group 2). As showed in Fig. 3.3., the first value of lnP(X) at which the curve reaches the plateau (-57513.6) and the highest ΔK value (714.6) were found for K = 2, indicating that the genetic structure produced by dividing all genotypes into two groups is the most likely. For K = 2, most individuals (152 of 216) had admixture coefficients over 0.8 (Fig. 3.4). It is also remarkable that 21 of 25 phorophytes had individuals of both groups (Table 3.1, Fig. 3.5).

63 Chapter 3. RESULTS

Fig. 3.1. A) Spatial distribution of the 714 trees with dbh > 1 cm located in the study area. Red and grey crosses represent Epidendrum rhopalostele phorophytes non-phorophytes trees respectively. B) Spatial distribution of the 216 Epidendrum rhopalostele individuals located in the study area and their corresponding phorophytes (red crosses). Green and black circles correspond to the two genetic clusters detected by Structure (group 1 and group 2 respectively).

Chapter 3. Fine scale genetic structure of E. rhopalostele

Table 3.1. Number of individuals of Epidendrum rhopalostele per phorophyte belonging to group 1 and 2 according to the admixture model with correlated alleles implemented in Structure. Assigment to group is based on the admixture coefficient usisng a threshold value of 0.5

ID Phorophyte Group 1 Group 2 Total Tree 1 11 18 29 Tree 2 1 2 3 Tree 3 1 2 3 Tree 4 6 11 17 Tree 5 9 11 20 Tree 6 2 5 7 Tree 7 3 5 8 Tree 8 2 2 4 Tree 9 1 1 2 Tree 10 1 0 1 Tree 11 10 9 19 Tree 12 1 5 6 Tree 13 4 6 10 Tree 14 3 6 9 Tree 15 2 2 4 Tree 16 1 3 4 Tree 17 12 20 32 Tree 18 1 3 4 Tree 19 0 3 3 Tree 20 6 4 10 Tree 21 2 6 8 Tree 22 3 4 7 Tree 23 0 1 1 Tree 24 0 2 2 Tree 25 2 1 3 Total 84 132 216

Spatial autocorrelation When all individuals were analyzed together no genetic spatial autocorrelation was observed (data not shown). However, when the data set was divided according to the genetic groups assigned by Structure, significant spatial autocorrelograms were obtained (ω1 = 148.5, p-value = 0.001; ω2 = 131.8, p- value = 0.001) evidencing the presence of fine-scale genetic structure within clusters. For group 1 (Fig. 3.6A) a positive genetic correlation was found among individuals belonging to the first three distance classes (0-3m) and to the eighth and ninth distance classes (7-9m). On the contrary, a negative genetic correlation was found among individuals belonging to the fifth distance class (4-5m) and from the eleventh to the thirteenth distance classes (10-13m). Similarly, for group 2 (Fig. 3.6B), a positive genetic correlation was found among individuals belonging to the first (0-1m) and third (2-3 m) distance classes and to the eighth and ninth distance classes (7-9m). By contrast, a negative genetic correlation was found from the eleventh to the fourteen distance classes (10-14m). Significant differences were detected between both autocorrelograms (ω = 55.5, p-value = 0.001). In particular for the distance classes 1 (ω=19.0, p-value = 0.001), 2 (ω=6.9, p-value = 0.009), 5 (ω=9.6, p-value = 0.003), 12 (ω=30.6, p-value = 0.001) and 15 (ω=7.9, p-value = 0.001).

67 Chapter 3. DISCUSSION

A) 0.10

0.05

r 0.00

-0.05

-0.10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Distance class

B) 0.10 0.05

r 0.00 -0.05 -0.10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Distance class

Fig. 3.6. Autocorrelograms for group 1 (A) and group 2 (B) showing the genetic correlation coefficient (r) as a function of geographical distance. Distance classes range from 1 to 15 m with 1m intervals. Dashed lines represent upper and lower 95% confidence limits, whereas the solid line represent r (spatial autocorrelation coefficient) DISCUSSION

The present study is among the first to analyze the fine spatial genetic structure of an epiphytic orchid population occurring in the cloud forest. In this population with 239 E. rhopalostele individuals located in only 25 of 714 possible phorophytes, two different genetic groups were found. The finding of these two groups suggests contributions of genes from other Epidendrum species that could be hybridizing and that coexist in this forest fragment, despite the absence of an evident morphological differentiation between analyzed individuals as well as some gene flux limitation between the two groups. Spatial distribution of the individuals belonging to these two genetic groups was independent of their distribution in phorophytes, suggesting that phorophyte type is not a factor in explaining their distribution. Considering the nonexistent spatial genetic structure when all individuals of the forest fragment are taken into account, we could imply that seed dispersal by wind is effective at this spatial scale and that as a result genetic diversity is randomly distributed. However, the presence of a significant spatial genetic structure compatible with an isolation-by-distance pattern found in the two genetic groups, when they are analyzed separately, could be explained by an alternative seed dispersal mechanism dependent of water run-off that would operate at a short-distance.

68 Chapter 3. Fine scale genetic structure of E. rhopalostele

The fine scale genetic analysis of the E. rhopalostele individuals studied in the montane forest fragment showed a structure in two genetic groups. As it can be observed in, group 1 and 2 did not have a clear spatial delimitation and individuals belonging to both groups were found in all phorophytes. Thus, it is remarkable that the distribution of orchid individuals in phorophytes was independent of its assignment to one genetic group or another, that is, in the same phorophyte it was possible to find orchids that were assigned to group 1 and 2 with varying ranges of membership coefficient of each individual (Fig. 3.4). The contribution of each genetic group to the phorophytes was quite even, although for certain phorophytes (e.g., 19, 23 and 24) group1 was predominant (Table 3.1). This result might be explained by the existence of a group that would represent “pure” individuals of E. rhopalostele and a second group generated from contributions from individuals of other Epidendrum species (E. madsenii, E. falcisepalum) that could be hybridizing E. rhopalostele (Marques et al., 2014) and would be changing population genetic structure. Hybridization events are quite common in some orchid genera such as Epidendrum (Pinheiro & Cozzolino, 2013; Pinheiro et al., 2013) contributing to genetic variation and may be a relevant evolutionary force that is acting in this species (Vega et al., 2013). In order to maintain such genetic structure there must be some isolation between the two clusters. This isolation may take place due to phenological differences in the flowering time (De hert et al., 2012) in such way that only the genotypes of E. rhopalostele that overlap their flowering time with the other Epidendrum species would be able to hybridize (Pinheiro et al., 2010). Another explanation might be the existence of subtle morphological changes or chemical differences in the hybridized group (Devey et al., 2007; Peakall & Whitehead, 2014; Menz et al., 2015) that may attract pollinators that are different from those that normally pollinate the “pure” E. rhopalostele individuals.

Can spatial genetic structure explain the dispersal mechanisms that are taking place in E. rhopalostele? Although the lack of spatial genetic structure taking all individuals as a whole could suggest that the species has a long-distance seed dispersal mechanism, the existence of two genetic clusters that taken separately show a significant pattern of spatial genetic structure, with individuals being more genetically similar at short distances and more dissimilar at longer distances, suggests otherwise. Taking into consideration that this species is a deceptive orchid we can assume that pollinators are not specialists and that are likely to move considerable distances (Hágsater & Soto-Arenas, 2005; Vega et al., 2013; Marques et al., 2014). Therefore, the existence of spatial genetic structure in the two groups, with an isolation-by-distance pattern can only be explained by a short-distance seed dispersal profile. These results would imply that seed dispersal events take place at short distance within genetic groups. Regardless of occasional wind dispersal at longer distances, as described in other orchids (Arditti & Ghani, 2000; Rasmussen & Rasmussen, 2009), predominant seed dispersal may take place by water run-off to nearby branches or trees caused by the constant rain and fog condensation that is

69 Chapter 3. Bibliography

characteristic of this forest. Therefore, these two seed dispersal processes at short and long distance would be taking place in this species.

Implication for conservation E. rhopalostele is located in mountain forest where deforestation and fragmentation is happening with consequences for population located in forest. The genetic structure found indicate that there are two distinct genetic groups in this species, than should be have some reproductive isolation lo keep them distinct individuals even coexist in same trees, further these groups are cryptic. Therefore additional studies are necessary to verify if genetic structure depends or is caused by hybridization process with other Epidendrum species. The existence of a fine-scale genetic structure in population has implications when establishing an in situ conservation strategy: it is necessary preserve all forest fragments and its interaction (pollinators, mycorrhiza), since the loss of an area within could involve loss of important genetic diversity in the specie. For reintroduction programs distances between individuals in phorohpytes should be considered, for ensure seed dispersal mechanisms to short and long distance.

