ECOLE NATIONALE SUPÉRIEURE AGRONOMIQUE DE MONTPELLIER
MONTPELLIER SUPAGRO
THÈSE
Pour l’obtention du grade de
DOCTEUR EN SCIENCES AGRONOMIQUES
Ecole Doctorale: Biologie des Systèmes Intégrés, Agronomie – Environnement.
Discipline: Ressources Phytogénétiques et Interactions Biologiques.
présentée et soutenue publiquement
Par
John OCAMPO PÉREZ
le 16 avril 2007
Titre :
Étude la diversité génétique du genre Passiflora L. (Passifloraceae) et de sa
distribution en Colombie
Study of the genetic diversity of genus Passiflora L. (Passifloraceae) and its
distribution in Colombia
JURY
Dr. Rosemary GILLESPIE, Professeur, Université de Californie, Berkeley, U.S.A.
Dr. André CHARRIER, Professeur, Montpellier SupAgro, France
Dr. Laure CIVEYREL, Maître de conférence, U. Paul Sabatier, Toulouse, France
Dr. Daniel DEBOUCK, Directeur Unité Ressources Génétiques, CIAT, Colombie
Dr. Geo COPPENS d’EECKENBRUGGE, CIRAD, Montpellier, France
Dr. Philippe FELDMANN, CIRAD, Montpellier, France
Examinateur
Examinateur
Rapporteur
Rapporteur
Co-Directeur
Directeur
DEDICATED TO MY FAMILY
ACKNOWLEDGEMENTS
This study was supported by grants from the Gines-Mera Fellowship Foundation
(CBN-CIAT), Bioversity International (formerly IPGRI), the Centre de Cooperation
Internationale en Recherche Agronomique pour le Développement. (CIRAD), the
International Center for Tropical Agriculture (CIAT), the Colombian Ministry for
Environment (MMA), the Research Center of the National Federation of Coffee
Growers of Colombia (Cenicafé) and the Instituto Colombiano para el Desarrollo de
la Investigación (COLCIENCIAS).
I would especially like to express my gratitude to Geo Coppens d’Eeckenbrugge, who
has played a critical role in this project from its inception. He introduced me to the
Passiflora world, and has been a wonderful advisor throughout the course of this
study. His door was always open and he constantly provided me with much needed
support and encouragement. Also, Geo and his family provided me with a place to
stay and support during visits in Montpellier.
I am particularly grateful to Drs. Philippe Feldmann (CIRAD), André Charrier
(ENSAM), Ange-Marie Risterucci (CIRAD), Laurence Pascal and Bertrand Schatz
(CNRS - CEFE), members of my committee for their scientific advices, comments
and corrections of the manuscript. Special thanks to Drs Laure Civeyrel (Université
Paul Sabatier), Daniel Debouck (CIAT) and Rosemery Gillespie (Califonia of
University) for having accepted to be part of the jury. I am deeply indebted also to
Drs. Andrew Jarvis (CIAT) and Xavier Scheldeman (Bioversity International) for
their comments and valuable suggestions regarding geographic data analysis.
I want to thank Daniel Franco and Mario Ruiz for helping in the installation of living
collections (Paraguaicito Experimental Station - Cenicafé), and to German Arroyave
and Juan G. Contreras (PASSICOL S.A) for providing the living collection facilities
for maracuja. Also, I wish to acknowledge Christian Houel (Passiflora National
Collection, Blois, France) and Doyle McKey (CNRS - CEFE) for assistance in
obtaining plant material for DNA extraction.
I am deeply indebted also to María Restrepo, Felipe Barrera, Lina Farfán and Cristian
Olaya of the Caldas University (Colombia) for assistance in collecting field data.
Thanks are also due to Mauricio Villegas (Cenicafé), Vicky Barney (Bioversity),
Odelio Soto (S.Agricultura, Tenerife), Robinsón Galindo (PN.Catatumbo), Carlos
Solarte (Cenicafé), Carolina Alcázar (Proselva), Hernando Criollo (U.Nariño), María
Giraldo (Umata, Filandia), José F. Restrepo (U. Caldas), Edgar Díaz (JB. San Jorge,
Ibagué), Sergio Ocampo (Aguas de Manizales), Camilo Palacios (PN. Guanenta),
Hector Jiménez (Umata, Fredonia), Cesar Londoño (Umata, Marsella), Rafael
Izquierdo (Bioversity), Creuci Caetano (U.Nacional), Mike Salazar (CIAT), Alvinxon
Castro (U.Chocó), Alvaro Mejia (CIAT) and Gustavo Morales (JB. José Celestino
Mutis) for accompanying me on several collecting trips. José O. Velásquez and
Segundo Pablo Guaspud (Casa Mutis) was my assistant and traveling companion
during my field trip to Mariquita (Tolima) and is a very dear friend. Their ecological
knowledge and experience in traveling around Mariquita made for a flawless trip.
I am enormously grateful also to Colombian farmers contacted in the fieldwork for
their continuous help and availability in localizing a great part of the observed plant
material.
I would like to thank the following Colombian and international herbaria for
providing me access to the specimens: Álvaro Fernández Pérez, Fundación
Universitaria de Popayán (AFP); Universidad del Cauca (CUP); Universidad
Tecnológica del Chocó (CHOCO); Centro Internacional de Agricultura Tropical
(CIAT); Instituto Amazónico de Investigaciones Científicas (SINCHI); Universidad
Nacional sede Bogotá (COL); Universidad del Valle (CUVC); Departamento de
Recursos Naturales, Universidad de Caldas (FAUC); Federico Medem,
Instituto
Alexander von Humboldt (FMD); Jardín Botánico Eloy Valenzuela (HEV);
Universidad de Antioquia (HUA), Universidad del Quindío (HUQ); Jardín Botánico
José Antonio Uribe (JAUM); Gabriel Gutiérrez (MEDEL); Universidad de Nariño
(PSO); Universidad Surcolombiana (SURCO); Universidad del Tolima (TOLI);
Universidad Industrial de Santander (UIS); José Cuatrecasas Arumi, Universidad
Nacional sede Palmira (VALLE); Muséum National d'Histoire Naturelle, Paris,
France (P); the Real Jardín Botánico, Madrid, Spain (MA); Royal Botanic Garden,
Kew, England, UK (K); Missouri Botanical Garden, USA (MO); The Natural History
Museum, Chicago (F); New York Botanical Garden, USA (NY); Marie Selby
Botanicals Gardens, USA (SEL). The three latter were visited on line.
In the office of Bioversity International (for the Americas) several people have
supported me in other aspects through this thesis in the last four years, particularly
Ana Luisa Triana, for her collaboration and friendship. Dimary Libreros, Adriana
Sánchez, Monica Macias, Angela Cardona, Elcy Lozano, Samir Patiño, Jesus Salcedo,
Margarita Baena and Emelda Usma have also shared theirs friendship and
congeniality, and helped in the logistics of my work. I want also to express my
gratitude to former director, Dr. Ramon Lastra, and the current director, Dr. Marleni
Ramírez, who have deposited in me their confidence and support.
I am especially grateful to Edgardo Alpizar (Costa Rica), Rommel Montufar
(Ecuador), Fabio Parrado and Genny (Bolivia), Gaston Loor (Ecuador), Mauricio
Soto (Colombia), and to the Venezolians Elvis Portillo, Climaco Alvarez, José
Bustamante, Douglas Rodríguez and Graciela Sepulcre for sharing immemorial
moments during my stay in Montpellier. In CIAT, I also wish to express my sincerest
thanks to Teresa Sánchez and Rosa González, who have offered me their friendship
and advices.
Finally, my thanks are due to my family for their love and support. I profoundly thank
my wife, Carolina who has been a constant source of love, patience, and
encouragement throughout the last four years and to baby who comes in way. My
parents have continually believed in me and helped me in innumerable ways in
special my mother Rocío.
SUMMARY
Given its economic importance, the characterization of genus Passiflora is seen
as a priority for Andean countries and specific strategies are needed to optimize its use
and conservation. The objective of this thesis was to study the distribution and
organization of Passifloraceae in Colombia, with a triple aim of exploring the diversity of
available genetic resources, evaluating the risk of their erosion and their potential as an
indicator group for the conservation of biodiversity, particularly in the coffee growing
zone.
Colombian
Passifloraceae were listed, gathering and georeferencing 3,930
records, for a total of 167 species. Forty-two produce an edible fruit, and nine are
commercially cultivated. Most of the 58 endemic species, including 37 narrow endemics,
are Andean species of subgenera Tacsonia and Decaloba. Applying the UICN criteria,
70% of the species appear threatened and three extinct. When compared with other
regions, the Andes of Colombia and Ecuador constitute the center of Passiflora diversity,
whose elevational distribution shows a small peak below 500 m, and two higher ones at
1000-2,000 and 2,500-3,000 m. This pattern corresponds to divergent adaptive trends
among infrageneric divisions, subgenus Tacsonia contributing markedly to the highest
peak.
The
climatic data associated with our 3,930 records allowed modeling and
summing the distribution of 80 species, so predicting the distribution of species richness.
Nine areas with very high richness, but no particular endemism, were identified in the
Andean region, three of which correspond to collection gaps as they do not appear on the
map of observed diversity. Their striking correspondence with coffee growing zone
ecotopes imposes a strategy integrating agricultural and environmental management at
the landscape level for preserving this threatened richness as well as a region of particular
importance for the country. Both aspects may be combined if Passifloraceae can be used
as an indicator of biodiversity in this region, which seems justified by their diversity and
characteristics, including multiple ecological interactions with many organisms.
Morphological
variation was studied in 124 accessions from 60 Passiflora
species and eight subgenera with 127 descriptors. Twenty-four quantitative traits showing
i
high variation among subgenera were selected. The three principal components of
variation are associated with (i) flower length; (ii) flower width and bract shape; and (iii)
peduncle branching, stem width and leaf length. The projection of accessions in the
resulting tridimensional space consistently separates subgenera. A similar selection of 32
qualitative traits, and four categorized quantitative traits, allowed classifying our sample
consistently. Most discriminating characters include size of stems and leaves, presence of
tendrils, number and distribution of extrafloral nectaries, dimensions and general shape of
bracts, width and length of flowers and corona complexity. Our results support seven of
the eight Killip’s subgenera of our sample, but no infrasubgeneric classifications.
However, the new classification of subgenus Decaloba by Feuillet & MacDougal was
partly supported. They converge on many points with previous phylogenetic results
obtained with DNA sequences, although the latter group subgenera Tacsonia and
Distephana with subgenus Passiflora.
The chloroplast and mitochondrial DNA diversity of 213 genotypes belonging
to 151 Passiflora species and 15 subgenera recognized by Killip was studied by PCRRFLP, identifying 280 haplotypes for cpDNA and 372 for mtDNA. The principal coordinate analysis on cpDNA data allowed visualizing a strong separation of subgenera
Apodogyne, Decaloba, Murucuja, Pseudomurucuja and Psilanthus (constituting the
“Decaloba group”), while the neighbor-joining cluster analysis showed three wellsupported clusters within Passiflora, corresponding to the three major divisions of the
taxonomy proposed by Feuillet & MacDougal. The first one, named the “Passiflora
group”, includes subgenera Calopathanthus, Deidamioides, Distephana, Dysosmia,
Dysosmioides, Manicata, Passiflora, Tacsonia, and Tacsonioides, with a very loose
substructure and considerable intraspecific variation. The second one includes subgenus
Astrophea, and the third is the ‘Decaloba group’. The outgroup species, take an
undefined position among the Passiflora clusters. The phenogram from mtDNA data
separates four moderately supported clusters. As for cpDNA, a first one corresponds to
the ‘Decaloba group’. The other are different, as subgenera Astrophea and
Tryphostemmatoides appear integrated within the ‘Passiflora group’, while subgenus
Tacsonia forms a uniform distinct cluster, close to another one comprising species of
Passiflora series Kermesinae, Simplicifoliae, Lobatae, and Menispermifoliae. The
ii
analyses of cpDNA and mtDNA give different pictures of the Passiflora diversity, in the
position of the outgroup, the relative position of four subgenera, and the relationships
between species, which we attribute to different rates of evolution and modes of
transmission of the chloroplastic and mitochondrial genomes, combined with reticulate
evolution in the genus.
Keywords: Passiflora L., Colombia, Andes, coffee growing zone, distribution,
biodiversity,
endemism,
morphological
mitochondria.
iii
descriptors,
PCR-RFLP,
chloroplast,
RÉSUMÉ
De par son importance économique, la caractérisation du genre Passiflora est
considérée prioritaire par les pays andins, impliquant des stratégies spécifiques pour
optimiser son utilisation et sa conservation. L’objectif de cette thèse est d’étudier la
distribution et l'organisation des Passifloraceae en Colombie, dans le triple but d'explorer
la diversité des ressources génétiques disponibles, d’évaluer le risque de leur érosion
ainsi que leur potentiel comme groupe indicateur pour la conservation de la biodiversité,
en particulier dans la région caféière.
La
liste des espèces colombiennes a été établie en rassemblant et localisant
3.930 registres, pour un total de 167 espèces, dont 42 produisent un fruit comestible et
neuf sont cultivées commercialement. La plupart des 58 espèces endémiques, y compris
37 endémiques étroites, sont des espèces andines des sous-genres Tacsonia et Decaloba.
En appliquant les critères de l'UICN, 70% des espèces semblent menacées et trois
éteintes. Par comparaison avec d'autres régions, les Andes de la Colombie et de
l'Equateur constituent le centre de diversité de Passiflora, dont la distribution altitudinale
montre un petit pic en-deçà de 500 m, et deux plus élevés, vers 1.000-2.000 et 2.5003.000 m. Ce patron reflète des tendances adaptatives divergentes entre divisions
infragénériques, le sous-genre Tacsonia contribuant nettement au dernier pic.
Les données climatiques associées à nos 3.930 registres ont permis de modéliser
la distribution de 80 espèces et de prédire la distribution de la richesse d'espèces. Neuf
zones de richesse très élevée, mais sans niveau d’endémisme particulier, ont été
identifiées dans la région andine. Trois d’entre elles n’ont pas été prospectées. La
correspondance entre ces neuf zones et des ecotopes caféiers impose une stratégie
intégrant la gestion agricole et environnementale au niveau du paysage pour préserver en
même temps cette richesse biologique et une région d'importance particulière pour le
pays. Les deux aspects peuvent être combinés si les Passifloraceae peuvent y être
employées comme indicateurs de biodiversité, ce qui semble justifié par leur diversité et
leurs caractéristiques, notamment leurs interactions écologiques avec de nombreux
organismes.
iv
La variation morphologique a été étudiée parmi 124 accessions de 60 espèces et
huit sous-genres de Passiflora, avec 127 descripteurs. Vingt-quatre traits quantitatifs
variant entre sous-genres ont été retenus. Les trois composants principaux de leur
variation sont associés à (i) la longueur de fleur ; (ii) la largeur de fleur et la forme des
bractées ; et (iii) la bifurcation du pédoncule, la largeur de tige et la longueur de feuille.
La projection des accessions dans l'espace tridimensionnel correspondant sépare bien les
sous-genres. Un tri semblable de 32 traits qualitatifs, et de quatre traits quantitatifs
catégorisés, a permis de classer notre échantillon de façon cohérente. Les caractères
distinctifs incluent la taille des tiges et des feuilles, la présence de vrilles, le nombre et la
distribution des nectaires extrafloraux, les dimensions et la forme générale des bractées,
la largeur et la longueur des fleurs et la complexité de la couronne. Nos résultats appuient
sept des huit sous-genres de Killip inclus dans l’échantillon, mais aucune classification
infrasubgénérique, sauf, pour partie, la nouvelle classification du sous-genre Decaloba
par Feuillet & MacDougal. Ils convergent sur de nombreux points avec des études
phylogénétiques à partir de séquences d'ADN, bien que celles-ci unissent les sous-genres
Tacsonia et Distephana au sous-genre Passiflora.
L’analyse, par PCR-RFLP, de l'ADN chloroplastique et mitochondrial de 213
accessions de 151 espèces de Passiflora et 15 sous-genres a permis d’identifier 280
haplotypes pour le premier et 372 pour le second. L’analyse factorielle des données
d’ADNcp sépare fortement les sous-genres Apodogyne, Decaloba, Murucuja,
Pseudomurucuja et Psilanthus (constituant le "groupe Decaloba"), tandis que la
classification par la méthode du neighbor-joining met en évidence trois groupes majeurs
dans Passiflora, correspondant aux trois divisions principales de la taxonomie proposée
par Feuillet & MacDougal. Le premier, appelé "groupe Passiflora", inclut les sous-genres
Calopathanthus, Deidamioides, Distephana, Dysosmia, Dysosmioides, Manicata,
Passiflora, Tacsonia et Tacsonioides, avec une sous-structure très lâche et une variation
intraspécifique considérable. Le second inclut le sous-genre Astrophea, et le troisième est
le "groupe Decaloba". Les espèces du groupe extérieur prennent une position indéfinie
entre les groupes de Passiflora. Le phénogramme résultant de l’analyse de l’ADNmt
sépare quatre groupes. Comme pour l’ADNcp, un premier correspond au "groupe
v
Decaloba".
Les
autres
sont
différents,
car
les
sous-genres
Astrophea
et
Tryphostemmatoides sont intégrés avec le ‘groupe Passiflora’, tandis que le sous-genre
Tacsonia forme un groupe différencié et homogène, proche d’un groupe plus varié
d’espèces des séries Kermesinae, Simplicifoliae, Lobatae et Menispermifoliae du sousgenre Passiflora. Les analyses des deux génomes, chloroplastique et mitochondrial,
donnent donc des images divergentes de la diversité de Passiflora, par la position des
groupes extérieurs, la position relative de quatre sous-genres, et les relations entre
espèces. Ces divergences semblent liées à des différences dans les taux d'évolution et
dans leur mode de transmission, en relation avec une évolution réticulée dans le genre.
Mots-clés: Passiflora L., Colombie, Andes, région caféière, distribution, biodiversité,
endémisme, descripteurs morphologiques, PCR-RFLP, chloroplastes, mitochondries.
vi
RESUMEN
Teniendo en cuenta su importancia económica, la caracterización del género
Passiflora es considerada una prioridad por los países andinos y se necesitan estrategias
específicas para optimizar su uso y conservación. El objetivo de esta tesis fue estudiar la
distribución y organización de las Passifloraceae de Colombia, con un triple objetivo de
analizar la diversidad de los recursos genéticos disponibles, evaluar el riesgo de su
erosión así como su potencial como grupo indicador para la protección de la
biodiversidad, particularmente en la zona cafetera.
Las
especies colombianas fueron inventoriadas con base en 3.930 registros,
para un total de 167 especies, de las cuales 42 producen fruto comestible y nueve son
cultivadas comercialmente. La mayoría de las 58 especies endémicas, incluyendo 37 de
distribución restringida, son principalmente andinas, de los subgéneros Tacsonia y
Decaloba. Aplicando los criterios de la UICN, 70 % de las especies aparecen en peligro y
tres son consideradas extintas. En comparación con otras regiones, los Andes de
Colombia y Ecuador constituyen el centro de diversidad de Passiflora, cuya distribución
altitudinal muestra un leve pico debajo de 500 m, y dos más marcados en los 1.0002.000 y 2.500-3.000 m. Este patrón refleja tendencias adaptativas según divisiones
infragenéricas, con particular contribución del subgénero Tacsonia al tercer pico.
Los
datos climáticos relacionados con los 3.930 registros permitieron
modelizar y sumar la distribución de 80 especies, y predecir la distribución de su
diversidad. Nueve zonas ultra-diversas fueron identificadas en la región andina, tres de
ellas subexploradas. Estas áreas no muestran un nivel de endemismo particular. La
correspondencia entre ellas y los ecotopos cafeteros impone una estrategia de
conservación que integre la gestión agrícola y el medio ambiente a escala del paisaje para
preservar al mismo tiempo esta riqueza biológica y una región de especial importancia
para el país. Ambos aspectos pueden ser combinados si se emplean las Passifloraceae
como un grupo indicador de biodiversidad en esta región. Esto es justificado por su
diversidad y características particulares, incluyendo sus múltiples interacciones con
diferentes organismos.
vii
La variación morfológica fue estudiada en 124 accesiones de 60 especies de
Passiflora, y ocho subgéneros de Killip, usando 127 descriptores. Por su diferenciación
entre subgéneros, 24 descriptores cuantitativos fueron seleccionados para un análisis de
componentes principales. Los tres primeros componentes están relacionados con (i)
longitud de la flor; (ii) ancho de la flor y forma de la bráctea; (iii) bifurcación del
pedúnculo, diámetro del tallo y longitud de la hoja. La proyección de las accesiones en un
espacio tridimensional muestra una clara separación entre subgéneros. Una selección de
32 caracteres cualitativos y cuatro cuantitativos categorizados ha permitido clasificar la
muestra de manera coherente. Los caracteres discriminantes incluyen tamaño del tallo y
la hoja, presencia de zarcillos, número y distribución de los nectarios extraflorales,
dimensiones y forma general de la bráctea, ancho y longitud de flor y complejidad de la
corona. Los resultados soportan siete de los ocho subgéneros de Killip representados en
la muestra, pero ninguna clasificación infra-subgenérica, con la excepción parcial del
subgénero Decaloba sensu Feuillet & MacDougal. El análisis converge en muchos
puntos con los estudios filogenéticos realizados a partir de secuencias de ADN, aunque
éstos agrupan los subgéneros Tacsonia y Distephana con el subgénero Passiflora.
Los
genomas cloroplástico y mitocondrial de 213 representantes de 151
especies y 15 subgéneros de Passiflora fueron estudiados por PCR - RFLP, identificando
280 haplotipos para ADNcp y 372 para ADNmt. El análisis factorial de los datos de
ADNcp permitió visualizar una fuerte separación de los subgéneros Apodogyne,
Decaloba, Murucuja, Pseudomurucuja y Psilanthus (constituyendo el "grupo
Decaloba"), mientras que la clasificación por el método de neighbor joining pone en
evidencia tres grupos bien soportados dentro de Passiflora, correspondiendo a tres
divisiones principales de la clasificación propuesta por Feuillet & MacDougal. El
primero, llamado "grupo
Passiflora", incluye los subgéneros Calopathanthus,
Deidamioides, Distephana, Dysosmia, Dysosmioides, Manicata, Passiflora, Tacsonia, y
Tacsonioides, con una débil subestructura y una diferenciación intra-específica
considerable. El segundo incluye el subgénero Astrophea, y el tercero el "grupo
Decaloba". Las especies del grupo externo toman una posición indeterminada entre los
grupos de Passiflora. El fenograma del análisis de ADNmt separa cuatro grupos
viii
parcialmente soportados, uno de ellos correspondiendo al mismo "grupo Decaloba" del
ADNcp. Los subgéneros Astrophea y Tryphostemmatoides aparecen integrados dentro
del "grupo Passiflora", mientras que el subgénero Tacsonia constituye un grupo distinto
y homogéneo, cerca de otro grupo que incluye especies de las series Kermesinae,
Simplicifoliae, Lobatae, y Menispermifoliae del subgénero Passiflora. Así, los análisis de
los dos genomas, cloroplástico y mitocondrial, muestran imágenes divergentes de la
diversidad de Passiflora, esencialmente por la posición del grupo externo, la posición
relativa de los cuatro subgéneros, y las relaciones entre especies. Estas divergencias
parecen ligadas a las diferencias en las tasas de evolución y el modo de transmisión de los
genomas, así como a una evolución reticulada en el género.
Palabras clave: Passiflora L., Colombia, Andes, zona cafetera, distribución,
biodiversidad,
endemismo,
descriptores
morfológicos,
mitocondria.
ix
PCR-RFLP,
cloroplasto,
OBJECTIVES
General objectives
To study the distribution and organization of the diversity of the Passifloraceae of
Colombia, with a triple aim of exploring the diversity of available genetic
resources, evaluating the risk of their erosion and their potential as an indicator
group for the conservation of biodiversity, particularly in the Colombian coffee
growing zone.
Specific objectives
•
To map the distribution and the diversity of the Passifloraceae species of
Colombia using a Geographic Information System (GIS).
•
To characterize the morphologic and molecular diversity at inter- and intraspecific levels.
•
To evaluate the potential of Passiflora as an indicator group to evaluate the risks
of biodiversity erosion and take them into account in the development of
strategies for in situ genetic resources management and the conservation of the
corresponding natural habitats.
x
TABLE OF CONTENTS
Summary……………………………………………………………………………..
i
Résumé………………………………………………………………………………..
iv
Resumen………………………………………………………………………………
vii
Objectives……………………………………………………………………………..
x
Table of contents……………………………………………………………………..
xi
List of tables…………………………………………………………………………..
xvii
List of figures…………………………………………………………………………
xix
_______CHAPTER I
GENERAL INTRODUCTION………...……………………………………………
1
I.1. Introduction……………………………………………………………………...
2
I.1.1. History………………………………………………………………………….
2
I.1.2. Taxonomy and general distribution…………………………..………………
3
I.1.3. Botany…………………………………………………………………………..
4
I.1.4. Biology of reproduction……………………………………………………….
6
I.1.5. Uses……………………………………………………………………………..
10
I.1.6. Cultivated species……………………………………………………………...
12
I.1.7. Breeding………………………………………………………………………...
13
I.1.8. Passiflora research and prospects, in Colombia and in the Andean region..
15
I.1.9. Structure of the thesis……………………………………………..…………..
17
_______CHAPTER II
DIVERSITY OF COLOMBIAN PASSIFLORACEAE: BIOGEOGRAPHY
AND AN UPDATED LIST FOR CONSERVATION…..…....................................
18
II.1. Diversity of Colombian Passifloraceae: biogeography and an updated list
for conservation………………………………………………………………………
19
II.1.1 Abstract…..……………………………………………………………………
19
xi
II.1.2. Introduction…..………………………………………………………………
20
II.1.3. Materials and methods……….………………………………………………
24
II.1.3.1. Study area..............................................................................................
24
II.I.3.2. Herbarium and literature data…….…………………………………..
24
II.I.3.3. Expeditions and samples collected…………………………………….
25
II.I.3.4. Threat status of Passifloraceae…..…..………………………………...
25
II.1.4. Results…………………..……………………………………………………..
26
II.1.4.1. Data collecting……………………………………………..…………
26
II.1.4.2. Distribution of species richness…………………………..…………...
26
II.1.4.3. New Passifloraceae checklist for Colombia…………….…………….
28
II.1.4.4. Endemism………….…………………………………………………..
32
II.1.4.5. Threatened species………………….…………………………………
34
II.1.5. Discussion……………..………………………………………………………
35
II.1.6. Conclusions……………………..……………………………………………..
64
II.1.7. Acknowledgements……….…………………………………………………..
64
II.1.8. Appendix 1. Synonymy …………….………..………………………………..
65
_________CHAPTER III
DISTRIBUTION, DIVERSITY AND IN SITU CONSERVATION OF
COLOMBIAN PASSIFLORACEAE…………………….........................................
67
III.1. Distribution, diversity and in situ conservation of Colombian
Passifloraceae...............................................................................................................
68
III.1.2. Abstract............................................................................................................
68
III.1.2. Introduction.....................................................................................................
69
III.1.3. Materials and methods....................................................................................
73
III.1.3.1. Geography and climate.........................................................................
73
III.1.3.2. Herbarium, field and literature data.....................................................
75
III.1.3.3. Species distribution and richness..........................................................
75
III.1.3.4. Description of climatic preferences......................................................
75
III.1.3.5. Potential species distribution................................................................
76
xii
III.1.4. Results and discussion.………….………….……………………………….
77
III.1.4.1. Distribution of observations.................................................................
77
III.1.4.2. Species richness….……..………………………..…………………...
80
III.1.4.3. Species diversity….…………………………………………………..
81
III.1.4.4. Distribution by altitude.........................................................................
81
III.1.4.5. Climatic requirements.…………..……………………………………
84
III.1.4.6. Areas of distribution - endemic species................................................
86
III.1.4.7. Potential distribution of species and species assemblages….……….
94
III.1.4.8. Conservation of Passifloraceae and their habitat.…………………..
96
III.1.4.9. Passifloraceae as indicators of biodiversity…….……………………
98
III.1.5. Conclusions…….…………………………………………………………….
100
III.1.6. Acknowledgements..........................................................................................
101
_______CHAPTER IV
A PHENETIC ANALYSIS OF MORPHOLOGICAL DIVERSITY IN THE
GENUS Passiflora L………………………………………………………………….
102
IV.1. A phenetic analysis of morphological diversity in the genus Passiflora L…
103
IV.1.1. Abstract……………………..……………………………………………......
103
IV.1.2. Introduction……..……………………………………………………….......
104
IV.1.3. Materials and methods……………………………………………………...
110
IV.1.3.1. Plant materials………………………………………………………...
110
IV.1.3.2. Data collection………………………………………………………...
112
IV.1.3.3. Analyses of quantitative variation…….…..…………………………...
112
IV.1.3.4. Cluster analyses on qualitative data………………………………….
119
IV.1.4. Results and Discussion……………………………………………………....
121
IV.1.4.1. Quantitative variation………………………………………………....
121
IV.1.4.2. Correlations and principal components analysis (PCA)……………...
123
IV.1.4.3. Qualitative variation among and within subgenera…………………..
126
IV.1.4.4. Cluster analysis on the reduced descriptor list………………………..
129
IV.1.4.5. Cluster analysis on the global descriptor dataset……………………..
133
xiii
IV.1.4.6. The “Passiflora cluster”…………………………………………….....
133
IV.1.4.7. The “Tacsonia cluster”……………………………………………......
134
IV.1.4.8. The “Decaloba cluster”…………………………………………….....
135
IV.1.4.9. The “Astrophea cluster”……………………………………………....
137
IV.1.4.10. Morphological and molecular diversity……………………………...
141
IV.1.5. Conclusions…………………………………………………………………..
145
IV.1.6 Acknowledgements…………………………………………………………..
147
_____ __CHAPTER V
CHLOROPLAST AND MITOCHONDRIAL DNA VARIATION IN THE
GENUS Passiflora L. (PASSIFLORACEAE) AS REVEALED BY PCR-RFLP...
148
V.1. Chloroplast and mitochondrial DNA variation in the genus Passiflora L.
(Passifloraceae) as revealed by PCR-RFLP………………………………………...
149
V.1.1. Abstract…..........................................................................................................
149
V.1.2. Introduction…………………………………………………………………...
150
V.1.3. Materials and methods.....................................................................................
157
V.1.3.1. Taxon sampling………………………………………………………
157
V.1.3.2. DNA Extraction and PCR-RFLP analyses…………………………...
162
V.1.3.3. Data analysis…………………………………………………………
164
V.1.4. Results…………………………………………………………………………
165
V.1.4.1. PCR amplification……………………………………………………
165
V.1.4.2. Restriction analysis…………………………………………………..
165
V.1.4.3. PCR-RFLP haplotypes……………………………………………….
167
V.1.4.4. Principal co-ordinates analysis……………………………………...