Bibliography

70 Chapter 3. Fine scale genetic structure of E. rhopalostele

Colling G, Hemmer P, Bonniot A, Hermant S, Matthies D. 2010. Population genetic structure of wild daffodils (Narcissus peudonarcissus L.) at different spatial scales. Plant Systematics and Evolution 287: 99-111. De hert K, Jacquemyn H, Van Glabeke S, Roldán-Ruiz I, Vandepitte K, Leus L, Honnay O. 2012. Reproductive isolation and hybridization in sympatric populations of three Dactylorhiza species (Orchidaceae) with different ploidy levels. Annals of Botany 109: 4, 709-720. Dearnaley JD. 2007. Further advances in orchid mycorrhizal research. Mycorrhiza 17: 6, 475-486. Devey DS, Bateman RM, Fay MF, Hawkins JA. 2007. Friends or Relatives? Phylogenetics and Species Delimitation in the Controversial European Orchid Genus Ophrys. Annals of Botany 101: 3, 385-402. Epperson BK. 2000. Spatial genetic structure and non-equilibrium demographics within plant populations. Plant Species Biology 15: 269-279, 1-11. Epperson BK. 1993. Recent advances in correlation studies of spatial patterns of genetic variation. Evolutionary Biology 27: 95-155. Escudero A, Iriondo J, Torres M. 2002. Spatial analysis of genetic diversity as a tool for plant conservation. Biological Conservation 113: 351-365. Evanno G, Regnaut S, Goudet J. 2005. Detecting the number of clusters of individuals using the software structure: a simulation study. Molecular Ecology 14: 8, 2611- 2620. Falush D, Stephens M, Pritchard J. 2007. Inference of population structure using multilocus genotype data: dominant markers and null alleles. Molecular Ecology Notes 7: 4, 574-578. Falush D, Stephens M, Pritchard JK. 2003. Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics 164: 1567-1587. Forrest AD, Hollingsworth ML, Hollingsworth PM, Sydes C, Bateman RM. 2004. Population genetic structure in European populations of Spiranthes romanzoffiana set in the context of other genetic studies on orchids. Heredity 92: 3, 218-227. Hágsater E, Soto-Arenas MA. 2005. Epidendrum L. In: Pridgeon AM, Cribb PJ, Chase MW, Rasmussen FN (eds.), Genera Orchidacearum Volume 4, Epidendroideae (Part one). Oxford University Press, Oxford, pp. 236-251. Hágsater E, Dodson C, Sánchez L, Cervantes L, Dressler RL, Silverstone-Sopkin P. 2001. Icones Orchidacearum. The genus Epidendrum. A third century of new species in Epidendrum, fasc 4, part 3 edn. Herbario AMO, D.F. México, México. Hamrick JL, Murawski D, Nason J. 1993. The influence of seed dispersal mechanisms on the genetic structure of tropical tree populations. Vegetatio 107: 108, 281-297. Huff DR, Peakall R, Smouse PE. 1993. RAPD variation within and among populations of outcrossing buffalograss (Buchloe dactyloides (Nutt.) Engelm). Theoretical and Applied Genetics 86: 927-934. Jacquemyn H, Brys R, Vandepitte K, Honnay O, Roldán-Ruiz I. 2006. Fine-scale genetic structure of life history stages in the food-deceptive orchid Orchis purpurea. Molecular Ecology 15: 10, 2801-2808. Jacquemyn H, Wiegand T, Vandepitte K, Brys R, Roldán-Ruiz I, Honnay O. 2009. Multigenerational analysis of spatial structure in the terrestrial, food‐deceptive orchid Orchis mascula . Journal of Ecology 97: 2, 206-216. Jacquemyn H, Brys R, Vandepitte K, Honnay O, Roldán-Ruiz I, Wiegand T. 2007. A spatially explicit analysis of seedling recruitment in the terrestrial orchid Orchis purpurea . New Phytologist 176: 2, 448-459. Jakobsson M, Rosenberg NA. 2007. CLUMPP: a cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics 23: 14, 1801-1806.

71 Chapter 3. Bibliography

Jersáková J, Malinová T. 2007. Spatial aspects of seed dispersal and seedling recruitment in orchids. New Phytologist 176: 2, 237-241. Kartzinel TR, Trapnell DW, Shefferson RP. 2013. Critical importance of large native trees for conservation of a rare neotropical epiphyte. Journal of Ecology 101: 6, 1429-1438. Le Corre V, Roussel G, Zanetto A, Kremer A. 1998. Geographical structure of gene diversity in Quercus petraea (Matt.) Liebl. III. Patterns of variation identified by geostatistical analyses. Heredity 80: 4, 464-473. Machon N, Bardin P, Mazer SJ, Moret J, Godelle B, Austerlitz F. 2003. Relationship between genetic structure and seed and pollen dispersal in the endangered orchid Spiranthes spiralis . New Phytologist 157: 3, 677-687. Marques I, Draper D, Riofrío ML, Naranjo C. 2014. Multiple hybridization events, polyploidy and low postmating isolation entangle the evolution of neotropical species of Epidendrum (Orchidaceae). BMC Evolutionary Biology 14: 1, 20. McCormick MK, Jacquemyn H. 2014. What constrains the distribution of orchid populations? New Phytologist 202: 2, 392-400. Menz M, Phillips R, Anthony J, Bohman B, Dixon KW, Peakall R. 2015. Ecological and genetic evidence for cryptic ecotypes in a rare sexually deceptive orchid, Drakaea elastica. Botanical Journal of the Linnean Society 177: 1, 124-140. Nathan R, Muller-Landau HC. 2000. Spatial patterns of seed dispersal, their determinants and consequences for recruitment. Tree 15: 7, 278-285. Ouborg NJ, Piquot Y, van Groenendael JM. 1999. Population genetics, molecular markers and the study of dispersal in plants. Journal of Ecology 87: 551-568. Peakall R, Smouse P. 2012. GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research—an update. Bioinformatics 28: 19, 2537-2539. Peakall R, Smouse P. 2006. Genalex 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes 6: 1, 288-295. Peakall R, Ruibal M, Lindenmayer DB. 2003. Spatial autocorrelation analysis offers new insights into gene flow in the Australian bush rat, Rattus fuscipes. Evolution; International Journal of Organic Evolution 57: 1182-1195. Peakall R, Whitehead MR. 2014. Floral odour chemistry defines species boundaries and underpins strong reproductive isolation in sexually deceptive orchids. Annals of Botany 113: 2, 341-355. Pinheiro F, Cozzolino S. 2013. Epidendrum (Orchidaceae) as a model system in ecological and evolutionary studies in Neotropics. Taxon 62: 77-88. Pinheiro F, Cozzolino S, Barros F, Gouveia T, Suzuki RM, Fay MF, Palma-Silva C. 2013. Phylogeographic structure and outbreeding depression reveal early stages of reproductive isolation in the Neotropical orchid Epidendrum denticulatum. Evolution 67: 2024-2039. Pinheiro F, Barros F, Palma-Silva C, Meyer D, Fay MF, Suzuki RM, Lexer C, Cozzolino S. 2010. Hybridization and introgression across different ploidy levels in the Neotropical orchids Epidendrum fulgens and E. puniceoluteum (Orchidaceae). Molecular Ecology 19: 3981-3994. Pritchard J, Stephens M, Donnelly P. 2000. Inference of population structure using multilocus genotype data. Genetics 155: 2, 945-959. Rasmussen HN, Rasmussen FN. 2009. Orchid mycorrhiza: implications of a mycophagous life style. Oikos 118: 3, 334-345. Rasmussen HN, Rasmussen FN. 2007. Trophic relationships in orchid mycorrhiza- diversity and implications for conservation. Lankesteriana 7: 334-341. Riofrío ML, Naranjo CJ, Iriondo JM, Torres ME. 2007. Spatial structure of Pleurothallis, Masdevallia, Lepanthes and Epidendrum epiphytic orchids in a fragment of montane cloud forest in south Ecuador. Lankesteriana 7: 1-2, 102-106.

72 Chapter 3. Fine scale genetic structure of E. rhopalostele

Riofrío ML, Cruz D, Torres E, De la Cruz M, Iriondo JM, Suárez JP. 2013. Mycorrhizal preferences and fine spatial structure of the epiphytic orchid Epidendrum rhopalostele. American Journal of Botany 100: 12, 2339-2348. Smouse P, Peakall R, Gonzales E. 2008. A heterogeneity test for fine-scale genetic structure. 17: 3389-3400. Smouse PE, Peakall R. 1999. Spatial autocorrelation analysis of individual multiallele and multilocus genetic structure. Heredity 82: 561-573. Vega Y, Marques I, Castro S, Loureiro J. 2013. Outcomes of extensive hybridization and introgression in Epidendrum (Orchidaceae): can we rely on species boundaries? PloS One 8: 11, e80662. Vekemans X, Hardy OJ. 2004. New insights from fine-scale spatial genetic structure analyses in plant populations. Molecular Ecology 13: 4, 921-935. Vekemans X, Beauwens T, Lemaire M, Roldán-Ruiz I. 2002. Data from amplified fragment length polymorphism (AFLP) markers show indication of size homoplasy and of a relationship between degree of homoplasy and fragment size. Molecular Ecology 11: 1, 139-151. Vos P, Hogers R, Bleeker M, Reijans M, Van de Lee T, Hornes M, Frijters A, Pot J, Peleman J, Kuiper M. 1995. AFLP: a new technique for DNA fingerprinting. Nucleic Acids Research 23: 21, 4407. Waser NM, Mitchell RJ. 1990. Nectar Standing Crops in Delphinium Nelsonii Flowers: Spatial Autocorrelation among Plants? Ecology 71: 1, 116-123. Williams CF, Guries RP. 1994. Genetic consequences of seed dispersal in three sympatric forest herbs. I. Hierarchial population-genetic structure. Evolution; International Journal of Organic Evolution 48: 791-805. Winkler M, Hülber K, Hietz P. 2009. Population dynamics of epiphytic orchids in a metapopulation context. Annals of Botany 104: 5, 995-1004.