170
V.1.4.5. Cluster analysis………………………………………………………
170
V.1.5. Discussion……………………………………………………………………..
181
V.1.5.1. Chloroplast DNA diversity…………………………………………..
181
V.1.5.2. Mitonchondrial DNA diversity……………………………………….
183
V.1.5.3. Divergences in the evolutions of chloroplast and mitochondrial
genomes………………………………………………………………………..
xiv
184
V.1.5.4. Diversity of organellar genomes and Passiflora systematics………...
187
V.1.6. Conclusions……………………………………………………………………
191
V.1.7. Acknowledgements............................................................................................
192
General discussion……………………………………………………………………
193
1. Discusion……………………………………………………………………….
194
1.1. Biogeography and conservation………………………………………..
194
1.2. Morphologycal and molecular diversity………………………………..
195
1.3. Importance of reticulate evolution in Passiflora……………………….
197
1.4. Phylogeography………………………………………………………...
199
Conclusions and futeres prospect…………………………………………………...
201
1. Conclusions……………………………………………………………...
202
2. Futures prospect………………………………………………………..
204
Bibliography.................................................................................................................
206
Congress communications..........................................................................................
230
Annexes.........................................................................................................................
233
Annex 1. Infrageneric classification according to Killip (1938) with emends
of Escobar (1988, 1989) and MacDougal (1994)…………………..…………
234
Annex 2. Infrageneric classification according to Feuillet & MacDougal
(2003)………………………………………………………………………….
236
Annex 3. Species cultivated in Colombia………………….………………….
238
A.3.1. Passiflora edulis Sims……………………………………………...
238
A.3.2. Passiflora ligularis Juss……………………………………………
239
A.3.3. Passiflora tripartita var. mollissima (Kunth) Holm-Niel. & Jørg....
240
A.3.4. Passiflora tarminiana Coppens & Barney........................................
240
A.3.5. Passiflora quadrangularis L……………….…………………........
242
A.3.6. Passiflora maliformis L…………………………………………....
242
A.3.7. Passiflora alata Curtis......................................................................
242
A.3.8. Passiflora popenovii Killip………………………………………...
243
Annex 4. Passiflora molecular diversity. Dendrograms obtained in previous
studies.................................................................................................................
xv
245
A.4a. Phylogenetic tree sensu Muschner et al. (2003)…………………...
245
A.4b. Phylogenetic tree sensu Yockteng (2003).………………………...
246
A.4c. Phylogenetic tree sensu Yockteng & Nadot (2004)……………….
247
A.4d. Phylogenetic tree sensu Hansen et al. (2006)……………………..
248
Annex 5. List of morphological descriptors in the genus Passiflora L………
249
xvi
LIST OF TABLES
______CHAPTER I
Table 1: Breeding objectives in passion fruits…………..…………………………..
15
__CHAPTER II
Table 1. Distribution of Passifloraceae by biogeographic region: number of species
(bold) and proportion of shared species…………........................................
28
Table 2. Number of observations and species of Passifloraceae in the 32
Colombian departments…………………………………………………….
31
Table 3. Number of Passifloraceae species in Colombia and the Neotropics……….
32
Table 4. List of 167 Passifloraceae species of Colombia……………………………
41
_____CHAPTER III
Table 1. Number of observations, species, rare and endemic Passifloraceae species
by Colombian division. Source for department areas………………….......
79
Table 2. Factor loadings, eigenvalues and percentages of variance for the first four
components, resulting from the PCA analysis on 19 bioclimatic
parameters for the 3,930 collection points (Colombian Passifloraceae)…...
85
Table 3. Total number of Passifloraceae present in Colombia. Number of
observations, Maximum distance (MaxD) and Circular area (CA) for each
species...........................................................................................................
89
_____CHAPTER IV
Table 1. List of accessions used in the present study. Taxonomy according to Killip
(1938) and emends by Escobar (1988, 1989) and MacDougal (1994).........
113
Table 2. List of 127 descriptors used in the morphological characterization study....
119
Table 3. Mean values and coefficients of variation for the whole sample and for the
different subgenera…………………………………………………………
122
Table 4. Factor loadings from principal component analysis (varimax normalized
rotation) on 24 quantitative descriptors……..…………………………...
xvii
125
Table 5. Variation for 32 qualitative and four categorized quantitative descriptors
in the different subgenera sampled…..…………………………………..
128
_____CHAPTER V
Table 1. List of species used in this study according to classification by Killip
(1938), Escobar (1988) and MacDougal (1994)…………………………..
157
Table 2. DNA sequence and type of primer pairs used in the present study by
Demesure et al. (1995)…………………………………………………….
163
Table 3. Numbers of haplotypes and fragments for each combination primer/
enzyme…………………………………………………………………….
165
Table 4. Global distribution of the haplotypes among the genera and subgenera
studied……………………………………………………………………..
xviii
169
LIST OF FIGURES
___ _CHAPTER I
Figure 1. First drawings of Passiflora representing the passion of Christ…………
2
Figure 2. Distribution of genus Passiflora in the world……………………………
3
Figure 3. Floral elements in the subgenera Passiflora and Tacsonia………………
5
Figure 4. Passiflora pollinators…………………………………………………….
8
Figure 5. Flower of Passiflora (probably P. ligularis) made in gold by a
precolombian goldsmith of Malagana culture in Colombia….……....…
12
_____CHAPTER II
Figure 1. Map of distribution of Passifloraceae specimens for 3,930 collections on
five biogeographic regions in Colombia. Points on the maps represent
sites of collection……………………………………………………….
27
Figure 2. Diagram comparing the similarity in contribution of Passifloraceae
species to the floras of the Colombian biogeographic regions (Jaccard
distance)…………………………………………………………………
28
Figure 3. Colombian endemic species……………………………………………...
33
Figure 4. Percentual number of the threat status of 165 Passifloraceae native
species under the IUCN criteria…………………….…………………..
34
Figure 5. Distribution of species richness of Passifloraceae in American countries.
38
CHAPTER III
Figure 1. Colombia’s geopolitical division in 32 departments and biogeographic
division in five regions………………………………………………….
74
Figure 2. Collection localities of Passifloraceae specimens used in this study,
among Colombian departments. Points on the maps represent sites of
collection……………………………………………………………….
78
Figure 3. Species richness observed for Passifloraceae in 1x1 km grid cells in
Colombia (165 species). Points on the maps represent sites of
collection………………………………………………………………..
xix
82
Figure 4. Distribution of total species richness (within circles) and species relative
diversity in relation to altitude in Colombia (3,930 observations), for
Passiflora and five infrageneric groups………………………………...
84
Figure 5. Distribution of Passifloraceae species in the Principal plane for climatic
variables, with indication of genera (Ancistrothyrsus and Dilkea) and
subgenera of genus Passiflora….............………………………………
86
Figure 6. Extent of Passifloraceae species distribution in Colombia: circular area
(CA50) vs. maximum distance (MaxD)…………………..……………..
88
Figure 7. Modeled distribution of Colombian Passifloraceae diversity based on
data from 80 species presenting more than 10 observations. Ellipses
individualize high richness spots mentioned in the text………………..
95
Figure 8. Distribution of protected areas in Colombia, showing poor
correspondence with areas of high Passifloraceae diversity……………
97
Figure 9. Correspondence between Passifloraceae high richness spots and coffee
growing zone ecotopes………………………………………………….
99
______CHAPTER IV
Figure 1. Schema of a flowering branch of Passiflora vitifolia Kunth ………..…...
105
Figure 2. Variation in shape and color among species from nine of Killip’s
subgenera…..............................................................................................
111
Figure 3. Relative variance components for 57 quantitative descriptors. Bold
characters are used for traits displaying more than 50% of variance
among subgenera………….…………………………………………….
123
Figure 4. Tridimensional plot of the scores of Passiflora accessions for the first
three components of quantitative variation. Colors refer to subgeneric
classification……………………………………………………...............................
Figure 5. Dendrogram obtained with first set of qualitative data.............................
Figure 6. Dendrogram obtained on complete set of qualitative data….…………....
126
132
137
Figure 6a. First part of the dendrogram obtained on complet set of qualitative
data…….………………………………………………………………..
Figure 6b. Second part of the dendrogram obtained on complet set of qualitative
xx
138
139
data……………………………………………………………………...
Figure 6c. Third part of the dendrogram obtained on complet set of qualitative
data……………………………………………………………….……..
140
Figure 7. Morphological affinity between typical representatives of P. rubra (a)
and P. capsularis (b). Accessions from Colombia (a, Calarcá, Quindío
– b, Cartago, Valle del Cauca) ................................................................
141
_____ _CHAPTER V
Figure 1. General schema of the PCR-RFLP markers……………………………..
164
Figure 2. Interspecific variation for cpDNA (a) and mtDNA (b)………………….
166
Figure 3. Principal co-ordinates on cpDNA data (PC1-PC2 and TS1-TS2 regions)
estimated with 268 CAPS markers…….…………………….................
171
Figure 4. Phenogram derived from on cpDNA (PC1-PC2 and TS1-TS2 regions)
data illustrating the distribution of the different Passiflora subgenera
studied………………………………….……………………………….. 173
Figure 4a. Cluster analysis on cpDNA data, ‘Decaloba group’, subgenera
Astrophea and Tryphostemmatoides, and outgroup genera.………….… 174
Figure 4b. Cluster analysis on cpDNA data, ‘Passiflora group’. Subgenus
Tacsonia in clear-brown. ……………………………………...………..
175
Figure 5. Phenogram derived from on mtDNA (N41-N42 and N1B-N1C regions)
data illustrating the distribution of the different Passiflora subgenera
studied....................................................................................................... 178
Fugure 5a. Cluster analysis on mtDNA data, ‘Passiflora group’ and subgenera
Astrophea and Tryphostemmatoides……………………………………. 179
Figure 5b. Cluster analysis on mtDNA data, ‘Decaloba group’…………………...
180
Figure 6. Probable relationships among main haploid numbers known in
Passiflora subgenera and other Passifloraceae genera as proposed by
De Melo et al., (2001)…………..………………………………………. 191
________ _ANNEXES
Figure 1. Species cultivated in Colombia: (a) P. edulis f. flavicarpa; (b) P. edulis 241
xxi
f. edulis; (c) P. ligularis; (d) P. tripartita var. mollissima;
(e) P. tarminiana………………………………………..………………
Figure 2. Species cultivated in Colombia: (a) P. quadrangularis; (b) P. alata; (c)
P. maliformis; (d) P. popenovii……………………………....................
244
All rights reserved
Front cover: Passiflora trinervia (Juss.) pollinated by a sword-billed hummingbird
(Ensifera ensifera Boissoneau) in the Colombian Andes (drawing by Jesus Salcedo).
Photographs by:
John Ocampo
Passiflora alata Curtis, P. antioquiensis Karst., P. arborea Spreng., P. bogotensis Benth, P.
caerulea L., P. capsularis L., P. coriacea Juss., P. edulis f. edulis Sims, P. edulis f. flavicarpa
Degener, P. emarginata Humb. & Bonpl., P. erytrophylla Mast., P. flexipes Triana & Planch., P.
foetida, L., P. lanata (Juss.) Poir., P. linearistipula Escobar, P. magdalenae Triana & Planch., P.
maliformis L., P. manicata (Juss) Pers., P. parritae (Mast.) Bailey, P. popenovii Killip, P.
quadrangularis L., P. rubra L., P. sphaerocarpa Triana & Planch., P. tarminiana Coppens &
Barney, P. tenerifensis Escobar, P. trinervia (Juss.) Poir., P. tripartita var. mollissima HolmNielsen & Jørgensen, P. vitifolia Kunth.
Alvinxon Castro
Passiflora arbelaezii Uribe
Geo Coppens d’Eeckenbrugge
Passiflora ligularis Juss.
Gustavo Morales
Passiflora longipes Juss.
xxii
CHAPTER I
____________________________________
General introduction
Chapter I: General introduction
_______________________________________________________________________
I.1. Introduction
I.1.1. History
Some plants have received attention from man because they are important to him as
sources of food, shelter, medicine, or even narcotics. But the first time passion flowers
caught the attention of Europeans, it was for another reason, for the Spaniards who first
encountered these plants in the New World in the 16th century, saw in them the elements
of the passion of Christ, and a sign that the New World would successfully be converted
to Christianity (Uribe 1955a). This religious symbolism gave these plants their common
name of Flos Passionis, or “passion flowers”. The Spanish Jesuit Juan Romero presented
the first drawing to Pope Paul V (Camollo Borgense) in 1608. A few years later, many
similar drawings were made available to a wider audience in Italy and Germany (Kugler
& King, 2004). The botanical features in these drawings were transformed to support the
religious interpretation very explicitly (Figure 1).
Figure 1. First drawings of Passiflora representing the passion of Christ: left, by Dominican monk Simone
Parlasca's 1609; right, by Eugenio Petrelli in 1610 (http://www.flwildflowers.com/passiflora.html).
2
Chapter I: General introduction
_______________________________________________________________________
I.1.2. Taxonomy and general distribution
The family Passifloraceae is placed in the order Malpighiales (Judd et al., 2002), and
divided in two tribes, Paropsieae and Passifloreae, with ca. 630 species distributed
around the world. Of the 20 currently recognized genera, four are present in America
(Ancistrothyrsus, Dilkea, Mitostemma and Passiflora). Passiflora is the largest one,
comprising approximately 525 species, mostly distributed in the tropical regions of
America, from the coastal zones up to 3,800 m in Andean páramos (Holm-Nielsen et al.,
1988). Only 22 species grow in the Old World, in the tropical and sub-tropical regions of
Southeast Asia and the Austral Pacific (Figure 2).
Figure 2. Distribution of genus Passiflora in the world (by John Ocampo).
According to Killip (1938), the genus Passiflora was created by Linnaeus in 1753, who
described 24 species in his Species Plantarum, a number increased to 35 by Lamarck
(1789). The first extensive monograph of the family was published by Cavanilles in 1790
with 43 species treated. He was followed by authors like Jussieu (1805), De Candolle
(1822, 1828), Masters (1872, 1877), and Harms (1898, 1925), who described about 250
species divided into 21 sections (Killip, 1938). In his monograph of 1938, The American
Species of Passifloraceae, Killip made the most extensive description of New World
species, classifying 355 species into four and 22 subgenera, on the basis of floral
3
Chapter I: General introduction
_______________________________________________________________________
morphology (Annex 1). In Colombia, the priest Uribe (1954, 1955a, 1955b, 1957, 1958,
1972) described several new species and Escobar (1986, 1987, 1988a, 1988b, 1989, 1990,
1990 inedited, 1994) revised subgenera Distephana, Manicata (syn. Granadillastrum),
Rathea and Tacsonia, including Tacsoniopsis in the latter, and described one additional
subgenus, Porphyropathanthus. She passed away in 1993, leaving an inedited document
on her revision of subgenus Astrophea. In the last decade, MacDougal and Feuillet have
published many papers including the description of about 15 new species, mainly of the
subgenera Decaloba and Astrophea (MacDougal, 1992, 1994, 2006; Feuillet, 2002,
2004). Recently, Feuillet & MacDougal (2003; Annex 2) proposed a new infrageneric
classification in Passiflora. According to their proposal, based on morphological
characters, four subgenera would be recognized: Astrophea and Deidamioides, from
South and Central America, Decaloba, from America, Southeast Asia and Australia, and
Passiflora, exclusively from America (Ulmer & MacDougal, 2004). Additionally, they
downgraded genus Tetrastylis as a section of subgenus Deidamioides.
I.1.3. Botany
The plants of genus Passiflora are mostly climbers with herbaceous or woody stems and
axillary tendrils, or very rarely arbustive or arborescent. Their leaves are alternate,
generally simple, entire, lobed or palmate. Stipules are generally present at the base of
petioles; the tendrils are also axillary, arising from sterile pedicels. Passion flowers are
bisexual or unisexual, regular. Figure 3 shows the different elements of the flowers of
two species of subgenera Passiflora and Tacsonia. The large receptacle is often hollowed
out like a cup or basin, and bears numerous filamentous or annular appendages between
the corolla and stamens, which may be brightly colored and form a conspicuous corona
of great diversity. The calyx is composed of 3–5 free or basely connate, imbricate sepals,
and the corolla of 3–5 free or basely connate petals, which may be absent in rare cases.
The 3–5 (10) stamens are inserted either at the bottom of the perianth, or at the base or
top of the gynophore; their filaments are subulate or filiform, free or monoadelphous,
sheathing the gynophore; the anthers are versatile, introrse, two-celled, with a
4
Chapter I: General introduction
_______________________________________________________________________
longitudinal dehiscence. The ovary is superior, more or less stipitate, very rarely sessile,
unilocular, consisting of 3–5 united carpels. The styles are equal in number to the
placentae, cohering at the base, distinct at the top, spreading, simple or branched, or
separate; stigma are clavate or peltate, sometimes bilobed; The ovules are numerous,
anatropous, 1–2 seriate, attached to 3–5 parietal linear placentae by longer or shorter
funicles, enlarged into a cupule at the umbilicus. The fruit is a 1-celled, indehiscent berry
or a capsule with 3–5 semi-placentiferous valves. Seeds are numerous, with a funicle
dilated into a pulpy cupuliform or saccate aril; their testa is crustaceous, foveolate, easily
separable from the membranous endopleura, which bears a longitudinal raphe. The
straight embryo occupies the axis of a fleshy dotted albumen; the cotyledons are
foliaceous and flat, the radicle cylindric, near the hilum, centrifugal.
Figure 3. Floral elements in the subgenera Passiflora (left, P. ligularis) and Tacsonia (right, P. tripartita
var. mollissima). Photograph by Geo Coppens d’Eeckenbrugge.
5
Chapter I: General introduction
_______________________________________________________________________
I.1.4. Biology of reproduction
The high variability of flower shape and colors in Passiflora corresponds to several
pollination syndromes, showing adaptations to insects, birds and bats (Figure 4). The
pollinator behavior may be influenced by morphological as well as chemical floral
features (Varassin et al., 2001). Hypanthium length, corolla and corona color, stigma
position, and concentrations of sugars or salt in their nectar rewards are fundamental for
each pollinator type. Small bees and wasps are common pollinators of small-flowered
species, particularly those of subgenera Astrophea, Decaloba and Tryphostemmatoides.
Species with flowers of medium or large size are pollinated by large wasps (Xylocopa
spp.) mainly in subgenus Passiflora. Insect pollination is generally associated with a
well-developed corona, with concentric combination of white, yellow and/or purple
colors. The width and depth of the flower, the resistance of the operculum closing the
nectary chamber, and the distance between the corona and sexual organs are correlated
with the size of the insect. Hummingbird pollination syndrome is the second most
prevalent one. It is associated with dominant orange, pink or red colors and a tubular
general morphology resulting from a tubular hypanthium and/or an erect corona
prolonging the floral tube around the androgynophore. Such syndromes are dominant in
particular groups. The highest specialization is presented by Andean species of subgenus
Tacsonia and P. trinervia (subgenus Psilanthus), whose very long hypanthia are
essentially adapted to the sword-billed hummingbird Ensifera ensifera Boissoneau
(Buchert & Mogens, 2001). The same syndrome is displayed in the bright-red flowers of
subgenera Distephana and Murucuja, pollinated by smaller hummingbird lowland
species, such as Phaethornis superciliosus L. and Trochilus polytmus L. (Snow, 1982).
Some species, for example P. kermesina and P. coccinea, are also visited and apparently
pollinated by Heliconius butterflies (Benson et al., 1976). Bat-pollination has been
described in several species from the Amazon, such as P. mucronata Lam. and P. ovalis
Vell. ex Roemer, pollinated by the phyllostomid Glossophaga soricina Pallas, as well as
in P. penduliflora Bert, whose most frequent visitor is the Greater Antillean long-tongued
bat Monophyllus redmani Leach (Sazima & Sazima, 1978; Kay, 2001). The
corresponding pollination syndrome is mainly constituted by nocturnal anthesis, white
6
Chapter I: General introduction
_______________________________________________________________________
flowers, the release of an unpleasant odor and a particular arrangement of stigmas and
styles.
In many species, rapid movements of the stigmas are observed. When the flower opens,
the styles are erect; they then bend, bringing the stigma surface close to the stamens,
enabling contact with the pollinator; the styles straighten up again before the flower
closes. In some flowers, however, this process is incomplete or even absent, precluding
natural pollination (Fouqué & Fouqué, 1980; Escobar, 1985; Ruggiero et al., 1976). The
female fertility of these flowers is relatively low, some exhibiting incomplete
development of the pistil, while male fertility remains unchanged. The proportion of
these male functional flowers varies during flowering, being apparently dependent on the
resources the plant has previously expended in fruit production (May & Spears, 1988).
Most species open their flowers in the morning, until the end of the afternoon, favoring
pollen transfer by diurnal pollinators. In bat-pollinated species anthesis logically takes
place during the night. In P. penduliflora, it starts in the early evening and ends in the
early morning, giving hummingbirds the possibility to contribute to pollination (Kay,
2001).
High variability in pollen morphology has been described in more than 200 Passiflora
species (Erdtman, 1952; Presting, 1969; García et al., 2002; Barrios et al., 2005). Pollen
grains are generally of medium size, prolate to oblate-spherical, reticulated, with luminae
of irregular diameter.
Passiflora species are very generally allogamous. Cross-pollination is favored by the
placement of the anthers below the stigma, large, heavy and sticky pollen grains (Nishida,
1958), and frequent physiological self-incompatibility (Bruckner et al., 1999). The genus
Passiflora contains both self-compatible and self-incompatible species (Vasconcellos,
1991). Ho & Shii (1986) observed that the self-incompatibility reaction takes place in the
stigma papillae. Their diallel crosses suggested that it is controlled sporophytically by a
single multiallelic locus. Bruckner et al. (1995) first confirmed this result, however they
later found evidence for a second site of rejection in the style (Rêgo et al., 1998, 2000),
7
Chapter I: General introduction
_______________________________________________________________________
which appeared consistent with the action of a second locus acting at the gametophytic
level (Suassuma et al., 2003). According to Escobar (1992), self-compatibility is the rule
in subgenera Tacsonia and Manicata, whose species set seedy fruits after controlled selfpollination.
a
b
c
d
Figure 4. Passiflora pollinators: a. honey bee on P. sphaerocarpa; b. sword-billed hummingbird on P.
mixta; c. large wasp on P. caerulea; d. bat on P. ovalis.
As for self-incompatibility, interspecific compatibility appears to be quite variable among
Passiflora species. Spontaneous hybrids have been observed in subgenus Tacsonia and
their experimental obtention is relatively easy (Escobar, 1985; Schöniger, 1986).
However fertility may decline considerably in subsequent generations, F2 and backcrosses
(R1), due to reduced flowering or fruiting, exacerbated by poor seed germination and high
mortality (Schöniger, 1986). Nonetheless, interspecific compatibility among the most
8
Chapter I: General introduction
_______________________________________________________________________
common species of subgenus Tacsonia is consistent with evidences of interspecific
introgression (Primot et al., 2005; Segura et al., 2005). Hybridization has also been
attempted among species of subgenus Passiflora to produce new ornamentals of for
resistance breeding. Interspecific incompatibility appears more frequent than in subgenus
Tacsonia. P. edulis has often been crossed with the closely related P. incarnata, with
variable results. Fertile hybrids have been obtained by some researchers (Beal, 1972;
Winks et al., 1988) and sterile hybrids by others (Knight, 1991), who resorted to
chromosomal doubling to partially restore fertility. Ruberté-Torres & Martín (1974)
obtained hybrids from 42 parental combinations of seven passion fruit species. Most
hybrids were vigorous with intermediate characteristics of foliage, flowers and fruits.
They varied in degree of sterility of pollen and seeds. Payán & Martín (1975) used double
pollinations, with compatible and incompatible pollens on different stigmas of the same
flowers, and applied hormones to delay flower abscission and overcome incompatibility
barriers in 28 crosses involving nine species. More recent reports by Junqueira et al.
(2005) appear more optimistic on the feasibility of interspecific crosses, showing
positive results from crosses involving 15 species of subgenera Passiflora and
Distephana (P. coccinea).
Chromosome numbers are now available for more than 150 Passiflora species (Snow &
MacDougal, 1993; De Melo et al., 2001). Most species of Passiflora are diploid, with
2n = 12, 18, 20 or 24 chromosomes, although some tetraploids (2n = 24), hexaploids (2n
= 36) and octoploids (2n = 72) have been observed (Snow & MacDougal, 1993; De Melo
et al. 2001). Several chromosome base numbers (n = 3, 6, 9) have been proposed for the
genus, however in the absence of a clear understanding of the phylogenetic relationships
among species, there has been no consensus (Storey, 1950; Raven, 1975; Snow &
MacDougal, 1993; Yockteng, 2003). According to De Melo et al. (2001), n = 6 is the
most probable base number for the genus, whereas n = 9, n = 10 and n = 12 were
considered secondary. However, the second most probable base number n = 12, appears
to have played an important role in the evolution of the group because it is better
represented in other genera of the family (De Melo & Guerra, 2003). Yockteng (2003)
and Yockteng & Nadot (2004) presented an alternative hypothesis where n = 12 is the
9
Chapter I: General introduction
_______________________________________________________________________
ancestral chromosome number of Passiflora, as genus Adenia (2n = 24) was placed as a
sister clade to Passiflora in their phylogenetic study of the ncpGS sequences.
In Passiflora, biparental plastid genome transmission was first suspected by Corriveau &
Coleman (1988), following epifluorescence microscopy observations on P. edulis. Do et
al. (1992) showed that RFLP markers of cpDNA were mostly inherited maternally in
crosses between yellow and purple maracuja when the former (P. edulis f. flavicarpa)
was used as female parent, and biparental in the reciprocal cross, suggesting asymmetric
post-fertilization exclusion processes. Transmission was paternal in the hybrid ‘P.
coccinea Aubl x P. edulis f. flavicarpa’. Recently, Mráček (2005) observed biparental
transmission between P. menispermifolia Kunth and P. oerstedii Mast., and Muschner et
al. (2006) established paternal transmission of cpDNA in four interspecific hybrids of
subgenera Passiflora and Dysosmia and maternal transmission in an interspecific hybrid
of subgenus Decaloba. All mtDNA were maternally transmitted in the same five hybrids.
More recently, Hansen et al. (2007) found paternal or biparental inheritance of cpDNA,
in 16 interspecific crosses and maternal inheritance in an intraspecific progeny of
P. costaricensis, the only species of subgenus Decaloba in their sample.
I.1.5. Uses
The discovery of several thousands years old seeds of Passiflora from archaeological
sites in Virginia, North America, provides strong evidences of the pre-historic use of the
fruits by the Amerindian people (Gremillion, 1989). In Colombia, 200 years before
Christ, the Malagana people were inspired by a passion flower for a gold jewel (Figure 5;
BRC 2007). Most Passiflora species present an ornamental interest, thanks to their
spectacular shapes and colors, others are of pharmaceutical interest for their sedative,
antispasmodic, antibacterial and insecticidal properties, but they are even more
appreciated for their edible fruits, which are consumed in natura or in preparations such
as juices, sherbets and ice cream.
10
Chapter I: General introduction
_______________________________________________________________________
Several species of Passiflora have been used extensively in the traditional therapeutics in
many countries. P. edulis has been used as a sedative, diuretic, anthelmintic, antidiarrheal, stimulant, tonic and also in the treatment of hypertension, menopausal
symptoms, colic of infants in South America (Chopra et al., 1956; Kirtikar & Basu, 1975;
Mowrey, 1993). In Nagaland (India), fresh leaves of Passiflora edulis are boiled in a little
amount of water and the extract is drunk for the treatment of dysentery and hypertension
(Jamir et al., 1999). The extract of P. alata (fragrant granadilla) has been used as an
anxiolytic, sedative, diuretic and an analgesic in Brazil (Oga et al., 1984). In the West
Indies, Mexico, the Netherlands and South America, the roots of P. caerulea L., have
been used as a sedative and vermifuge. In Italy, the plant has been used as an antispasmodic and sedative (Hickey & King, 1988; Dharwan et al., 2004). P. foetida L. leaf
infusion has been used to treat hysteria and insomnia in Nigeria (Nwosu, 1999). In India
this plant (P. foetida) is widely cultivated and its leaves are applied on the head for
giddiness and headache; a decoction is given in cases of biliousness and asthma and the
fruit is used as an emetic (Kirtikar & Basu, 1975). In Brazil, the same herb is used in the
form of lotions or poultices for erysipelas and skin diseases with inflammation (Chopra et
al., 1944). P. incarnata L. is a popular traditional European remedy (Handler, 1962) as
well as a homoeopathic medicine (Rawat, 1987) for insomnia, anxiety, and has been used
as a sedative tea in North America (Bergner, 1995). The juice of P. maliformis L. is used
for intermittent fevers in Brazil. P. quadrangularis (giant granadilla) is used throughout
the Caribbean as a sedative and for headaches. Leaf infusion is taken for high blood
pressure and diabetes (Seaforth et al., 1983). In Central America, stems of the aerial parts
of P. sexflora Juss. and P. vitifolia Kunth have been used against snakebites (Morton,
1981).
11
Chapter I: General introduction
_______________________________________________________________________
Figure 5. Flower of Passiflora (probably P. ligularis)
made in gold by a precolombian goldsmith of Malagana
culture in Colombia, b.C. 200 - 200 a.C (Banco de la
República de Colombia, Museo del Oro, Bogotá D.C).
I.1.6. Cultivated species
More than 80 Passiflora species produce an edible fruit, the most interesting ones
belonging to subgenera Passiflora and Tacsonia (Martin & Nakasone, 1970; Coppens
d’Eeckenbrugge, 2003). The two botanicals forms of P. edulis Sims, edulis (purple
maracuja) and flavicarpa Degener (yellow maracuja), are by far the most important fruit
crops in the family, with a world production estimated at ca. 640.000 tons
(http://www.passionfruitjuice.com) and a permanent presence on the international
market. Other cultivated passion fruits are P. tripartita var. mollissima (Kunth) HolmNielsen & Jørgensen (curuba de Castilla), P. tarminiana Coppens & Barney (curuba
India), P. ligularis (sweet granadilla), P. maliformis (granadilla de piedra, cholupa),
P. quadrangularis (giant granadilla), P. popenovii Killip (granadilla de Quijos), P. alata
(fragrant granadilla) and P. laurifolia L. (golden apple) (Annex 3). These eight species
are mainly commercialized in South American for the local and national markets,
principally in Colombia and Brazil, with incursions on the international market. Passion
fruits are consumed fresh or processed into juices, sherbets, ice cream, and components
of industrial pastry and candies. The most important commercial species are susceptible
to a large number of pests and diseases, with considerable negative effects on production.
Thanks to the high number of species producing and edible fruit of commercial size, the
12
Chapter I: General introduction
_______________________________________________________________________
genus Passiflora has high potential for crop diversification and economic development,
which induced research institutions in the Andean countries to prioritize their
characterization and the evaluation of wild and cultivated populations (Debouck &
Libreros, 1995), and develop strategies for conservation and improvement of these
genetic resources.