Chapter 4

4. Mycorrhizal preferences of epiphytic orchid Epidendrum rhopalostele (Orchidaceae)

“El contenido de este capítulo sirvió para la publicación del artículo: “Riofrío, M., Cruz, D., Torres, E., de la Cruz, M., Iriondo, J.M., Suárez, J.P. Mycorrhizal preferences and fine spatial structure of epiphytic orchid Epidendrum rhopalostele. American Journal of Botany (2013) 100(12): 1–10”.

INTRODUCTION

The establishment and survival of epiphytic orchids depend on factors such as the environmental conditions of the forest that sustain them and, perhaps most importantly, on the presence of appropriate fungi. Because orchid seeds are minute and contain few stored reserves, colonization of a seed by a compatible fungus is essential for germination and early seedling development (Arditti & Ghani, 2000; Smith & Read, 2008). In epiphytic orchids, the stronger limitations

75 Chapter 4. INTRODUCTION

to water and nutrients derived from their particular habitat may have increased the frequency of mycorrhizal associations that extend beyond the germination stage (Martos et al., 2012). Knowing the identity of associated mycorrhizal fungi thus becomes an important task in orchid conservation strategies (Batty et al., 2002; Dearnaley et al., 2012) especially when species require specific fungi (McCormick et al., 2004). The identification of orchid mycorrhizal fungi has been studied for a long time, and more recently DNA-based molecular techniques have been used (Dearnaley et al., 2012). Orchids with different trophic strategies have been found to be associated with different groups of fungi (Rasmussen & Rasmussen, 2007; Rasmussen & Rasmussen, 2009). Mycoheterotrophic (MH) orchids are associated with a diverse group of Ascomycota or Basidiomycota fungi (Bidartondo et al., 2004; Girlanda et al., 2006; Martos et al., 2009). However, autotrophic orchids generally associate with a limited range of Basidiomycota, mainly Agaricomycotina including members of Tulasnellaceae (Suárez et al., 2006; Kottke et al., 2008), Sebacinales (Suárez et al., 2008; Kottke et al., 2008) and Ceratobasidiaceae (Otero et al., 2002; Otero et al., 2007b; Graham & Dearnaley, 2012). Nevertheless, Pucciniomycotina with members of Atractiellales have also been recently noted to form mycorrhizas (Kottke et al., 2010). Concerning the type of habitat, both epiphytic and photosynthetic terrestrial orchid species have been reported to be associated mainly to Tulasnellaceae (Martos et al., 2012). Previous studies have shown that “specificity”, understood as the phylogenetic diversity of the fungi associated with a particular plant species (Taylor et al., 2002) is difficult to predict. In the case of MH orchids, high specificity has been reported between them and their fungal partners (Taylor & Bruns, 1997; McKendrick et al., 2002; Selosse et al., 2002; Selosse et al., 2004; Bidartondo & Read, 2008; Barrett et al., 2010), while specificity varies for photosynthetic orchids. Association with a narrow or even dominant group has been shown for many photosynthetic species, both terrestrial (McCormick et al., 2004; Shefferson et al., 2005; McCormick et al., 2006; Shefferson et al., 2007; Shefferson et al., 2008; Roche et al., 2010; Yuan et al., 2010) and epiphytic orchids (Otero et al., 2002; Otero et al., 2004; Otero et al., 2005; Suárez et al., 2008; Graham & Dearnaley, 2012). Nevertheless, associations between other photosynthetic species and a wide range of fungi (mycorrhizal generalists) have been described (Stark et al., 2009; Jacquemyn et al., 2010). Even closely related orchid species appear to have different degrees of specificity (Shefferson et al., 2007). Degree of mycorrhizal specificity may have important consequences on orchid distribution and conservation. Thus, orchid rarity and vulnerability could be enhanced by their specificity for certain mycorrhizal fungi of rare or patchy distribution (e.g., Swarts et al., 2010). In contrast, orchid species associated with a broad range of fungi (or whose mycorrhiza are common and widespread) would be less predisposed to be endangered, as their seeds would have greater probability of encountering a compatible fungus after dispersal (Bonnardeaux et al., 2007). In any case, it is important to note that orchid distribution is limited not only by the presence of compatible mycorrhizal fungi, but also by its abundance (Diez, 2007; McCormick et al., 2012) and by other factors such as pollination

76 Chapter 4. Mycorrizhal preference of E. rhopalostele

(Pauw & Bond, 2011), seed dispersal (Jacquemyn et al., 2009; Winkler et al., 2009), or environmental conditions (Těšitelová et al., 2012). Although many studies have been published regarding the identification of orchid mycorrhiza, little is still known about the intrapopulation variation in mycorrhizal associations. However, this information is important in orchid conservation, as conservation actions are mainly undertaken at the population level. In the present chapter, we studied the mycorrhizal associations in one population of Epidendrum rhopalostele Hágsater & Dodson, an epiphytic orchid species of South America. Specifically, we asked the following questions: (1) Which fungi are associated with E. rhopalostele? If several species are detected, what are the phylogenetic relationships among them? (2) Do plants associate with one or more fungi simultaneously? (3) Do neighboring orchids share the same mycorrhizal fungus? (4) Is there any association between the identity of mycorrhizal fungi and the type of phorophyte? As found in other epiphytic orchids, we hypothesized that the mycorrhizal fungi of E. rhopalostele would fit in a narrow number of clades belonging to members of Tulasnellaceae, Sebacinales, and Ceratobasidiaceae. MATERIAL AND METHODS

Study species and site Epidendrum rhopalostele is a photosynthetic, epiphytic species native to Ecuador, Peru and Bolivia in evergreen montane forests. It belongs to the alpicolum group, with a single racemose apical inflorescence of 10–25 light green flowers (Hágsater et al., 2001). It is very similar to Epidendrum dialychilum Hágsater & Dodson, both having the lip free from the column, but differs from the latter by its linear-lanceolate and acuminate lip; the long, filiform, acuminate petals; and the stigmatic cavity only in the apical third of the column. Its seed dispersal mechanism and reproductive biology have not been studied and are unknown. The study site is located on the eastern slope of Cordillera Real in the Andes of southern Ecuador on the border of Podocarpus National Park along the Loja- Zamora Road in Zamora-Chinchipe province, at around 2250 m a.s.l. This regenerated forest, ~35-yr old and classified as evergreen, upper montane forest (Beck et al., 2008a), covers an area of ~ 1 ha. The most diverse and abundant seed plant families are Orchidaceae, Melastomataceae, Ericaceae, and Rubiaceae. Mean annual precipitation is 2193 mm, and mean annual temperature is 20.8°C (4.7–25.5°C). A moderate rainy season typically extends from April to July. On average, there is 1 mo with less than 100 mm of rainfall during the driest part of the year, from October to January. Shorter dry spells of 1–2 wk are more frequent. Fog is common at this elevation and provides an additional water surplus of 9.6% (Bendix et al., 2008).

Sampling One adult plant of E. rhopalostele was randomly selected on each host tree for root sampling between January and March 2009 and georeferenced as described

77 Chapter 4. MATERIAL AND METHODS

in chapter 2. Three to four roots per individual plant were packed in aluminum foil to prevent desiccation and transported to the laboratory on the sampling day. To ensure correct identification, we collected all orchid samples from plants in bloom.

Light microscopy Light microscopy was used to select material with coils of hyphae or pelotons. Transversal sections were cut in the middle part of each root sample by hand using a razor blade. Sections were stained with a methyl blue (0.05% w/v) solution for 3 min and examined at 100× to 1000× (Zeiss Axioskop 2). To reduce contamination with nonmycorrhizal fungi, we removed the velamen using a stereomicroscope, and the remaining cortical tissue was collected in microtubes for DNA isolation.

DNA isolation, PCR, cloning, and sequencing DNA was extracted from colonized root pieces of ~1–2 cm long using a Plant DNAeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer’s instructions. A PCR was conducted to amplify the ITS-5.8S rDNA region with the universal primers ITS1 (5'-TCC GTA GGT GAA CCT GCG-3') (White et al., 1990) and TW14 (5'-GCT ATC CTG AGG GAA ACT TC-3') (Cullings, 1994). This primer combination was chosen to increase the possibility of identifying the entire diversity of fungi within the root. PCR reactions were conducted in 20-µL reaction volume containing 0.5 pmol of each primer, 0.8 µg/µL of bovine serum albumin (SIGMA-USA)(Iotti & Zambonelli, 2006)and 10 µL of the Phusion High-Fidelity PCR Master Mix (Finnzymes, Espoo, Finland). In each PCR, a negative control of PCR mix without DNA template was included. The thermocycler program used for reactions was as follows: 98°C for 30 s (1 cycle); denaturation at 98°C for 10 s, annealing at 60°C for 20 s, and elongation at 72°C for 30 s (30 cycles); and a final extension at 72°C for 7 min. Success of PCR amplification was tested in 1% agarose gel. PCR products were cloned with Zero Blunt TOPO PCR cloning kit (Invitrogen) following the manufacturer’s protocol. To verify the DNA insert, we randomly selected 12 growing colonies for each sample, for a direct PCR with M13F and M13R primers modified according to Krüger et al. (2009). PCR conditions were as follows: 95°C for 5 min (1 cycle); denaturation at 94°C for 30 s, annealing 65°C for 30 s, elongation at 72°C for 2 min (35 cycles); and final extension at 72°C for 10 min. Success of the PCR was tested as mentioned before. The colonies with inserts were purified following S.N.A.P. miniprep kit (Invitrogen) protocol. Sequencing was carried out using primer M13F.