I.1.7. Breeding
The genetic variability in the genus Passiflora is very wide, both within the genus and
within the most cultivated species. This variability can be used by breeders in their efforts
to improve traits in the species of subgenera Passiflora and Tacsonia, which are diploid
with 2n = 18 chromosomes. In spite of the success of a few flag species, most species
with an edible fruit are still poorly known. Few species have diffused out of America and
only the maracuja, the sweet maracuja, the granadilla, the stone granadilla and the
curubas have been objects of intensive cultivation. Many wild species of Passiflora
present characteristics of interest for breeding programs but the hybridization of these
species with cultivated species is not always possible.
Several levels of passion fruit selection are observed in relation to the coexistence of all
domestication stages. Many species, which are rarely cultivated but simply harvested,
have never undergone artificial selection. Cultivation of species grown in home gardens
has resulted in the creation of some unique types, such as the large-fruited giant
granadilla (P. quadrangularis var. macrocarpa), which bears exceptionally large fruit. In
commercialized species, such as the maracuja, sweet granadilla and curubas (tacsos),
growers carry out phenotypic or mass selection when establishing or renewing new plots,
in function of their observations or, sometimes, in function of an ideotype imposed by the
local market. Seeds are collected from a small number of good quality fruits plucked
from one or two high performing plants. Given the size of the plots, the total population
is small and the selection intensity is low, especially if the plantation renewal cycle is
long. This practice and seed exchanges maintain considerable variability in the
populations.
13
Chapter I: General introduction
_______________________________________________________________________
Modern breeding has been focused on P. edulis. The first work took place in the
developed tropical or subtropical regions where the commercial cultivation of P. edulis
was initiated, but with very limited genetic resources and without even knowing the
existing variation. These few institutional works concentrated on the clonal propagation
of hybrids between yellow and purple maracujas, obtained on a very narrow genetic base,
in Australia, Hawai and Florida (Knight, 1992; Vanderplank, 2000; Winks et al., 1988).
The main breeding objectives were early flowering, improved yield, resistance to pests
and diseases (Fusarium, Phytophthora, Alternaria, Xanthomonas and virus), cold
tolerance and fruit quality (Table 1). In spite of the lack of basic knowledge on genetic
resources, complex technical approaches have been explored directly, including
interspecific hybridizations, in Puerto Rico and Australia (Beal, 1972; Payán & Martín,
1975; Winks et al., 1988), biotechnologies, with in vitro tissue culture (Drew 1991;
Cancino & Hodson, 1994), somatic hybridization (Dornelas et al., 1995; Otoni et al.,
1995; Barbosa & Vieira, 1997) and genetic transformation with A. tumefaciens (Manders
et al., 1994; Silva, 1998; Hall et al., 2000).
As the commercial cultivation of P. edulis started in tropical America, their place of
origin, with materials repatriated from Hawaii, propagation by seedlings became the rule.
This cultural practice, the increase in production areas and the abundance of natural
pollinators increased genetic variation, widening the basis for later breeding programs.
The most significant ones have taken place in Brazil, first country in production as well
as consumption of yellow maracuja and sweet maracuja. The most advanced Brazilian
breeding programs have used progeny-testing for the obtention of synthetic populations,
aiming at a better productivity, quality and homogeneity of the crop, while maintaining
sufficient genetic diversity for efficient cross-pollinations (Coppens d’Eeckenbrugge et
al., 2001; Meletti et al., 2005). Several cultivars have been proposed for consumption in
natura (e.g. ‘IAC-273’ and ‘IAC-277’; Meletti et al., 2000) and for the agroindustry (e.g.
‘IAC-275’ and ‘CPATU-Casca Fina’; Nascimento et al., 2003).
14
Chapter I: General introduction
_______________________________________________________________________
Table 1: Breeding objectives in passion fruits.
Objective
Productivity
Fruit quality
•
•
•
•
•
•
•
•
•
•
Resistance to fungal diseases
Resistance to bacterial diseases
•
•
•
•
•
•
Resistance to viral diseases
Resistance to nematodes
Acclimatization
•
•
•
•
Selection criteria
Vigor, early production, productivity (flowering,
fruiting).
Self-fertility if clonal selections (out of dated).
Adequate and uniform size.
Yield of pulp and/or juice.
Sugar/acidity ratio (fresh or processed) and pulp
aroma.
Color of pulp, dark yellow for maracujas, orange to
salmon for curubas and giant granadilla.
Small seeds.
Pericarp colored, turgescent (smooth), durable and
resistant during transportation.
Alternaria alternata, A. passiflorae and other species
for maracujas and curubas.
Fusarium oxysporum and F. solani for the purple
maracuja and sweet granadilla.
Phytophtora nicotianae and P. cinnamoni for
maracujas.
Septoria passiflorae for the purple maracuja
Oidium sp. for curubas.
Colletotrichum gloeosporioides (anthracnose) and
Cladosporium herborum, especially for curubas.
Xanthomonas passiflorae and X. campestris pv.
Passiflorae for maracujas, P. alata and giant
granadilla.
Potyviruses for maracujas (PMV and SMV) and
sweet granadilla (ringspot).
Tymovirus and closterovirus for maracujas.
Meloidogyne incognita and M. javanica for
maracujas, giant granadilla, P. alata and curubas
Rotylenchulus reniformis for the purple maracuja.
Cold resistance for yellow and purple maracujas and
frost resistance for curubas.
I.1.8. Passiflora research and prospects, in Colombia and in the Andean region
There is little factual information available on the Andean passion fruits. This must be
generated by comparative tests in the areas where these species are available. Research
institutions in Bolivia, Peru, Ecuador, Colombia, and Venezuela could provide
information that may be useful as the basis for industrial development. The future of
these fruits will depend upon horticultural development. The production of pulp and
15
Chapter I: General introduction
_______________________________________________________________________
concentrate has extremely good prospects if commercial-scale production can be
established and maintained.
In the last ten years, institutions of Andean countries, grouped in a net promoted by
Bioversity International (formerly IPGRI), have started a systematic exploration of the
genetic resources of Andean Passiflora species, which has allowed the constitution of
collections of the most common species. Nowadays, important collections exist in the
five countries: Bolivia, Colombia, Ecuador, Peru and Venezuela. Scientists from the
region have investigated the genetic variability of the cultivated curubas and related wild
species, building the bases for conventional breeding programs. However, similar efforts
on the warm climate species have only been carried out in Brazil.
The main problems encountered along these research efforts on passion fruit genetic
resources have included the difficulty of approaching such a wide morphological and
genetic diversity at both intra- and interspecific levels, with related taxonomical problems
and inconsistencies, and practical problems for its effective conservation. The specific
adaptations and high level of endemism of species originating from a wide range of
habitats impose in situ conservation strategies, implying a good knowledge of the
distribution of the different species and the repartition of their diversity.
In Colombia, several collaborative projects have been focused on Passifloraceae. The
Interamerican Development Bank (BID) has supported a regional project, coordinated by
Bioversity International in 1994-1997. Colciencias funded in 1999-2001 the national
project “Conservación y utilización de los recursos genéticos de pasifloras”, developed
by French and Colombian scientists at the IPGRI-Americas office. In 2004, the same
group developed a study of diversity of the Passifloraceae and Caricaceae in the
Colombian coffee growing zone. All these projects have generated a considerable amount
of information on morphology, cytology, palynology, molecular diversity, and
biogeography of Passiflora, strengthening the national collections and providing most of
the material for the work presented in this thesis.
16
Chapter I: General introduction
_______________________________________________________________________
I.1.9. Structure of the thesis
Our research is presented in five more chapters. The first two bear on Passiflora
geography in Colombia. The first one consists of an inventory of species for Colombia,
their distribution among biogeographic regions and administrative divisions, and
considers their status of conservation. The second presents a more detailed analysis of the
ecogeographic drivers of Passiflora diversity, through distribution modeling in relation to
climatic adaptation of the different species, and gives orientations for their in situ
conservation. The next two chapters analyze Passiflora diversity, the third chapter
concerning morphological diversity and the fourth one genetic diversity. Each of these
four chapters is treated independently, with the structure of a scientific article. A wider
general discussion is the object of the final chapter.
17
CHAPTER II
____________________________________
Diversity of Colombian Passifloraceae:
biogeography and an updated list for conservation
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
II.1 Diversity of Colombian Passifloraceae: biogeography and an
updated list for conservation
John Ocampo Pérez1, Geo Coppens d’Eeckenbrugge2, María Restrepo1, Andy Jarvis1,3, Mike
Salazar1, and Creuci Caetano1,4.
1
Bioversity International (formerly IPGRI), Regional Office for the Americas, A.A. 6713,
Cali, Colombia.
2
CIRAD/FLHOR, UPR ‘Gestion des ressources génétiques et dynamiques sociales’,
Campus CNRS/Cefe, 1919 route de Mende, 34293 Montpellier, France.
3
International Center for Tropical Agriculture (CIAT), A.A. 6713, Cali, Colombia.
4
Universidad Nacional de Colombia Sede Palmira. Facultad de Ciencias Agropecuarias.
Kra. 32 Chapinero, vía Candelaria. Palmira, Valle del Cauca, Colombia.
Submitted and accepted in Biota Colombiana
II.1.1. Abstract
The list of Colombian Passifloraceae was revised, using 3,930 records from literature,
herbaria, and field observations. It includes 167 species, 165 of them native, which is
equivalent to 27% of the family. Our list brings more details on species distribution and
presents 26 species new to Colombia. Passiflora is the most important genus, with 162
species. When compared with other regions, the Andes of Colombia and Ecuador
constitute its center of diversity. Within Colombia, the highest diversity is also
concentrated in the Andean region, with 81% of the species, particularly in the
departments of Antioquia, Valle del Cauca, Cundinamarca, Quindío, Risaralda, and
Caldas. The highest number of species is found at elevations between 1000 and 2,000 m.
Most common species thrive in disturbed habitats, such as borders of roads, cultivated
plots, and secondary forest. Most of the 58 endemic species are high Andean and belong
to subgenera Tacsonia and Decaloba. Forty-two species produce an edible fruit, and nine
are commercially cultivated. Among the species reported, 70% are threatened to some
degree and three are considered extinct. Colombia may still harbor many unknown
species in poorly explored departments. A better knowledge of Passiflora diversity and
its distribution is needed to develop its economic potential. The urgent task of conserving
this threatened richness must target the conservation of these resources as well as their
habitat. Both aspects may be combined if Passifloraceae can be used as an indicator of
biodiversity in the Andean region, which seems justified by their multiple ecological
interactions with many organisms.
Keywords: biodiversity, Colombia, Neotropics, Passifloraceae, passionflowers.
19
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
II.1.2. Introduction
The Passifloraceae consist of 18 genera and approximately 630 species, distributed
throughout the tropics from the coastal zones up to 3,800 m in Andean páramos (HolmNielsen et al., 1988). In America, it is represented by four genera (Ancistrothyrsus,
Dilkea, Mitostemma and Passiflora). Passiflora, with about 525 species is numerically
and economically the most important genus of the family; it is distributed mainly in the
New World (Ulmer & MacDougal, 2004), with only 22 Old World species of subgenus
Decaloba (syn. Plectostemma sensu Killip), in the tropical and sub-tropical regions of
Southeast Asia and Austral Pacific. Passionflowers are generally perennial lianas or
herbaceous vines climbing by tendrils, although some are trees, shrubs, or even annuals.
Their wide morphological variation appears to result from the diversity of their habitats
as well as their coevolutionary relationships with many organisms, including protective
ants (Apple & Feener, 2001), herbivores (particularly Heliconius spp. butterflies; Gilbert,
1982), pollinators, and the plant communities providing them physical support and access
to sunlight. Pollination is mainly carried out by insects and birds; several species are batpollinated (Endress, 1994; Büchert & Mogens, 2001) and a few species exhibit elements
of the carnivory syndrome (Radhamani et al., 1995). Many species are cultivated for their
edible fruit, as ornamentals, or for their medicinal properties (Ulmer & MacDougal,
2004; Coppens d’Eeckenbrugge, 2003; Martin & Nakasone, 1970; Dharwan et al., 2004).
P. edulis Sims (maracuja) is by far the best known and economically most important
species of the family.
When Spanish missionaries arrived in South America in the 16th century, they felt
passionflowers were a good omen for their mission. In their unique morphology, they
saw the elements of the Passion of Jesus Christ and a sign that the New World would
successfully be converted to Christianity (Killip, 1938; Uribe, 1955a). This religious
symbolism gave the plant their common name of Flos Passionis, or “passion flowers”.
The Latin translation by Pluckenet (1696; cited by Escobar, 1988a) was accepted for the
genus Passiflora created by Linnaeus in 1753, who described 24 species in his Species
Plantarum (cited by Killip, 1938), a number increased to 35 by Lamarck (1789; cited by
20
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Killip, 1938). The first extensive monograph of the family was published by Cavanilles
in 1780 (cited by Killip, 1938), with 43 species treated. They were followed by authors
like Jussieu (1805), De Candolle’s (1828), Masters (1872) and Harms (1925), who
described about 250 species divided into 21 sections (Killip, 1938). In his 1938
monograph, The American Species of Passifloraceae, Killip made the most extensive
description of the New World species, classifying 355 species into 17 genera and 22
subgenera, based on floral morphology (Annex 1). In Colombia, the priest Uribe (1954,
1955a, 1955b, 1957, 1958, 1972) described several new species and Escobar (1986, 1987,
1988a, 1988b, 1989, 1990, 1990 inedited, 1994) revised the subgenera Distephana,
Manicata (syn. Granadillastrum), Rathea and Tacsonia, including Tacsoniopsis in the
latter, and described one additional subgenus, Porphyropathanthus. She passed away in
1993, leaving an inedited document on her revision of subgenus Astrophea. MacDougal
revised Killip’s subgenus Plectostemma in 1994, restoring its ancient name Decaloba. In
the last decade, MacDougal and Feuillet have published many papers including the
description of about 15 new species, mainly of the subgenera Decaloba and Astrophea
(MacDougal, 1992, 1994, 2006; Feuillet, 2002, 2004). Recently, Feuillet & MacDougal
(2003; Annex 2) proposed a new infrageneric classification in Passiflora. According to
this proposal, only based on morphological characters, four subgenera would be
recognized: Astrophea and Deidamioides, from South and Central America, Decaloba,
from America, Southeast Asia and Australia, and Passiflora, exclusively from America
(Ulmer & MacDougal, 2004). Additionally, they proposed to downgrade genus
Tetrastylis as a section of subgenus Deidamioides. Recent molecular analyses partly
support the reduction in the number of subgenera (Muschner et al., 2003; Yockteng,
2003; Yockteng & Nadot, 2004; Hansen et al., 2006; see Annex 4a-d), with the existence
of at least three major groups, corresponding globally to subgenera Decaloba, Passiflora
and Astrophea of the new proposal. On the other hand, molecular data from the different
studies are not always consistent on the relative placement of these groups, and their
results are less clear at lower levels, with inconsistent grouping of particular species and
poor correspondence with some well established morphological divisions. In addition, the
monophyly of Passiflora has not been established, and the study of Muschner et al.
(2003) even raises some doubts about it. Clearly, more studies, involving more numerous
21
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
species samples, are needed before re-evaluating such a complex and fast evolving group
as Passiflora.
Colombia’s location and variety of ecosystems places it in second position for
biodiversity (MacNeely et al., 1990). The country is divided into five main biogeographic
regions: Amazonian, Andean, Caribbean, Orinoquian, and Pacific. The Andean region
presents a highly varied topography (100-5,400 m) with three main mountain ranges.