Phylogenetic analyses of Epidendrum rhopalostele mycobionts We used BLAST against the NCBI nucleotide database (GenBank; http://www.ncbi.nlm.nih.gov/) to detect published sequences with a high similarity. The majority of obtained sequences were closely related to

78 Chapter 4. Mycorrizhal preference of E. rhopalostele

Tulasnellaceae. Due to the heterogeneity of the Tulasnellaceae sequences, the ITS-5.8S rDNA region could not be aligned over the whole data set. Therefore, we used the 5.8S region (160 bp) to calculate a first phylogenetic tree of a wider phylogenetic spectrum. Subsequently, we carried out phylogenetic analyses for each subset of related sequences. The ITS-5.8S rDNA region was analyzed considering sequences derived from this study and other closely related sequences from GenBank. Sequences were aligned with G-INS-I strategy implemented in MAFFT version 5.667 (Katoh et al., 2005) Phylogenetic analyses were performed with neighbor-joining (NJ), Bayesian likelihood, and maximum-likelihood (ML) analyses. Neighbor-joining was implemented in the program PAUP* (Swofford, 2002) using the BIONJ modification (Gascuel, 1997). Branch support was tested with 1000 bootstrap replicates. The Bayesian likelihood was based on Markov chain Monte Carlo (MCMC) method as implemented in MrBayes, version 3.1.2 (Huelsenbeck & Ronquist, 2001)We ran two independent MCMC analyses, each involving four incrementally heated Markov chains over 4 million generations and using random starting trees with GTR+I+G substitution model. Trees were sampled over 100 generations resulting in a total of 40 000 trees in each run from which 30 000 were used to compute a pooled majority rule consensus tree. For heuristic analysis, the program PHyML version 2.4.4 (Guindon & Gascuel, 2003; Guindon et al., 2005) was used with GTR+I+G DNA substitution model. Gamma distribution was approximated with four discrete rate categories; all model parameters were estimated using maximum likelihood. Branch support was inferred from 1000 bootstrap replicates.

Spatial and association analyses of mycorrhizal fungi Spatial point pattern analysis was used to evaluate if neighboring orchids share the same mycorrhizal fungus. Spatial segregation of mycorrhiza clades was tested by means of differences between the univariate and bivariate K-functions [Ki(r)- Kj(r), Ki(r)-Kij(r) and Kj(r)-Kij(r)] (Dixon, 2002). Because we did not detect heterogeneity at the scale of the study, we tested the observed differences of K functions against the null model of random labeling. Spatial analysis was carried out with K1K2 functions of ecespa package (De La Cruz et al., 2008) in the R environment (R Development Core Team., 2011). To evaluate if the identity of mycorrhizal fungi is related to a type of phorophyte, we employed contingency table analysis (Agresti, 2002). In particular we tested the association of the obtained mycorrhiza clades with the status (dead or alive) and the character (shade-tolerant or pioneer) of phorophyte. Trees without branches or leaves, many of which showed clear signs of bark decomposition, and large branches fallen on the ground were classified as dead. Functional categories (shade-tolerant or pioneer) were assigned according to literature (Finegan, 1992; Poorter et al., 2006).

79 Chapter 4. RESULTS

RESULTS

Light microscopy examination showed that the roots of 22 of the 25 sampled individuals of E. rhopalostele were colonized by mycorrhizal fungi forming coils. Vital and collapsed pelotons were present in the cortical cells, and abundant hyphae were found in the velamen (Fig. 4.1.).

Fig. 4.1. Cross section of a mycorrhizal Epidendrum rhopalostele root stained with methyl blue (×100). Fungal hyphae (hv) traverse the velamen (V) and epidermis (E) and form tight coils called pelotons within cortical cells (CC). After a few days, pelotons lyse and the lysed products are absorbed by the host cells. (Single arrow: vital peloton; double arrow: collapsed peloton.

Identification and phylogenetic analysis of Epidendrum rhopalostele mycobionts PCR amplification of the ITS-5.8S rDNA region was successful for 19 of the 22 analyzed mycorrhizal samples. A total of 47 sequences were used; identical sequences from the same roots, chimeric sequences and low-quality sequences were discarded. BLAST searches in GenBank showed that 44 sequences were close to Tulasnella. The sequences obtained in this study are available in GenBank under accession numbers JF346765–JF346853. Additionally, two other sequences belonging to other Basidiomycota close to Hyphoderma (GenBank

80 Chapter 4. Mycorrizhal preference of E. rhopalostele

DQ873597.1) and Infundibura (GenBank AJ406404.1) and one sequence close to the Ascomycota Hyalodendriella (GenBank EU040232.1) were detected in the same root samples where Tulasnella was present. Finally, two identical sequences belonging to E. rhopalostele were identified (GenBank KC165027, KC165028). The phylogenetic analyses of the sole 5.8S region showed the sequences from E. rhopalostele mycobionts in two distinct clades within Tulasnellaceae DNA alignment (Appendix 4.1). The phylogenetic analyses of ITS1-5.8S-ITS2 for each subset increased phylogenetic resolution. Clade A (Fig. 4.2) was supported by 54/100/100 (MCMC / heuristic ML / BIONJ, respectively), and 63% of all sequences were found in this subset. Within clade A, sequences shared 99% similarity in the region ITS1-5.8S-ITS2, except in the case of clone JF346765 with just 98% similarity. The closest related sequences to clade A, supported by values 56/85/94 (MCMC / heuristic ML / BIONJ) (Fig. 4.2) were Tulasnellaceae from epiphytic orchids (subtribe Pleurothallidinae) Stelis concinna Lindl. and Pleurothallis lilijae Foldats, with 95% similarity, and Stelis superbiens Lindl. and Stelis hallii Lindl. with 84 to 95% similarity. Clade B was included in the second subset (Fig. 4.3) with a support of 94/99/100 (MCMC / Heuristic ML / BIONJ), including 37% of Tulasnella sequences. All sequences of clade B shared a similarity of 99% in ITS1-5.8S-ITS2, except for the sequence JF346853 with 95% similarity compared to the rest. Sequences in clade B are close to Tulasnella asymmetrica Warcup & P.H.B.Talbot (DQ520101 and DQ388046.1, with 84% similarity), forming a clade supported by 100/99/100 (MCMC / heuristic ML / BIONJ). Tulasnellaceae sequences displayed in clade A were present in nine orchid individuals, and Tulasnellaceae sequences from clade B were present in six orchids, whereas clades A and B were simultaneously found in four orchid individuals.

81 Chapter 4. RESULTS

96/62/72 from Epidendrum rhopalostele 227-C125 JF346776 from Epidendrum rhopalostele 198-C118 JF346777 99/61/62 from Epidendrum rhopalostele 104-C061 JF346783 from Epidendrum rhopalostele 103-C051 JF346775 99/84/82 from Epidendrum rhopalostele 34-C005 JF346780 87/56/54 from Epidendrum rhopalostele 34-C003 JF346778 from Epidendrum rhopalostele 103-C049 JF346839 98/67/67 from Epidendrum rhopalostele 57-C021 JF346790 from Epidendrum rhopalostele 59-C030 JF346794 from Epidendrum rhopalostele 88-C037 JF346787 from Epidendrum rhopalostele 88-C033 JF346766 from Epidendrum rhopalostele 59-C025 JF346814 from Epidendrum rhopalostele 119-C071 JF346799 from Epidendrum rhopalostele 119-C068 JF346796 A from Epidendrum rhopalostele 151-C097 JF346802 from Epidendrum rhopalostele 151-C093 JF346807 from Epidendrum rhopalostele 253-C139 JF346810 from Epidendrum rhopalostele 104-C056 JF346816 from Epidendrum rhopalostele 241-C134 JF346812 from Epidendrum rhopalostele 34-C001 JF346808 from Epidendrum rhopalostele 103-C050 JF346820 from Epidendrum rhopalostele 253-C142 JF346800 from Epidendrum rhopalostele 119-C069 JF346817 from Epidendrum rhopalostele 241-C131 JF346818 54/100/100 from Epidendrum rhopalostele 103-C053 JF346769 - /- /77 from Epidendrum rhopalostele 88-C036 JF346765 from Epidendrum rhopalostele 147-C091 JF346819 from Pleurothallis lilijae DQ178098

84/48/ - from Stelis superbiens DQ178101 89/79/88 from Pleurothallis lilijae DQ178100 56/85/94 from Stelis concinna DQ178084 66/99/81 from Stelis concinna DQ178092 from Stelis halli DQ178103 from Stelis halli DQ178104 - / - /56 - / - /69 from Stelis superbiens DQ178105 from Stelis concinna DQ178109 -/63/97 from Stelis concinna DQ178108 73/99/100 from Stelis concinna DQ178110