Thus, the Eastern, Central and Western Cordilleras separate two large inter-Andean
valleys from the Pacific Coast to the West and the Orinoquian ‘Llanos’ to the East. The
uplift of the Andes created new habitats and increased local isolation, favoring high
speciation rates in many taxa. In Passiflora, a particularly striking example is given by
subgenus Tacsonia, whose beautiful and large-flowered species are strictly adapted to
high altitudes in cloud forest (2,000-3,800 m) and pollination by the sword-billed
hummingbird Ensifera ensifera Lesson, which shows the same distribution (Büchert &
Mogens, 2001). As a result of this habitat diversification, the Colombian flora includes
one of the world most diverse groups of vascular plants, with 51,220 documented species
(May, 1992; UNEP-WCMC, 2004). However, Colombia has undergone recent
transformation of large parts of its natural ecosystems, in particular in the Andean region.
Seventy percent of the Andes, an area that is vital to the conservation of Colombia’s
water supply, have been deforested as a result of both agricultural colonization and
human migration (World Press Review, 1993). Destruction of natural habitats has
drastically affected many species distributions, often reducing their historical ranges to a
set of small, fragmented populations (Brooks et al., 2002). Such habitat alteration is
predicted to lead to substantial extinction in the near future.
In this context of rapid erosion of biodiversity, Passifloraceae are highly interesting, not
only for their fast radiation and spectacular variation in morphology and reproductive
biology. Indeed, as stated above, this family is exemplative from the standpoint of
coevolution in many respects, such as their particular relationship with specialized
herbivores, ants and other nectar feeding insects; most importantly, they are parasites of
structure, as they depend on many very different species for their support, from low
22
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
shrubs in disturbed habitats to high trees in primary forests. They are mainly perennials,
but their life cycle is much shorter than that of their support. They are sensitive to longterm changes in the ecosystem (dependence on trees) as well as short-to medium-term
changes (by their other adaptive traits). Thus, they should constitute an excellent
indicator group for the monitoring of biodiversity in Colombia. In addition, Colombia
presents a long tradition of diversity in fruit production and consumption, and it is the
country with the highest number of marketed passion fruit species, so the study of
Passiflora diversity must also be thought in terms of conservation of genetic resources of
important or promising fruit crops.
The last inventory by Hernández & Bernal (2000) recorded 141 Passifloraceae species
distributed in all the biogeographic regions. Forty-eight of them are endemic to
Colombia, mostly living in the Andean region. This inventory was based on the study of
specimens from five herbaria (COL, HUA, JAUM, MEDEL and MO) and the citations
made in publications compiled by several authors that have worked on the family.
Several recent, collaborative projects have been focused on Passifloraceae. The
Interamerican Development Bank (BID) has supported a regional project, coordinated by
IPGRI (currently Bioversity International) in 1994-1997. Colciencias funded in 19992001 the national project “Conservación y utilización de los recursos genéticos de
pasifloras”, developed by French and Colombian scientists at the IPGRI Americas office.
In 2004, the same group developed a study of diversity of the Passifloraceae and
Caricaceae in the Colombian coffee growing zone. All these projects have generated a
considerable amount of information on morphology, cytology, palynology, molecular
diversity, and biogeography of Passiflora, providing most of the material for the present
inventory and allowing us to supplement and update the list of Hernández & Bernal
(2000) with new information, such as species new to science or to the country and
elements of ethnobotanical information. In addition, the use of a Geographic Information
System (GIS) allowed us to re-assess the conservation status of Colombian
Passifloraceae.
23
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
II.1.3. Materials and methods
II.1.3.1. Study area
Colombia is situated in the north of South America, between 12° 26’ 46” N and 4° 13’
30” S and between 66° 50’ 54” W and 79° 02’ 33” W, covering an area of 1,141,748
km2, with an altitudinal range from the sea level to 5,775 m (http://www.igac.gov.co).
The main administrative division defines 32 departments and geographers recognize five
biogeographic regions (Hernández et al., 1991).
II.1.3.2. Herbarium and literature data
The data set consists of information gathered from specimen labels from 18 Colombian
herbaria (AFP, CAUP, CDMB, CHOCO, COL, COAH, CUVC, FAUC, FMB, HUA,
HUQ, JAUM, MEDEL, PSO, SURCO, TOLI, VALLE, UIS) and five herbaria in other
countries (K, MA, MO, NY, P). These collections were gathered between 1750 and 2006.
Most specimens were verified or identified, using the keys and descriptions of Killip
(1938), and amendments by Holm-Nielsen et al. (1988), Escobar (1988a, 1994),
MacDougal (1994) and Tillet (2003). A synonymy list, based on the general list of
Feuillet & MacDougal (2003), is given in the Appendix. When possible, voucher label
information was used to assign geographic coordinates to specimens, using gazetteers and
topographic maps of Colombia (scale 1:50,000 and 1:250,000). The database was
supplemented with materials mentioned in species descriptions, essentially those of Killip
(1938, 1960), Uribe (1955a), and Escobar (1988a,b, 1989, 1990, 1990 inedited, 1994).
Collection records with obviously inaccurate or doubtful data were excluded from the
analysis. Coordinates were further checked by plotting all species on a dot map, using the
DIVA-GIS 5.2 software (Hijmans et al., 2001). Finally, we followed the infrageneric
classification by Killip (1938) with the amendments of Escobar (1988, 1989) and
MacDougal (1994).
24
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
II.1.3.3. Expeditions and samples collected
The dot map of all geo-referenced specimens was used to plan germplasm collecting
trips. The prioritization of explored areas followed three criteria: permission of access
(unfortunately not obtained for protected areas), richness of species and collection gaps.
The collecting trips were carried out during 2003-2006, covering 555 localities in 17
departments, between 0 and 4,200 m of altitude. The explorations were concentrated in
the Andean region, in watersheds, wild forest areas, cultivated fields and road borders.
Data were recorded for each collected specimen, including locality names, elevation,
geographic coordinates using a hand-held GPS device, status (wild, cultivated or
introduced), and ethnobotanical information (if any). These passport data were recorded
and tabulated. Finally, the Geographic Information System software DIVA-GIS 5.2 was
used to generate a dot map of the distribution of accessions collected/observed during the
expedition.
II.1.3.4. Threat status of Passifloraceae
The distribution area of each native species was characterized by the maximum distance
(MaxD) and the circular area (CA50), following the method of Hijmans et al. (2001). This
methodology has been applied in a number of studies to provide quantitative assessment
of the distribution area required by the Red List criteria, for example by Maxted et al.
(2005). MaxD is the largest distance between any pair of observations of one species.
CA50 is the total surface within a 50-km radius around all the observations for a same
species. These methods were supplemented with historical records of each taxon and
subjected to the Red List criteria of the World Conservation Union (IUCN 2003, 2004),
involving complex combinations of quantitative observations concerning the size and
structure of the population, the range and fragmentation of its distribution (extent of
occurrence and area of occupancy), as well as the intensity of their past or foreseeable
variation. Along these lines, we considered that CA50 under 20,000 km2, MaxD under
100 km and number of observations under six, as well as the absence of records younger
than 100 years, are critical.
25
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
II.1.4. Results
II.1.4.1. Data collecting
A total of 3,330 herbaria and 45 literature data, concerning 120 species, were gathered
and georeferenced when coordinates were not directly available. The highest number of
species and specimens were found in the Colombian herbaria COL and HUA, with 1,056
and 976 records, respectively. During the collecting trips, most specimens were observed
in forest fragments, gallery forest and forest and road edges, mainly in the watersheds of
the coffee growing zone, between 1000 and 2,000 m. In all sites visited during the
expeditions, 87 Passifloraceae species were recorded. Five individuals could not be
identified. The dot map of Figure 1 shows the spatial distribution of our final dataset of
3,930 records for herbarium (3,330), literature (45) and field collections (555) of
Passifloraceae in the different biogeographic regions.
II.1.4.2. Distribution of species richness
The number of observations and species richness was highest on the Andean slopes (123
species), followed by the Amazonian region with 45 species (Table 1). The Orinoquian
was the poorest region, with only 19 species. The Andean and Caribbean regions share
the highest number of species (27). By contrast, the Pacific and Caribbean regions only
present four species in common. Figure 2 gives a synthetic image of the similarities in
species occurrence among regions, confirming a relative similarity between the
Amazonian and Orinoquian as well as between the Andean and Caribbean regions. The
Pacific Coast Passifloraceae appear relatively divergent. In the Andean region, Antioquia,
Valle del Cauca, Cundinamarca and Santander were the departments that displayed the
highest richness of specimens and species (Table 2). Considering their area, Quindío,
Risaralda and Caldas are even more diverse. The department of San Andrés and
Providencia (Caribbean islands) is only represented by P. biflora Lam. and P. pallida L.
26
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Figure 1. Map of distribution of Passifloraceae specimens for 3,930 collections in the five biogeographic
regions in Colombia. Points on the map represent sites of collection.
27
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Table 1. Distribution of Passifloraceae by biogeographic region. The diagonal gives their contribution in
species number (bold) and contribution to the country’s total. The other cells give the number and
proportion of shared species for each pair of regions.
Biogeographic region
Amazonian
Andean
Caribbean
Orinoquian
Pacific
amz
and
car
ori
pac
45 (28%)
21 (14%)
123 (76%)
9 (12%)
27 (20%)
38 (23%)
15 (31%)
7 (5%)
9 (19%)
19 (12%)
15 (23%)
14 (10%)
4 (6%)
9 (14%)
36 (22%)
II.1.4.3. New Passifloraceae checklist for Colombia
Table 3 gives the number of species for each genus and subgenus present in Colombia in
relation with the number of species present in the Neotropics. The updated inventory of
the Colombian species (Table 4) includes a total of 167 Passifloraceae species,
representing three genera, Ancistrothyrsus, Dilkea and Passiflora. This is equivalent to
27% of all Passifloraceae. The genus Passiflora is by far the most important with 162
species, representing 11 of Killip’s subgenera, and all the four subgenera defined in the
classification proposed by Feuillet & MacDougal (2003). The most abundant species
were P. vitifolia Kunth (359 specimens) and P. mixta L. (162 specimens), while 67
species (23%) were represented by a single specimen.
Figure 2. Diagram comparing the similarity in contribution of Passifloraceae species to the floras of the
Colombian biogeographic regions (Jaccard distance).
28
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
In the expeditions, we found some species that had not been collected in the last decades,
such as P. erytrophylla Mast., P. guazumaefolia Juss., and the semi-arborescent
P. mariquitensis Mutis ex Uribe. The latter was described in 1783 by José Celestino
Mutis during the Botanical Expedition of the “Nuevo Reino de Granada” in Mariquita
(Tolima). It was considered extinct by Uribe (1955a) and a synonym of P. pittieri Mast.
by Escobar (1990 inedited). However, we could verify that P. mariquitensis still exists, as
three specimens that we have collected in highly disturbed forest near Mariquita
corresponded very well to the type specimen, while they appeared morphologically
distinct from P. pittieri specimens from Costa Rica, Panama, and northwestern Colombia
in several traits (e.g. nectar shape, peduncle length, nerve shape). Similarly, we
maintained other species that had been considered synonyms by Hernández & Bernal
(2000), such as P. mollis H.B.K. (vs. P. cuspidifolia Harms), and P. hahnii Mast. (vs.
P. guatemalensis S. Watson), after checking the collected materials against the type
specimens. Ours list presents 26 species new to Colombia, from those recognized by
Killip (1960), Feuillet & MacDougal (2003) and Ulmer & McDougal (2004) and three
inedited from Escobar (1990) and Hernández (2003): Ancistrothyrsus antioquiensis L.K
Escobar (ined.), P. alata Curtis, P. andina Killip, P. bucaramangensis Killip,
P. candollei Tr. & Planch., P. chocoensis Gerlach & Ulmer, P. cincinnata Mast.,
P. hahnii (Fourn.) Mast., P. hirtiflora Jørgensen & Holm-Nielsen, P. killipiana
Cuatrecasas, P. lyra Planch. & Linden & ex Killip, P. megacoriacea Porter-Utley (ined.),
P. mollis Kunth, P. monadelpha Jørgensen & Holm-Nielsen, P. munchiquensis
Hernández (ined.), P. occidentalis Hernández (ined.), P. pallida L. (clearly separated
from P. suberosa by Porter-Utley, 2003), P. pillosissima Killip, P. popenovii Killip,
P. sodiroi Harms, P. tuberosa Jacq., P. rigidifolia Killip, P. tricuspis Mast.,
P. truxillensis Planch. & Lind. P. caerulea L., recently introduced from Brazil and
Argentina and cultivated as an ornamental, was not included in the counts of each
department. P. alata was not counted for Quindío and Valle del Cauca either, as the
material under cultivation was also introduced from Brazil. Nine more species occur
close to the Colombian international border (less than 100 km), and possibly exist also in
the country, although they have not been included in this inventory. Another important
result is the presence of the genera Ancistrothyrsus and Dilkea in the Andean and Pacific
29
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
regions, the former following the mention of A. antioquiensis by Escobar (1990 ined.).
Unfortunately, she passed away before publishing her monograph on arborescent
Passifloraceae.
Several botanical forms and varieties are mentioned for P. edulis Sims, P. cumbalensis
(Karst.) Harms, P. foetida L, P. ligularis Juss., P. longipes Juss., P. rugosa (Mast.) and
P. tripartita (Juss.) Poir. A total of 42 species with edible fruit are reported. Nine of them
are sold on the international, national and/or local markets, P. edulis f. flavicarpa
Degener and P. edulis f. edulis (introduced), P. ligularis, P. tripartita var. mollissima, P.
tarminiana Coppens & Barney, P. quadrangularis L., P. maliformis L., P. popenovii
Killip, P. nitida Kunth, and P. alata Curtis. Other species, such as P. antioquiensis H.
Karst., P. cumbalensis, P. laurifolia L., P. nitida Kunth, P. palenquensis Holm-Niels. &
Lawesson P. tiliifolia L., and P. pinnatistipula Cav. are cultivated in home gardens. Some
commonly cultivated species seem to depend on human activity for their propagation,
which suggests an advanced stage of domestication and/or an incomplete acclimatisation
following an ancient introduction. Thus, P. edulis f. flavicarpa, P. ligularis,
P. quadrangularis L., P. popenovii P. tripartita var. mollissima, and P. tarminiana, are
exceptionally found as feral plants. On the other hand, let us remind that the latter has
pullulated as an invasive plant in Hawaii and New Zealand. Another particular case is P.
edulis f. edulis, introduced from southern South America, which has naturalized at
intermediate to high altitudes, where it is not uncommon in the wild.
30
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Table 2. Number of observations and species of Passifloraceae in the 32 Colombian departments.