97/70/66 from Stelis concinna DQ178111 Tulasnella calospora DQ388043

50/43/53 Tulasnella calospora EF393621 60/ - /60 Tulasnella calospora DQ388044 98/82/84 from Goodyera pubescens AY373264 91/89/99 from Goodyera pubescens AY373270 from Goodyera pubescens AY373274 - /38/ - Tulasnella bifrons AY373290 84/62/54 91/74/87 Tulasnella calospora DQ388042 - / - /55 Tulasnella calospora EU218888 2 - / - /74 Tulasnella calospora DQ388045 97/100/100 Tulasnella calospora DQ388041 72/97/100 Tulasnella calospora AY373298 98/99/99 Tulasnella deliquescens AY373291 - /73/82 from Arundina graminifolia AJ313435 99/100/100 from Spathaglotis plicata AJ313450 77/57/69 - / 29 /52 from Spathaglotis plicata AJ313452 87/63/63 from Spathaglotis plicata AJ313453 - /50/71 from Spathaglotis plicata AJ313439 from Spathaglotis plicata AJ313437 from Epipactis gigantea AY634124 -/81/86 from Vanda miss Joaquin AJ313443 from Dendrobium crumenatun AJ313438 59/100/100 100/100/100 from Stelis halli DQ178114 from Stelis halli DQ178113 100/70/98 99/89/78 from Stelis superbiens DQ178115 75/-/- Tulasnella danica AY373297 Tulasnella asymmetrica DQ520101 Tulasnella albida AY373294 Tulasnella pruinosa AY373295

0.01 substitutions/site

Fig. 4.2. Clade A of Epidendrum rhopalostele mycobionts. Phylogenetic placement of Tulasnella sequences, inferred by Markov chain Monte Carlo (MCMC) analysis of the ITS-5.8S region from nuclear rDNA sequences. Numbers on branches correspond to MCMC analysis/heuristic maximum likelihood bootstrap/neighbor-joining bootstrap (only values >50% are shown). Note that genetic distances cannot be directly correlated to branch lengths in the trees, since highly diverse alignment regions were excluded for tree calculation. The tree was rooted with Tulasnella pruinosa AY373295.

82 Chapter 4. Mycorrizhal preference of E. rhopalostele

from Epidendrum rhopalostele 57-C023 JF346828

from Epidendrum rhopalostele 191-C114 JF346846

from Epidendrum rhopalostele 227-C127 JF346848

from Epidendrum rhopalostele 91-C044 JF346831

from Epidendrum rhopalostele 191-C109 JF346847

from Epidendrum rhopalostele 171-C102 JF346850

from Epidendrum rhopalostele 135-C075 JF346852

from Epidendrum rhopalostele 147-C085 JF346832

from Epidendrum rhopalostele 191-C116 JF346827 B

from Epidendrum rhopalostele 39-C014 JF346834

from Epidendrum rhopalostele 171-C104 JF346842

from Epidendrum rhopalostele 119-C070 JF346824 - /60/61 from Epidendrum rhopalostele 206-C080 JF346836

from Epidendrum rhopalostele 135-C073 JF346826 98/100/100 from Epidendrum rhopalostele 91-C042 JF346823

94/99/100 from Epidendrum rhopalostele 91-C045 JF346821

from Epidendrum rhopalostele 119-C064 JF346853 100/98/100 Tulasnella asymmetrica DQ520101.1 51/100/71

100 /100/100 Tulasnella asymmetrica DQ520101

Tulasnella asymmetrica DQ388046.1

from Stelis halli DQ178072 - / 98/79

- / - /84 from Stelis halli DQ178073 100/56/75 from Stelis halli DQ178071

- / - /94 from Stelis concinna DQ178075

100/100/100 from Stelis concinna DQ178076

from Stelis superbiens DQ178074 100/97/100 76/80/75 Tulasnella violea AY373303 100/100/99

69/60/99 Tulasnella albida AY373294 99/88/75 Tulasnella pruinosa AY373295

Tulasnella asymmetrica DQ388047 93/100/ -

100/100/100 from Pleurothallis lilijae DQ178070 from Pleurothallis lilijae DQ178069

Tulasnella sp. from Cryptothallus mirabilis AY192482 99/ - /100

100/ - /100 from Cryptothallus mirabilis AY192483

Tulasnella sp. from Pinus muricata AY192461

Tulasnella violea DQ520097 0.01 substitutions/site

Fig. 4.3. Clade B of Epidendrum rhopalostele mycobionts. Phylogenetic placement of Tulasnella sequences, clade B from Epidendrum rhopalostele, inferred by Markov chain Monte Carlo (MCMC) analysis of ITS-5.8S region from nuclear rDNA. The values that support the nodes correspond to MCMC analysis/heuristic maximum likelihood bootstrap/neighbor-joining bootstrap (only values >50% are shown; hyphens indicate absence of these values for that analysis). Note that genetic distances cannot be directly correlated with the length of the branches on the trees, since the highly diverse alignment regions were excluded from the construction of the tree. The tree was rooted with Tulasnella violea DQ520097.

Chapter 4. Mycorrizhal preference of E. rhopalostele

DISCUSSION

Although a high number of epiphytic orchid species have been recorded, studies on their mycorrhizal fungi are still scarce (Dearnaley et al., 2012) and, in most cases, only one or two individuals per population have been sampled. Our study provides a novel focus by assessing within-population variation of epiphytic orchid mycorrhizal interactions in a spatially explicit way. We thus found that E. rhopalostele can associate separately or simultaneously with two different clades of closely related Tulasnella. These clades are spatially randomly distributed showing no segregation patterns that would suggest a limited distribution of the fungi or competitive exclusion between clades. We have also found that this particular orchid is more likely than expected to be found on dead and fallen trees. Our results thus contribute to improving knowledge on epiphytic orchid species, providing relevant information on mycorrhizal fungi preference of E. rhopalostele, and identifying favorable environments at the population level.

Identification and phylogenetic relationships of fungi associated with E. rhopalostele Light microscopy examination showed that 22 of the 25 E. rhopalostele individuals were consistently colonized. Roots revealed the presence of vital and collapsed pelotons in the same tissue suggesting the possibility of subsequent reinfection in roots. Beyond the essential support that mycorrhizal fungi provide during seed germination, the presence of the fungi in adult orchid individuals may be useful to retain the fungus in the neighborhood to assure further seed germination, which could explain the high number of juveniles growing close to adults. The maintenance of this association in adult plants could also be advantageous in adverse seasons or under conditions of high shade, as plants could obtain part of organic carbon through their mycorrhizal fungi (Hudson, 1992; Dearnaley et al., 2012). DNA sequence analysis showed the presence of Tulasnella in E. rhopalostele roots, which agrees with reports for other species of the same tribe (tribe Epidendreae). Zettler and Hofer (1998) and Pereira et al. (2003), identified the anamorphic genus Epulorhiza from mycorrhizal roots of Epidendrum conopseum R.Br. and Epidendrum rigidum Jacq., respectively, and Suárez et al. (2006; 2008) identified Tulasnella in Stelis hallii, S. superbiens, S. concinna, and Pleurothallis lilijae. Our results showed that E. rhopalostele has a preference for two Tulasnella clades: (1) sequences of clade A were grouped with sequences of mycorrhizal fungi isolated from Stelis and Pleurothallis that correspond to clades A, B, and C reported by Suárez et al. (2006) all of which form a subset related to Tulasnella calospora Hadley, and (2) sequences of clade B were close to the clade of Tulasnella asymmetrica and to sequences from mycorrhizas isolated from Stelis and Pleurothallis that correspond to clades E and F of Suárez et al. (2006) and other Tulasnella species. Because the percentage of similarity between sequences of the same clade was over 95%, the mycobionts of each clade could be

85 Chapter 4. DISCUSSION

considered “genotypes” of the same Tulasnella species or operational taxonomic unit (OTU) (Lindner & Banik, 2011). These results suggest, therefore, that there is no absolute fungal specificity in E. rhopalostele, as it can form mycorrhizas with at least two Tulasnella species. Concerning the similarity of the mycorrhizal sequences isolated for E. rhopalostele and those reported for Stelis hallii, S. superbiens, S. concinna, and Pleurothallis lilijae, it is important to note that the study by Suárez et al. (2006) was carried out in a nearby site with the same forest type as E. rhopalostele. In addition to Tulasnella sequences, other Basidiomycota close to Infundibura and Hyphoderma and one Ascomycota close to Hyalodendriella were also found. Although Kottke et al. (2010) recently showed that other Atractiellales closely related to Infundibura formed mycorrhizas with several terrestrial and epiphytic orchid species, the presence of only one sequence does not provide enough evidence to conclude that these fungi form mycorrhizas with E. rhopalostele. In this sense, a highly diverse group of fungi has been reported to colonize only the velamen in epiphytic orchids, including mostly Ascomycota with members of Helotiales closest to Hyalodendriella as well as Polyporales such as Hyphoderma (Herrera et al., 2010). Thus, it would be interesting to extend this study to other known populations of E. rhopalostele occurring in Ecuador, Peru and Bolivia to confirm genus-level specificity to Tulasnella.

Do plants associate with one or more fungi simultaneously? Do neighboring orchids share the same mycorrhizal fungus? Four individuals of E. rhopalostele were found to be associated with two Tulasnella species at the same time, which supports the hypothesis of successive reinfections. Multiple fungal associations have also been described in some terrestrial orchid species, both photosynthetic (Jacquemyn et al., 2010; Jacquemyn et al., 2012; Martos et al., 2012) and mycoheterotrophic (Roy et al., 2009), and also in some epiphytic orchid species (Martos et al., 2012; Xing et al., 2013), although it is difficult to know if this is a common phenomenon because most studies do not provide this information. As suggested by Jacquemyn et al. (2012) associating with multiple fungi can provide an advantage if the associated fungi belong to different fungal lineages and have different nutritional and environmental requirements. Nevertheless, in our case, this is unlikely because the two mycorrhizal species found in the roots of E. rhopalostele are phylogenetically very close. Because seed colonization by a compatible fungus is essential for germination and early seedling development, the spatial distribution of E. rhopalostele depends on the presence of the fungi and, therefore, it could potentially be influenced by the spatial distribution of the fungi. The detailed spatial distribution of the two Tulasnella species associated with E. rhopalostele is unknown in this forest, and there is virtually no information about patterns of fungal small-scale spatial distribution in the environment (Phillips et al., 2011). However, some conclusions can be derived from the bivariate spatial analysis, which found no trend in neighboring orchids to share the same mycorrhizal fungus. The existence

86 Chapter 4. Mycorrizhal preference of E. rhopalostele

of segregation between clade A and clade B of Tulasnella could have indicated a limited spatial distribution of fungi. The absence of such a pattern and the fact that both Tulasnella clades could be found in the same tree, and even in the same plant, provide no evidence in support of a limited distribution of Tulasnella.

Final remarks From a conservation perspective, the knowledge provided in this study may help detect recruitment sites and support management measures in the forest where the study took place. Thus, fallen and dead trees could be left undisturbed to promote recruitment. Specific studies to characterize the spatial distribution of the two Tulasnella species in the forest and their ecological requirements would be advisable because, in the end, the factors that influence the presence and abundance of these fungi indirectly affect E. rhopalostele establishment and survival. Furthermore, studies on other known populations of this species are needed to complete the characterization of microsite suitability at the species level and to confirm genus-level specificity to Tulasnella.

Bibliography

Agresti A. 2002. Categorical data analysis, 2nd edn. Wiley, Hoboken, New Jersey, USA. Arditti J, Ghani AK. 2000. Numerical and physical properties of orchid seeds and their biological implications. New Phytologist 145: 3, 367-421. Barrett CF, Freudenstein JV, Taylor DL, Koljalg U. 2010. Rangewide analysis of fungal associations in the fully mycoheterotrophic Corallorhiza striata complex (Orchidaceae) reveals extreme specificity on ectomycorrhizal Tomentella (Thelephoraceae) across North America. American Journal of Botany 97: 4, 628-643. Batty AL, Dixon KW, Brundrett MC, Sivasithamparam K. 2002. Orchid conservation and mycorrizal associations. In: Sivasithamparam K, Dixon KW, Barret RL (eds.), Microorganisms in plant conservation and biodiversity. Kluwer, Dordrecht, Netherlands, pp. 195-226. Beck E, Makeschin F, Haubrich F, Richter M, Bendix J, Valarezo C. 2008. The ecosystem (Reserva Biológica San Francisco). In: Beck E, Bendix J, Kottke I, Makenschin F, Mosandl R (eds.), Gradients in a tropical mountain ecosystem of Ecuador. Ecological studies. Springer-Verlag Berlin Heidelberg, Germany, pp. 1- 14. Bendix J, Rollenbeck R, Fabian P, Emck P. 2008. Climate. In: Beck E, Bendix J, Kottke I, Makeschin F, Mosandl R (eds.), Gradients in a Tropical Mountain Ecosystem of Ecuador. Springer-Verlag Berlin Heidelberg, Germany, pp. 281-290. Bidartondo MI, Read DJ. 2008. Fungal specificity bottlenecks during orchid germination and development. Molecular Ecology 17: 3707-3716. Bidartondo MI, Burghardt B, Gebauer G, Bruns TD, Read DJ. 2004. Changing partners in the dark: isotopic and molecular evidence of ectomycorrhizal liaisons between forest orchids and trees. Proceedings of the Royal Society of London Biological Science Series B 271: 1799-1806.

87 Chapter 4. Bibliography

Bonnardeaux Y, Brundrett M, Batty AL, Dixon KW, Koch J, Sivasithamparam K. 2007. Diversity of mycorrhizal fungi of terrestrial orchids: compatibility webs, brief encounters, lasting relationships and alien invasions. Mycological Research 111: 1, 51-61. Cullings KW. 1994. Molecular phylogeny of the Monotropoideae (Ericaceae) with a note on the placement of the Pyroloideae. Journal of Evolutionary Biology 7: 4, 501-516. De La Cruz M, Romao RL, Escudero A, Maestre FT. 2008. Where do seedlings go? A spatio-temporal analysis of seedling mortality in a semi-arid gypsophyte. Ecography 31: 6, 720-730. Dearnaley JD, Martos F, Selosse MA. 2012. Orchid Mycorrhizas: Molecular ecology, physiology, evolution and conservation aspects. In: Hock B (ed.), Fungal associations. The mycota. Springer-Verlang, Berlin, Germany, pp. 207-230. Diez J. 2007. Hierarchical patterns of symbiotic orchid germination linked to adult proximity and environmental gradients. Journal of Ecology 95: 1, 159-170. Dixon PM. 2002. Ripley's K Function. In: El-Shaarawi A, Piergorsch W (eds.), The encyclopedia of Environmetrics. Wiley, New York, New York, USA, pp. 1976- 1803. Finegan B. 1992. Bases ecológicas para la silvicultura, Tema I. V. Curso internacional sobre silvicultura y manejo de bosques tropicales, Centro Agronómico Tropical de Investigación y Enseñanza, Turrialba, Costa Rica. Gascuel O. 1997. BIONJ: an improved version of the NJ algorithm based on a simple model of sequence data. Molecular Biology and Evolution 14: 7, 685-695. Girlanda M, Selosse MA, Cafasso D, Brilli F, Delfine S, Fabbian R, Ghignone S, Pinelli P, Segreto R, Loreto F. 2006. Inefficient photosynthesis in the Mediterranean orchid Limodorum abortivum is mirrored by specific association to ectomycorrhizal Russulaceae. Molecular Ecology 15: 2, 491-504. Graham RR, Dearnaley JD. 2012. The rare Australian epiphytic orchid Sarcochilus weinthalii associates with a single species of Ceratobasidium. Fungal Diversity 54: 31-37. Guindon S, Lethiec F, Duroux P, Gascuel O. 2005. PHYML Online--a web server for fast maximum likelihood-based phylogenetic inference. Nucleic Acids Research 33: W557-W559. Guindon S, Gascuel O. 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology 52: 5, 696-704. Hágsater E, Dodson C, Sánchez L, Cervantes L, Dressler RL, Silverstone-Sopkin P. 2001. Icones Orchidacearum. The genus Epidendrum. A third century of new species in Epidendrum, fasc 4, part 3 edn. Herbario AMO, D.F. México, México. Herrera P, Suárez JP, Kottke I. 2010. Orchids keep the ascomycetes outside: A highly diverse group of ascomycetes colonizing the velamen of epiphytic orchids from a tropical mountain rainforest in Southern Ecuador. Mycology 1: 262-268. Hudson HJ. 1992. Fungi as mutualistic symbionts in endomycorrhizas. In: Hudson HJ (ed.), Fungal biology. Cambridge University Press, Cambridge, UK., pp. 214-241. Huelsenbeck JP, Ronquist F. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17: 8, 754-755. Iotti M, Zambonelli A. 2006. A quick and precise technique for identifiying ectomycorrhizas by PCR. Mycological Research 110: 60-65. Jacquemyn H, Deja A, De hert K, Cachapa Bailarote B, Lievens B. 2012. Variation in mycorrhizal associations with Tulasnelloid fungi among populations of five Dactylorhiza species. PLoS ONE 7: 8, e42212. Jacquemyn H, Honnay O, Cammue BPA, Brys R, Lievens B. 2010. Low specificity and nested subset structure characterize mycorrhizal associations in five closely related species of the genus Orchis. Molecular Ecology 19: 18, 4086-4095.