Department
Amazonas
Antioquia
Arauca
Atlántico
Bolívar
Boyacá
Caldas
Caquetá
Casanare
Cauca
Cesar
Chocó
Córdoba
Cundinamarca
Guainía
Guaviare
Huila
La Guajira
Magdalena
Meta
Nariño
Norte de Santander
Putumayo
Quindío
Risaralda
S. Andrés y Prov.
Santander
Sucre
Tolima
Valle del Cauca
Vaupés
Vichada
Abbreviation
Biogeographic region
Observation number
Species number
ama
ant
ara
at
bl
by
cl
cq
cs
cau
ce
cho
cor
cun
gn
gv
hu
lg
ma
met
na
ns
pu
qu
ri
sp
snt
suc
to
vc
va
vch
amz
and car pac
and ori
Car
and car
and ori
And
amz and
and ori
amz and pac
and car
and pac
and car
and ori
amz
amz
and
and car
car
amz and ori
and pac
and
amz and
and
and pac
car
and
car
and
and pac
amz
ori
87
784
10
18
33
145
245
47
4
161
13
211
33
419
16
27
62
21
71
85
170
79
56
150
68
4
203
6
213
420
35
16
19
70
6
7
17
36
36
18
4
42
10
40
9
53
10
14
22
12
31
24
44
36
26
38
24
2
48
3
44
56
20
9
The vernacular names are very diverse for each species. In the Amazonian region, we
noted several indigenous names for the species P. foetida var. gossypiifolia Desv. (Iñanaleeg, Murulale), P. holtii Killip (Guachique), P. nitida (Burucuña, Gemarundare,
Tuchica, Jino-Gojé), P. serratodigitata L. (Cipo-Cipo), P. vitifolia (Maloca de Fisi). In
the Cauca and Nariño departments (south of the Andean region) P. fimbriatistipula
Harms and P. ligularis are named Pachuaca and Awapit in the indigenous languages.
31
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Table 3. Number of Passifloraceae species in Colombia and the Neotropics.
Genus
Ancystrothyrsus
Dilkea
Mitostemma
Passiflora
Subgenus
Colombia
Astrophea
Decaloba
Dysosmia
Distephana
Manicata
Passiflora
Porphyropathanthus
Psilanthus
Rathea
Tacsonia
Tryphostemmatoides
All Passifloraceae
Neotropics
2
3
0
22
52
2
6
1
38
1
3
2
30
4
3
5
3
57
190
20
15
5
156
1
4
3
55
7
167
533
Among the species collected in our expeditions, we found several species growing very
commonly in disturbed habitats like road borders, secondary forest margins, and
especially riverbanks between 1000 and 2,000 m: P. adenopoda Moc, & Sessé ex DC.,
P. alnifolia Kunth, P. coriaceae Juss., P. capsularis L., P. rubra L, and P. suberosa L.
The latter two can even be considered weeds in the coffee plantations. At higher altitudes
(above 2,500 m), P. mixta is also very common in disturbed habitats.
II.1.4.4. Endemism
Among the 165 native species, 58 (36%) are endemic to the country. The largest
concentration of these occurs in the Andean region, principally in the Cordillera Central,
in the departments of Antioquia and Tolima. The elevation belt between 1,500 and
2,500 m presents the highest richness of endemic and rare species (≤ 5 observations).
Only eight of these were represented with only one specimen (e.g. P. cremastantha
Harms), while P. bogotensis Benth and P. antioquiensis were the most collected endemic
species, with 23 recorded specimens each. The proportion of endemic species varied
considerably among taxonomic groups, especially among the subgenera of Passiflora
(Table 4). Thus, Tacsonia (21), Decaloba (13), Passiflora (9) and Astrophea (7) present
the highest number of endemic species. Subgenus Tacsonia displays the highest richness
32
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
of endemic species in the Cordillera Central with eight species, mainly of the section
Colombiana characterized by a very long peduncle (P. flexipes Triana & Planch.,
P. linearistipula L.K. Escobar, P. quindiensis Killip and P. tenerifensis L.K. Escobar).
Twenty-one species (37%) are restricted to very small areas of one department. These are
located mainly in the departments of Antioquia (6), Tolima (4), Santander (3), Cauca (2),
while only one such narrow endemic is found for the departments of Bolivar, Boyacá,
Chocó, Caldas, Cauca, and Magdalena. Figure 3 shows 15 endemic species from five
subgenera.
a
b
c
d
e
f
g
h
i
j
k
l
m
n
o
Figure 3. Colombian endemic species (a) P. antioquiensis; (b) P. parritae; (c) P. flexipes; (d)
P. linearistipula; (e) P. lanata; (f) P. tenerifensis; (g) P. trinervia; (h) P. longipes; (i) P. erytrophylla; (j)
P. bogotensis; (k) P. magdalenae; (l) P. smithii; (m) P. arborea; (n) P sphaerocarpa; (o) P. emarginata.
33
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
II.1.4.5. Threatened species
Figure 4 shows the distribution of the 165 Colombian Passifloraceae native species,
according to their threat status under the criteria of the IUCN (2003, 2004). Seventy-one
percent of them present some threat degree, 10% being critically endangered (CR), 6.1%
vulnerable (VU) or endangered (EN). Four of the 16 critically endangered species are
endemic. All three extinct species (EX) belong to the Andean subgenus Tacsonia.
Unfortunately, the only two species of genus Ancistrothyrsus are included in the category
CR. Only 16% of the species were placed in the two categories LC and NT, ‘least
concern’ and ‘near threatened’. The species P. alata, P. megacoriacea and P. rigidifolia
are placed in the DD category because of deficient data. The 29.3% classified in ‘least
concern’, belong mostly to subgenera Decaloba and Passiflora with 18 and 14 species,
respectively.
Figure 4. Percentual number of the threat status of 165 Passifloraceae native species under the IUCN
criteria.
34
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
II.1.5. Discussion
Colombia has been subject to many studies focused on inventories of plant species
groups (Gentry, 1993; Silverstone-Sopkin & Ramos, 1995; Galeano et al., 1998; Rangel,
1995, 2002). Passifloraceae have been inventoried in taxonomical works by Escobar
(1998a, 1989, 1990 inedited) and Hernández & Bernal (2000). As compared to the latter,
we have added new information on geographical distribution of each taxon and extended
the list to a total of 167 Passifloraceae species, from three genera and the five
biogeographic regions, with reports of 26 species new to Colombia.
For obvious reasons, the quality of botanical inventories depends on the quality of
taxonomical work in this complex family. While the definition of genera and subgenera
should not affect significantly studies of the distribution of its diversity across the
Colombian territory, such work may be affected to some extent by poor definitions below
the subgenus level. Indeed, several morphological groups include species that are very
similar, and regularly reported as very difficult to distinguish from each other. In several
cases, experts may have underestimated intraspecific variation in widely distributed
species, or even intra-individual variation, splitting well known species in several new
species only distinguished by a few quantitative or color traits. Among the difficult
groups, let us mention particularly subgenus Astrophea, whose species tend to be less
well differentiated, at least in sterile specimens, by nectar gland position and number,
having only two at the junction of the lamina and petiole, while they may show
impressive intraspecific variation in pubescence and intra-individual variation in leaf size
and shape according to light exposure and whole tree development (heteroblasty). In
subgenus Decaloba also, several morphological groups demand great experience and
caution for their identification, even in the most common species such as P. capsularis
and P. rubra, which can be found in the same habitats. In the most difficult cases, several
species have even changed status several times. For instance, Killip merged
P. bauhinifolia H.B.K. with P. andreana Mast. in 1938, and restored it as a distinct
species in 1960, while Holm-Nielsen et al. (1988) merged P. bauhinifolia with another
close relative, P. alnifolia, a position that we have adopted here. On the other hand, a
35
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
couple of other species may also show very little morphological differentiation, as P.
mollis and P. cuspidifolia or P. hahnii and P. guatemalensis, but differ in their altitudinal
distribution, which confirms that they are distinct entities. Many new species of subgenus
Distephana are also questionable, as one of its two most common species, P. coccinea
Aubl., distributed in most of the Amazon, has been split in several species on the basis of
bract size, nectar gland numbers, and small variation in numbers and respective colors of
the corona series. Concerning Colombia, Vanderplank (2006) underlined that the
description of P. coccinea by Escobar (1988) matches perfectly that of P. miniata
Vanderplank, so he considered the latter a Colombian species. However, we have not
adhered to this opinion for several reasons. Vanderplank described it on material grown
in glasshouse and his report does not refer to the examination of Colombian materials.
The type and level of the differentiation described between the different new species and
P. coccinea is at most of the same order as morphological variation in other common
widespread species (e.g. P. vitifolia, P. foetida, P. suberosa, P. alnifolia, P. capsularis,
P. mixta, P. cumbalensis, P. maliformis, or P. emarginata). He reported a high level of
sexual compatibility with the other common Distephana species, P. vitifolia, which raises
the expectation of sexual compatibility with the even closer “true” P. coccinea. Thus we
have stuck to the treatment of P. coccinea by Escobar (1988), whose quantitative
description is more precise than the original by Aublet (1775) but not fundamentally
different. Within subgenus Passiflora, P. maliformis, P. serrulata and P. multiformis
constitute other cases of possible overclassification, as they are mostly differentiated by
the degree of lobation of their leaves, a trait that is quite variable in many other species,
including other Tiliifoliae, such as P. ligularis (Killip, 1938; obs. pers.). A wider
problematic group is the series Laurifoliae, with ten species in Colombia, always difficult
to identify from incomplete specimens. Although they probably constitute a very young
group and they exhibit a high number of common traits, species of subgenus Tacsonia are
relatively easy to differentiate. Particularly interesting are the endemics of section
Colombiana, from the center of the cordilleras, often characterized by a very long
peduncle and linear-lanceolate stipules, and from the northeast and up to the Venezuelan
Andes. Several authors have reported easy interspecific hybridization in subgenus
Tacsonia, involving cultivated as well as wild materials (Escobar, 1985). This
36
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
phenomenon,
by
producing
spontaneous
off-types, may have
led to some
overclassification in this subgenus. Indeed, of the 30 species reported here for Colombia,
five (P. cremastantha Harms, P. formosa Ulmer, P. pamplonensis Planch. & Linden ex
Triana & Planch., P. purdiei Killip, P. rigidifolia Killip) are known only from the type
material. Whether this is due to high endemism, ancient extinction, or off-types resulting
from hybridization cannot be ascertained, unless a second specimen is recorded, as we
did for P. linearistipula. It is important to note that P. formosa was described as a new
species from the same specimen considered as an off-type of P. lanata (Juss.) by Escobar
(1988). Overclassification may be suspected even in better-known species, as P. parritae
(Mast.) Bailey, and P. jardinensis L.K. Escobar. Indeed, in populations of the former, we
have observed sufficient morphological variation to include the few known specimens of
the latter species, which might simply represent a small isolated population. On the other
hand, most endemics of subgenus Tacsonia were found in highlands of difficult access,
and more species can reasonably be expected to be described from relatively poorly
explored areas such as the South of Tolima, Santander and Norte de Santander
departments.
Our list ranks Colombia as the country with the highest richness of Passifloraceae,
followed by Brazil with 127 species. Figure 5 allows comparisons for species richness
and relative diversity of passion flowers in the Neotropics, showing the strong influence
of latitude (typical of a tropical distribution) and topography on Passiflora diversity.
Colombian species richness and diversity is more than twice that of Peru and Venezuela,
two countries of similar surface and latitude. Given its much smaller area, Ecuador also
presents an impressive diversity. Thus, the northern Andes of Colombia and Ecuador
clearly constitute the center of diversity for the genus Passiflora. This is probably due to
the greater availability of habitats, especially at high elevations, in these two countries.
The presence of three Andean cordilleras in Colombia very probably played a significant
role. Indeed, radiation has been very active in the northern Andes, with particular
contribution of recent and fast evolving groups, such as subgenera Rathea and Tacsonia,
accounting for more than 41 highland species in Colombia and Ecuador. Among them, 21
37
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
(14%) species are endemic to Colombia. Colombian highlands are also rich in
representatives of subgenus Decaloba.
Figure 5. Distribution of Passifloraceae species richness in American countries, according to information
gathered from Killip (1938, 1960), Escobar (1988, 1989, 1990 inedited, 1994), Holm-Nielsen et al. (1988),
Jørgensen & León (1999), MacDougal (1994), Vanderplank (2000), Deginiani (2001), Tillet (2003), Ulmer
& MacDougal (2004), records of the herbaria cited in this study and many journal articles related with the
description of new species present in the America.
38
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
According to Escobar (1988a), 40% of the New World Passifloraceae are found in the
Andes. In Colombia, habitats between 1000 and 3,000 m account for only 27% of the
land area, yet 81% of the species of Passifloraceae are found there. With 123 species, the
Andean region concentrates the highest richness, mainly between 1000 and 2,000 m. The
Caribbean region shares the highest proportion of species (27) with the Andean region
(Table 1), which is mostly due to the presence of the Sierra Nevada de Santa Marta
mountain range in northern Colombia, with a steep gradient of elevation from the
Caribbean Sea to summits at 5,775 m. The increase of species richness and endemism
with the elevation is generally interpreted as a result of the increasing isolation and
decreasing habitat surface in high mountain regions, leading to small, fragmented
populations which are prone to speciation (Simpson, 1975; Jørgensen et al., 1995).
Another contribution to the particular species richness in Colombia and Ecuador is that of
the Pacific Coast region, continuous with the similar highly diverse ecosystems of
Central America (Chocó-Darién/Western Ecuador hotspot of Myers et al., 2000), and
receiving one of the highest rainfalls in the world, in strong contrast with the conditions
prevailing in the westerns Andes and coast of Peru that are arid or semi-arid, or the drier
and more contrasted climate of Venezuela. The Passifloraceae species composition of this
region appears both diverse and well-differentiated when compared with that of the other
biogeographic regions (Figure 2), heightening its interest. This is not surprising, as the
Chocó region is recognized as one of the most diverse biotas in the world, with nearly
40% endemism (Gentry, 1986).
Until recently, the genera Dilkea and Ancistrothyrsus were only known from the Amazon
basin. The description of A. antioquiensis by Escobar (1990 ined.) in the Andes and the
observation of Dilkea retusa in the Andes and Pacific regions extend their distribution to
other important biota.
The distribution of Passifloraceae has been drastically affected by deforestation,
principally in the Andean region. Its historical range corresponds to a region with a long
history of livestock and agriculture that now supports extensive plantations of coffee,
sugar cane, rice, bananas, and potatoes. According to our field observations, very
39
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
common species, such as P. adenopoda, P. alnifolia, P. capsularis, P. coriaceae,
P. rubra, P. suberosa, and P. mixta, are mostly species that thrive in secondary forest or
disturbed old-growth forest. Human disturbances may even have contributed to extend
their distribution, as reported in other plants (Svenning, 1998).
According to Myers et al. (2000) and Robbirt et al. (2006), rarity and endemism represent
two factors of particular significance in the consideration of risk of decline and
extinction. In this context, most Colombian Passifloraceae (70.7%) are under some threat
degree according to IUCN criteria. Only 29.3% fall in the ‘least concern’ category (LC),
which clearly illustrates the alarming situation for the family (Figure 3). Our results are
consistent with the Red List of plants published by the von Humboldt Institute (Calderón,
2005), based on the 141 species listed by Hernández & Bernal (2000), with similar
percentages for each category. However, this list only includes P. colombiana L.K
Escobar under the category of critically endangered species (CR), while ours places 16
species in this category.
Exploration for Passifloraceae was not possible in the protected areas of Colombia that
are of essential importance for the conservation of the country’s biodiversity, as we
lacked permission of access by the Colombian Ministry for Environment (MMA).
Another limiting factor of research for conservation purposes is the conflict situation in
many areas (Martin & Szuter, 1999; Dévalos, 2001).
Forests in the northern Andes are currently one of the major conservation priorities on a
global scale due to their fragility, biological richness, high rates of endemism and
multiple anthropogenic threats (Olson & Dinerstein, 1998). As Passifloraceae display
very high species richness, endemism and extinction risk in this area, and given their
multiple ecological interactions with many organisms, as well as their economic
potential, this family should constitute both an important target of conservation efforts
and a good indicator of their success.
40
Table 4. List of 167 Passifloraceae species of Colombia. Fifty-eigth endemic species are marked by an asterisk (*); twenty-six species new to Colombia by the
abbreviation ‘nr’; nine species probably present in the country are indicated between square brackets. New records, for a given biogeographic region, department
(abbreviated as in Tables 1 and 2) or elevation-range are indicated by bold letters. Abbreviations in bold letters in the ‘Notes’ column correspond to the plant
habits: shrub (Ab), tree (Ar), and climber (Tr). V.N and I.N. indicate vernacular and indigenous names, respectively.