88 Chapter 4. Mycorrizhal preference of E. rhopalostele

Jacquemyn H, Wiegand T, Vandepitte K, Brys R, Roldán-Ruiz I, Honnay O. 2009. Multigenerational analysis of spatial structure in the terrestrial, food‐deceptive orchid Orchis mascula . Journal of Ecology 97: 2, 206-216. Katoh K, Kuma K, Toh H, Miyata T. 2005. MAFFT version 5: improvement in accuracy of multiple sequence alignment. Nucleic Acids Research 33: 2, 511-518. Kottke I, Suárez JP, Herrera P, Cruz D, Bauer R, Haug I, Garnica S. 2010. Atractiellomycetes belonging to the "rust" lineagae (Pucciniomycotina) form mycorrhizae with terrestrial and epiphytic netropical orchids. Proceedings of the Royal Society of London Biological Science Series B 277: 1289-1298. Kottke I, Haug I, Setaro S, Suárez JP, Weiß M, Preusing M, Nebel M, Oberwinkler F. 2008. Guilds of mycorrhizal fungi and their relation to trees, ericads, orchids and liverworts in a neotropical mountain rain forest. Basic and Applied Ecology 9: 1, 13-23. Krüger M, Stockinger H, Krüger C, Schüßler A. 2009. DNA-based species-level detection of arbuscular mycorrhizal fungi: one PCR primer set for all AMF. New Phytologist 183: 212-223. Lindner DL, Banik MT. 2011. Intragenomic variation in the ITS rDNA region obscures phylogenetic relationships and inflates estimates of operational taxonomic units in genus Laetiporus. Mycologia 103: 731, 740. Martos F, Munoz F, Pailler T, Kottke I, Gonneau C, Selosse MA. 2012. The role of epiphytism in architecture and evolutionary constraint within mycorrhizal networks of tropical orchids. Molecular Ecology 21: 20, 5098-5109. Martos F, Dulormne M, Pailler T, Bonfante P, Faccio A, Fournel J, Dubois MP, Selosse MA. 2009. Independent recruitment of saprotrophic fungi as mycorrhizal partners by tropical achlorophyllous orchids. New Phytologist 184: 3, 668-681. McCormick MK, Lee Taylor D, Juhaszova K, Burnett RK, Whigham DF, O’Neill JP. 2012. Limitations on orchid recruitment: not a simple picture. Molecular Ecology 21: 6, 1511-1523. McCormick MK, Whigham DF, Sloan D, O'Malley K, Hodkinson B. 2006. Orchid- fungus fidelity: a marriage meant to last? Ecology 87: 4, 903-911. McCormick MK, Whigham DF, O'neill J. 2004. Mycorrhizal diversity in photosynthetic terrestrial orchids. New Phytologist 163: 2, 425-438. McKendrick SL, Leake J, Taylor D, Read DJ. 2002. Symbiotic germination and development of the myco-heterotrophic orchid Neottia nidus-avis in nature and its requirement for locally distributed Sebacina spp. New Phytologist 154: 233-247. Otero J, Flanagan N, Herre E, Ackerman JD, Bayman P. 2007. Widespread mycorrhizal specificity correlates to mycorrhizal function in the neotropical, epiphytic orchid Ionopsis utricularioides (Orchidaceae). American Journal of Botany 94: 12, 1944- 1950. Otero J, Ackerman JD, Bayman P. 2004. Differences in mycorrhizal preferences between two tropical orchids. Molecular Ecology 13: 8, 2393-404. Otero JT, Bayman P, Ackerman JD. 2005. Variation in mycorrhizal performance in the epiphytic orchid Tolumnia variegata in vitro: the potential for natural selection. Evolutionary Ecology 19: 1, 29-43. Otero JT, Ackerman JD, Bayman P. 2002. Diversity and host specificity of endophytic Rhizoctonia-like fungi from tropical orchids. American Journal of Botany 89: 11, 1852-1858. Pauw A, Bond W. 2011. Mutualisms matter: pollination rate limits the distribution of oil- secreting orchids. Oikos 120: 10, 1531-1538. Pereira O, Rollemberg C, Borges A, Matsuoka K, Kasuya M. 2003. Epulorhiza epiphytica sp. nov. isolated from mycorrhizal roots of epiphytic orchids in Brazil. Mycoscience 44: 2, 153-155.

89 Chapter 4. Bibliography

Phillips RD, Barrett MD, Dixon KW, Hopper SD. 2011. Do mycorrhizal symbioses cause rarity in orchids? Journal of Ecology 99: 858-869. Poorter L, Bongers L, Bongers F. 2006. Architecture of 54 moist-forest tree species: Traits, trade-offs, and functional groups. Ecology 87: 1289-1013. R Development Core Team. 2011. R: A language and environment for statistical computing, R Foundation for Statistical Computing, Vienna, Australia. Rasmussen HN, Rasmussen FN. 2009. Orchid mycorrhiza: implications of a mycophagous life style. Oikos 118: 3, 334-345. Rasmussen HN, Rasmussen FN. 2007. Trophic relationships in orchid mycorrhiza- diversity and implications for conservation. Lankesteriana 7: 334-341. Roche SA, Carter RJ, Peakall R, Smith LM, Whitehead MR, Linde CC. 2010. A narrow group of monophyletic Tulasnella (Tulasnellaceae) symbiont lineages are associated with multiple species of Chiloglottis (Orchidaceae): Implications for orchid diversity. American Journal of Botany 97: 8, 1313-1327. Roy M, Watthana S, Stier A, Richard F, Vessabutr S, Selosse MA. 2009. Two mycoheterotrophic orchids from Thailand tropical dipterocarpacean forests associate with a broad diversity of ectomycorrhizal fungi. BMC Biology 7: 51, . Selosse MA, Faccio A, Scappaticci G, Bofante P. 2004. Chlorophyllous and achlorophyllous specimens of Epipactis microphylla (Neottieae, Orchidaceae) are associated with ectomycorrhizal Septomycetes, including trufﬂes. Microbial Ecology 47: 416-426. Selosse MA, Weiß M, Jany JL, Tillier A. 2002. Communities and populations of sebacinoid basidiomycetes associated with the achlorophyllous orchid Neottia nidus-avis (L.) L.C.M. Rich. and neighbouring tree ectomycorrhizae. Molecular Ecology 11: 1831-1844. Shefferson RP, Kull T, Tali K. 2008. Mycorrhizal interactions of orchids colonizing Estonian mine tailings hills. American Journal of Botany 95: 2, 156-164. Shefferson RP, Taylor DL, Weiß M, Garnica S, McCormick MK, Adams S, Gray H, McFarland J, Kull T, Tali K, and others. 2007. The evolutionary history of mycorrhizal specificity among lady's slipper orchids. Evolution 61: 6, 1380-1390. Shefferson RP, Weiss M, Kull T, Taylor D. 2005. High specificity generally characterizes mycorrhizal association in rare lady's slipper orchids, genus Cypripedium. Molecular Ecology 14: 2, 613-626. Smith SE, Read DJ. 2008. Mycorrhizal symbiosis, 3rd edn. Academic Press, New York, New York, USA. Stark C, Babik W, Durka W. 2009. Fungi from the roots of the common terrestrial orchid Gymnadenia conopsea. Mycological Research 113: 952-959. Suárez JP, Weiss M, Abele A, Oberwinkler F, Kottke I. 2008. Members of Sebacinales subgroup B form mycorrhizae with epiphytic orchids in a neotropical mountain rain forest. Mycological Progress 7: 75-85. Suárez JP, Weiß M, Abele A, Garnica S, Oberwinkler F, Kottke I. 2006. Diverse tulasnelloid fungi form mycorrhizas with epiphytic orchids in an Andean cloud forest. Mycological Research 110: 11, 1257-1270. Swofford DL. 2002. PAUP*. Phylogenetic analysis using parsimony (*and other methods), version 4.0b10.0, Sinauer, Sunderland, Massachusetts, USA. Taylor DL, Bruns TD, Leake JR, Read DJ. 2002. Mycorrhizal specificity and function in myco-heterotrophic plants. In: van der Heijden MGA, Sanders I (eds.), Mycorrhizal ecology. Springer, Berlin, Germany, pp. 375-413. Taylor DL, Bruns TD. 1997. Independent, specialized invasions of ectomycorrhizal mutualism by two nonphotosynthetic orchids. Proceedings of the National Academy of Sciences 94: 4510-4515.

90 Chapter 4. Mycorrizhal preference of E. rhopalostele

Těšitelová T, Těšitel J, Jersáková J, Říhová G, Selosse MA. 2012. Symbiotic germination capability of four Epipactis (Orchidaceae) is broader than expected from adult ecology. American Journal of Botany 99: 1020-1032. White TJ, Bruns T, Lee S, Taylor JW. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ (eds.), PCR protocols and applications: A laboratory manual. Academic Press, San Diego, California, USA, pp. 315-322. Winkler M, Hülber K, Hietz P. 2009. Population dynamics of epiphytic orchids in a metapopulation context. Annals of Botany 104: 5, 995-1004. Xing X, Ma X, Deng Z, Chen J, Wu F, Guo S. 2013. Specificity and preference of mycorrhizal associations in two species of the genus Dendrobium (Orchidaceae). Mycorrhiza 23: 4, 317-324. Yuan L, Yang Z, Li S, Hu H, Huang J. 2010. Mycorrhizal specificity, preference, and plasticity of six slipper orchids from South Western China. Mycorrhiza 20: 8, 559- 568. Zettler LW, Hofer CJ. 1998. Propagation of the little club-spur orchid (Platanthera clavellata) by symbiotic seed germination and its ecological implications. Environmental and Experimental Botany 39: 189-195.

91 Chapter 4. Supplementary information

Supplementary information

from Epidendrum rhopalostele 88 C036 from Epidendrum rhopalostele 88 C033 from Epidendrum rhopalostele 88 C039 from Epidendrum rhopalostele 88 C035 from Epidendrum rhopalostele 59 C029 from Epidendrum rhopalostele 88 C034 from Epidendrum rhopalostele 88 C038 from Epidendrum rhopalostele 88 C037 from Epidendrum rhopalostele 59 C032 from Epidendrum rhopalostele 59 C027 from Epidendrum rhopalostele 57 C021 from Epidendrum rhopalostele 59 C026 from Epidendrum rhopalostele 59 C028 from Epidendrum rhopalostele 59 C031 from Epidendrum rhopalostele 59 C030 from Epidendrum rhopalostele 59 C025 from Epidendrum rhopalostele 198 C120 from Epidendrum rhopalostele 103 C055 from Epipendrum rhopalostele 103 C052 from Epidendrum rhopalostele 103 C048 from Epidendrum rhopalostele 103 C051 from Epidendrum rhopalostele 227 C125 from Epidendrum rhopalostele 198 C118 from Epidendrum rhopalostele 34 C003 from Epidendrum rhopalostele 34 C006 from Epidendrum rhopalostele 34 C005 from Epidendrum rhopalostele 111 C057 from Epidendrum rhopalostele 111 C062 from Epidendrum rhopalostele 111 C061 from Epidendrum rhopalostele 103 C049 from Epidendrum rhopalostele 103 C053 from Epidendrum rhopalostele 119 C067 from Epidendrum rhopalostele 119 C068 from Epidendrum rhopalostele 119 C065 from Epidendrum rhopalostele 119 C066 from Epidendrum rhopalostele 119 C071 from Epidendrum rhopalostele 253 C142 from Epidendrum rhopalostele 151 C094 from Epidendrum rhopalostele 151 C097 from Epidendrum rhopalostele 151 C098 from Epidendrum rhopalostele 151 C099 from Epidendrum rhopalostele 151 C095 from Epidendrum rhopalostele 151 C096 from Epidendrum rhopalostele 151 C093 from Epidendrum rhopalostele 34 C001 from Epidendrum rhopalostele 253 C138

-/95/99 from Epidendrum rhopalostele 253 C139 from Epidendrum rhopalostele 241 C132 from Epidendrum rhopalostele 241 C134 from Epidendrum rhopalostele 103 C054 from Epidendrum rhopalostele 111 C063 from Epidendrum rhopalostele 111 C056 from Epidendrum rhopalostele 119 C069 from Epidendrum rhopalostele 241 C131 from Epidendrum rhopalostele 147 C091 from Epidendrum rhopalostele 103 C050 from Stelis concinna DQ178111 from Stelis concinna DQ178110 from Stelis concinna DQ178109 from Pleurothalis lilijae DQ178098 from Stelis concinna DQ178108 from Stelis superbiens DQ178105 from Stelis halli DQ178103 from Stelis halli DQ178104 from Stelis concinna DQ178084 from Stelis superbiens DQ178101 from Stelis concinna DQ178092 97/94/82 from Pleurothalis lilijae DQ178100 from Stelis halli DQ178114

94/92/82 from Stelis halli DQ178113 from Stelis superbiens DQ178115 Tulasnella danica AY373297 Tulasnella calospora AY373298 Tulasnella calospora DQ388042 Tulasnella calospora DQ388045 Tulasnella calospora EU218888 2

75/-/55 Tulasnella deliquescens AY373291 Tulasnella calospora DQ388041 from Spathaglotis plicata AJ313450 Tulasnella calospora DQ388043 Tulasnella calospora DQ388044 Tulasnella calospora EF393621 Tulasnella bifrons AY373290 94/50/- from Goodyera pubescens AY373264 from Goodyera pubescens AY373270 from Goodyera pubescens AY373274 from Arundina graminifolia AJ313435 from Spathaglotis plicata AJ313452 from Spathaglotis plicata AJ313439 from Spathaglotis plicata AJ313453 from Spathaglotis plicata AJ313437 from Vanda miss Joaquin AJ313443 from Epipactis gigantea AY634124 from Dendrobium crumenatun AJ313438 Tulasnella asymmetrica DQ520101

100/100/100 Tulasnella asymmetrica DQ520094 Tulasnella asymmetrica DQ520101 Tulasnella asymmetrica DQ388041 from Epidendrum rhopalostele 91 C045 from Epidendrum rhopalostele 91 C043 from Epidendrum rhopalostele 91 C042 from Epidendrum rhopalostele 119 C070 from Epidendrum rhopalostele 135 C072 from Epidendrum rhopalostele 135 C073 from Epidendrum rhopalostele 191 C116 88/-/88 from Epidendrum rhopalostele 57 C023 from Epidendrum rhopalostele 135 C077 from Epidendrum rhopalostele 206 C083 from Epidendrum rhopalostele 91 C044 from Epidendrum rhopalostele 147 C085 from Epidendrum rhopalostele 135 C079 from Epidendrum rhopalostele 39 C014 from Epidendrum rhopalostele 206 C081 from Epidendrum rhopalostele 206 C080 -/88/55 from Epidendrum rhopalostele 171 C105 from Epidendrum rhopalostele 171 C101 from Epidendrum rhopalostele 171 C103 from Epidendrum rhopalostele 171 C106 62/-/- from Epidendrum rhopalostele 171 C108 from Epidendrum rhopalostele 171 C104 88/50 from Epidendrum rhopalostele 191 C110 from Epidendrum rhopalostele 191 C113 from Epidendrum rhopalostele 191 C112 from Epidendrum rhopalostele 191 C114 from Epidendrum rhopalostele 191 C109 from Epidendrum rhopalostele 227 C127 from Epidendrum rhopalostele 171 C107 from Epidendrum rhopalostele 171 C102 from Epidendrum rhopalostele 135 C076 from Epidendrum rhopalostele 135 C075 from Epidendrum rhopalostele 119 C064 Tulasnella asymmetrica DQ388047

100/65/71 Tulasnella albida AY373294 100/63/79 94/68/70 Tulasnella albida AY373294 Tulasnella violea AY373303 Tulasnella pruinosa AY373295 60/-/- 65/93/94 Tulasnella pruinosa AY373295

99/100/99 from Pleurothallis lilijae DQ178070

100/87/95 from Pleurothallis lilijae DQ178069

85/-/- 100/100/100 from Platanthera chlorantha AY634131 from Tipularia discolor AY373310 from Cryptothallus mirabilis AY192482 -/100/100 from Pinus muricata AY192461 from Cryptothallus mirabilis AY192483 from Stelis halli DQ178072 100/100/- from Stelis halli DQ178071 from Setlis halli DQ178073 from Stelis concinna DQ178075 from Stelis concinna DQ178076 from Stelis superbiens DQ178074

100/98/100 from Pleurothallis lilijae DQ178116 from Stelis superbiens DQ178117 Tulasnella violea DQ520097 Tulasnella tomaculum AY373296 -/100/100 Tulasnella violea AY373293 Tulasnella eichleriana AY373292 Multiclavula corynoides U66440 0.005 substitutions/site Appendix 4.1. DNA alignment of Epidendrum rhopalostele mycobionts. Tulasnellaceae sequences correspond to the 5.8S region.

92 CONCLUSIONES

CONCLUSIONES

Los resultados obtenidos en los distintos capítulos permiten proponer las siguientes conclusiones. Es importante hacer notar que la mayoría están referidas a la población objeto de estudio y, por tanto, su posible extrapolación debería realizarse teniendo en cuenta este hecho. 1. El estudio inicial de prospección mostró diferencias en los patrones de presencia y abundancia de las orquídeas Masdevallia, Lepanthes, Pleurothallis y Epidendrum; mientras que para los dos primeros géneros la distribución entre los árboles responde a un patrón aleatorio, para Pleurothallis y Epidendrum es de tipo agregado. 2. La capacidad para colonizar nuevos árboles por parte de Epidendrum rhopalostele es limitada, habida cuenta de la baja proporción de árboles ocupados y la distribución agrupada de los forófitos inferida del ajuste de los puntos a un modelo Poisson agrupado. 3. Epidendrum rhopalostele muestra preferencia por los árboles muertos y, en menor medida, por Clusia alata. Su mayor presencia en árboles de carácter umbrófilo y el hecho de ocupar la parte inferior de los troncos indican una preferencia por los microhábitats de baja luminosidad y elevada humedad. La presencia de esta orquídea también parece estar favorecida por las condiciones que ofrecen los árboles de corteza rugosa. Por el contrario, el grosor de los árboles no parece ser un factor condicionante. 4. La distribución espacial de E. rhopalostele se ajusta a un modelo Poisson agrupado sugiriendo que la dispersión de semillas se produce principalmente a cortas distancias (dentro de un mismo árbol y entre árboles próximos). La ausencia de un patrón de agregación de individuos juveniles en torno a individuos adultos no apoya la hipótesis de facilitación de adultos hacia juveniles. 5. La diversidad genética de la población se encuentra estructurada en dos grupos bien diferenciados que podrían corresponder a individuos puros de E. rhopalostele y a individuos de origen híbrido entre E. rhopalostele y Epidendrum madsenii o Epidendrum falcisepalum al haberse encontrado individuos de ambos grupos en un mismo árbol. Con el fin de confirmar la existencia de procesos de hibridación, sería recomendable secuenciar las regiones ITS de las otras dos especies de Epidendrum y cloroplásticas para compararlas con las de algún individuo de los grupos 1 y 2. 6. Dentro de cada grupo genético se ha detectado un patrón de aislamiento por distancia compatible con una limitación en la dispersión de semillas. Este resultado apoyaría la hipótesis de una dispersión facilitada por el agua de lluvia en favor de la hipótesis de dispersión por viento. 7. Epidendrum rhopalostele no presenta una especificidad absoluta en la relación orquídea-hongo micorrízico ya que al menos se ha encontrado

93 Chapter 4. CONCLUSIONES

asociada a dos especies de Tulasnella. El hecho de haber encontrado en un mismo individuo adulto las dos especies apoyaría la hipótesis de sucesivas reinfecciones. 8. Desde una perspectiva de conservación, las estrategias a desarrollar deben: a) favorecer la disponibilidad de micro-sitios con características como: árboles muertos, árboles de corteza rugosa, árboles tolerantes a la sombra, así como ejemplares de Clusia alata; b) contemplar la traslocación de individuos en la parte inferior del tronco (< 1 m del suelo) donde exista menor luminosidad y mayor humedad ambiental; c) en el caso de traslocar plantas sobre un forófito, colocar las plantas traslocadas dentro de un radio de 0,4 m de los ejemplares existentes para favorecer la presencia de unas condiciones microambientales adecuadas; d) disponer los árboles susceptibles de alojar plantas de E. rhopalostele a una distancia media en torno a 2.1 m para facilitar la colonización de nuevos árboles por parte de la orquídea.