Taxon
Biogeographic
Region
Genus Ancistrothyrsus Harms, 1931
Ancistrothyrsus antioquiensis L.K Escobar (ined), 1988
* nr
and
[Ancistrothyrsus hirtellus A.H. Gentry, 1992]
amz
Ancistrothyrsus tessmannii Harms, 1931
amz
Genus Dilkea Mast., 1871
Dilkea johannesii Barb. Rodr., 1885
Geopolitical
Distribution
Elevation
Collection for
Reference
90-800
Escobar &
Roldán 8819
(HUA) - Type
150-350
Gentry & Stein
47114 (MO) Isotype
ama pu
50-400
amz
va
Dilkea parviflora Killip, 1938
amz
Dilkea retusa Mast.,1871
amz and pac
Bibliographic
Reference
IUCN
Category
Notes
F.J Roldán
(com. pers),
Escobar (1990
inedited)
Gentry 1992
CR
Tr
Vester & Matapi
639 (COAH)
Holm-Nielsen et
al. 1988
CR
Tr
Reported in
the
Ecuadorian,
Peruvian and
Venezuelan
Amazon.
Tr
100-500
Soejarto 2461
(HUA)
Killip 1938
CR
Tr
ama cq va
100-500
Gentry 64981
(MO)
Holm-Nielsen et
al. 1988
LC
ama ant cho
cq gv met pu
snt va vc
100-500
López et al.
5947
(COAH)
Killip 1938;
Uribe 1955b;
Holm-Nielsen
1974; HolmNielsen et
al.1988
LC
Tr
V.N.: Canilla
de Tente,
Tripa de
Tente (ama).
Edible fruit
Tr
ant
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Genus Passiflora L., 1753
Subgenus Astrophea (DC.) Masters, 1871
Section Astrophea
Passiflora callistema L.K. Escobar, 1994 *
car
bl
100
E. Forero 487
(COL) - Type
Escobar 1990
Inéd., 1994
CR
Tr
Only known
from the type.
Section Botryastrophea
Passiflora holtii Killip, 1938
amz
ama cq gn va
150-500
Jaramillo 7890
(COL)
Killip 1938;
Escobar 1990
Inéd., 1994
LC/NT
Passiflora pyrrhantha Harms, 1926
amz
va
400-1000
Shultes &
Cabrera 12693
(COL)
Killip 1938;
Holm-Nielsen et
al.1988; Escobar
1990 Inéd., 1994
EN/CR
Tr
I.N:
Guachique,
Bejuco (ama).
Edible fruit
Tr
Passiflora securiclata Mast., 1893
amz ori
ara by gv va
vch
150-500
Betancourt et al.
9753 (COAH)
Killip 1960;
Escobar 1990
Inéd., 1994
LC
Tr
Passiflora spicata Mast., 1872
amz
gv
150-500
Cuatrecasas
7397 (COL)
VU
Tr
Passiflora spinosa (Poepp. & Endl.) Mast., 1871
amz and ori car
ama ant by cq
cor cun gn
met pu snt va
vch
150-500
Zarucchi 4279
(COL)
Killip 1938;
Holm-Nielsen et
al. 1988;
Escobar 1990
Inéd, 1994
Killip 1938;
Holm-Nielsen
1974; HolmNielsen et
al.1988; Escobar
1990 Inéd.,
1994
VU
Tr
V.N.:
Cocorella
(bl), Bejuco
campano (snt)
Section Dolichostemma
Passiflora citrifolia (Juss.) Mast.,1871
amz
va vch
85-500
Barbosa &
Zurucchi 2989
(COAH)
Killip 1838;
Escobar 1990
Inéd.
LC
Tr
42
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora haughtii Killip, 1938 *
and
snt
100-700
Haught 1635
(COL)
Killip 1938;
Escobar 1990
Inéd., 1994
CR
Ab
Passiflora mariquitensis Mutis ex Uribe, 1954 *
and
to
420-700
Ocampo et al.
55 (TOLI)
Killip 1938;
Escobar 1990
Inéd., 1994
CR
Ab
Formerly
considered
extinct.
Passiflora mutisii Killip, 1938 *
and
to
600
Mutis 2279
(MA) - Type
Killip 1938;
Escobar 1990
Inéd., 1994
EX
Tr
Passiflora pittieri Mast., 1897
pac
ant cho
50-1000
Gentry &
Aguirre 15318
(COL)
Killip 1938;
Escobar 1990
Inéd., 1994;
Gentry 1976
VU
Ab
Section Euastrophea
Passiflora arborea Spreng., 1826
and car
ant bl by cau
cl cun hu ma
na qu ri to vc
1000-2300
Humboldt &
Bonpland 5864
(P) - Type
Killip 1938;
Pérez 1956;
Holm-Nielsen et
al. 1988;
Escobar 1990
Inéd.
NT
Passiflora lindeniana Planch. ex Triana & Planch., 1873
and
cun ns snt
1000-2700
Linden 1409
(P) - Type
Escobar 1994
NT
Ar
V.N:
Cherimoyo
(vc),
Granadillo
arboreo (cun).
Edible fruit
Ab
Passiflora emarginata Humb. & Bonpl., 1813 *
and pac
cau cl cho na
vc
1500-2000
Humboldt &
Bonpland
(P) - Type
Killip 1938;
Escobar 1990
Inéd., 1994
LC
Passiflora engleriana Harms, 1894 *
and
ant
1500-2500
Escobar 8853
(COL)
Killip 1938;
Escobar 1990
Inéd., 1994
VU/EN
Passiflora macrophylla Spruce ex Mast., 1883
amz and pac
ant cau cho
pu na
60-1800
Alcázar &
Salgado 1203
(CAUP)
Killip 1938;
Holm-Nielsen et
al. 1988;
Escobar 1990
Inéd., 1994
43
LC
Ar
Edible fruit
Ar
Ab
V.N: Acaba
familia (cho)
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora putumayensis Killip, 1938
and
pu
1350-2500
Mora-0. 3438
(PSO)
Killip 1938;
Escobar 1990
Inéd., 1994
EN/CR
Ab
Passiflora sphaerocarpa Triana & Planch., 1873 *
and
ant ce cun na
ns qu ri snt to
vc
400-1700
Schlim 285 (P)
- Type
Killip 1938;
Uribe 1972;
Escobar 1990
Inéd., 1994
LC/NT
Passiflora tica Gomez-Laur. & L.D. Gómez, 1981
pac
ant cho
450-1500
Escobar 2192
(HUA)
Escobar 1990
Inéd., 1994
LC/NT
Ar
V.N: Gulupo
de Arbol
(cun),
Capafraile
(to). Edible
fruit
Ar
Section Pseudoastrophea
[Passiflora costata Mast., 1872]
amz
50-350
Spruce 1670 (K)
- Type
Killip 1938;
Escobar 1990
Inéd., 1994
Passiflora grandis Killip, 1938 *
and
1000-2000
Schlim 585
(K)
Escobar 1990
Inéd., 1994
EN/CR
Tr
Reported in
the Amazon
of Peru,
Brazil,
Guianas, and
Venezuela)
(confluence
of the rivers
Rio Negro
and
Casiquiare).
Ar
[Passiflora ovata Martin ex DC., 1828]
ori
0-150
Colector n.v.
Killip 1938;
Escobar 1990
Inéd., 1994
Passiflora phaeocaula Killip, 1927
amz ori
gn va vch
150-1100
Madriñan 893
(MO,GH)
Killip 1938;
Holm-Nielsen
1974; Escobar
1990 Inéd., 1994
LC/NT
Tr Ab
Passiflora skiantha Huber, 1960
amz
gv
150-500
Cuatrecasas
7366 (COL)
Killip 1938;
Escobar 1990
Inéd.
NT/VU
Tr
ns snt
44
Tr
Reported in
the Amazon
of Venezuela.
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
[Passiflora tessmannii Harms,1926]
amz
Passiflora venosa Rusby
and pac
Subgenus Decaloba (DC.) Rchb., 1828
Section Cieca
Passiflora apoda Harms,1929
50-500
Tessmann 4385
(N) - Type
Killip 1938;
Escobar 1990
Inéd., 1994
cho
50-450
Juncosa s.n.
(JAUM) n.v.
Killip 1938;
Escobar 1990
Inéd.
VU/EN
Tr
Reported in
the northern
Amazon of
Peru.
Tr
and
ant cau cl qu
na ri to vc
1900-3260
Hazen 9688
(MO) - Isotype
Killip 1938;
Hernández 2003
LC/NT
Tr
Passiflora coriacea Juss.,1805
and car pac
250-1500
Uribe 2565
(COL)
Croat 1978;
Holm-Nielsen et
al. 1988
LC
Passiflora holosericea L.,1753
car
ant by cau cl
cho cun hu
ma ns qu ri
snt to vc
at bl ce
0-1400
Cuadros-H 1882
(COL)
Killip 1938
LC/NT
Tr
Passiflora megacoriacea Porter-Utley, 2003 nr
car
bl
100-200
Killip & Smith
14415 (US)
Porter-Utley
2003
DD
Tr
Passiflora pallidaL., 1753 nr
car
at bl ma sp
0-200
Dugand &
Jaramillo 2844
(COL)
Porter-Utley
2003
LC
Tr
Appel
Monkey (sp)
Passiflora sodiroi Harms, 1922 nr
and
cau
1850-2150
Escobar et al.
4368
(PSO)
Holm-Nielsen et
al. 1988
EN/CR
Passiflora suberosa L., 1753
and car
ant cau cl cun
gv na ns qu
snt suc to vc
200-2200
Cuatrecasas
15930
(VALLE)
Holm-Nielsen et
al. 1988
LC
Tr
V.N.:
Curubita de
Monte (ant)
amz and car ori
pac
ama ant bl by
cau cl cho cor
cq cun gn gv
met na ns pu
qu snt to va
vc vch
0-1500
Killip &
Cuatrecasas
58988
(VALLE)
Holm-Nielsen
1974; HolmNielsen et al.
1988
LC
Tr
V.N: Rejito
(cun)
Section Decaloba
Series Auriculatae
Passiflora auriculata Kunth, 1817
45
Tr
V.N.: Ala de
Murcielago
Tr
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Series Sexflorae
Passiflora sexflora Juss., 1805
and
ant hu qu to
vc
1700-2300
Zurucchi et al.
5813
(CHOCO)
Holm-Nielsen et
al. 1988
NT/VU
Series Luteae
Passiflora filipes Benth., 1843
and
qu ri vc
950-1250
Silverstone 7205
(CUCV)
Holm-Nielsen et
al. 1988
VU
Tr
Series Miserae
Passiflora misera Kunth, 1817
and car ori pac
ant at ara bl
by cau cl cho
cor cun cs lg
ma met vc ns
0-1050
E. Forero 9936
(COL)
Killip 1938
LC
Tr
Passiflora tricuspis Mast., 1872 nr
and
met
1220-2000
Estrada et al.
146 (MA)
Killip 1938
CR
Tr
[Passifora trifasciata Lemaire, 1868]
amz
ama pu
130-1100
Brandbyge et
al.33556 (AAU)
Killip 1938;
Nielsen et al.
1988
Series Punctatae
Passiflora alnifolia Kunth, 1817
and car
ant by cau cl
cun ma na pu
qu ri snt to vc
1400-2500
Hno. Daniel
2803 (MEDEL)
Holm-Nielsen et
al. 1988
LC
Tr
Passiflora andreana Mast., 1883
and
ant cau cun
ma na qu ri
1500-3150
Garcia-B.12949
(COL)
Holm-Nielsen et
al. 1988
CR
Tr
Passiflora azeroana L. Uribe, 1955 *
and
by cun hu snt
2500-3000
Lozano 3718
(COL)
Uribe 1957
NT/VU
Tr
Passiflora biflora Lam., 1789
and car
ant at bl ce cl
cho cun hu
ma met na ns
ri sp snt to vc
0-1500
Garcia-B. 11720
(COL)
Killip1938;
Holm-Nielsen
1974; Croat
1978
46
Tr
V.N.: Corvejo
(na)
Tr
Reported in
the Amazon
of Ecuador,
Brazil and
Peru.
Ornamental
(qu)
LC
Tr
V.N.: Peyen
Papaya (sp),
Desjarretader
a (cun)
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora bogotensis Benth., 1845 *
and car
by cun hu lg
ma ns snt vc
2000-3700
Garcia-B. 15291
(COL)
Killip 1938;
Holm-Nielsen
1974
LC
Tr
V.N.: Curubo
macho (cun)
Passiflora bucaramangensis Killip, 1930 * nr
and
snt
1500-2600
Killip & Smith
16787
(MO) - Isotype
Killip 1930,
1938
EN
Tr
Passiflora candollei Tr. & Planch., 1873 nr
amz
ama
100
Rudas et al.
2180
(COL)
Killip 1938
NT
Tr
Passiflora chelidonea Mast., 1979
and car pac
ant ara cau
cho na ns pu
ri snt vc
900-3000
Cuatrecasas
12526
(COL)
Holm-Nielsen et
al. 1988
LC
Tr
Passiflora cuneata Willd.,1809
and car
ant by cho
cun ma met
ns snt vc
900-3000
Uribe 5973
(COL)
Killip 1938;
Hno. Daniel
1968; HolmNielsen 1974
LC
Passiflora cuspidifolia Harms, 1893
and
by cun snt
2000-3200
Prieto 302
(UIS)
Holm-Nielsen et
al. 1988
LC
Tr
V.N.:
Granadillita
de Monte
(ant)
Tr
Passiflora dawei Killip, 1930 *
and
cun snt
900-1600
Idrobo 2037
(COL)
VU/EN
Passiflora erytrophylla Mast.,1872 *
and
by cun
1600-2790
Ocampo et al.
54 (HUA)
Killip 1930,
1938;
Hernández 2003
Killip 1938;
Uribe 1955a
Passiflora lyra Planch. & Lind. ex Killip, 1846 nr
and
ant
400-840
MacDougal
4161 (HUA)
Killip 1938
NT/VU
Tr
Passiflora magdalenae Triana & Planch., 1873 *
and
cl cun to
200-1200
Uribe 2568
(COL)
Killip 1938;
Pérez 1956
NT/VU
Passiflora micropetala Mast., 1872
amz and
0-710
Perez-A.669
(COL)
Holm-Nielsen et
al. 1988
Passiflora mollis HBK., 1817 * nr
and
ama ant by
cho cq met pu
vc
ant cl cho qu
lg snt to vc
Tr
V.N.:
Granadillo del
Magdalena.
Tr
1400-2500
Humboldt &
Bonpland
(P) - Type
Killip 1938;
Hno. Daniel
1968
47
EN
LC
LC/NT
Tr
Tr
Not collected
since 1938.
Tr
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora monadelpha Jørgensen & Holm-Nielsen,
1987 nr
and
to vc
2800-3310
Escobar 4859
(HUA)
Holm-Nielsen et
al. 1988;
Hernández 2003
VU/EN
Tr
Passiflora munchiquensis.Hernández (ined), 2003 * nr
and
cau vc
1900-3200
Vargas 3909
(HUA)
NT/VU
Tr
Passiflora occidentalis Hernández (ined), 2003 * nr
and pac
cau cho na
pu vc
50-1200
Killip 39025
(COL)
LC/NT
Tr
Passiflora panamensis Killip, 1922
pac car
ant bl cho cor
0-500
Zarucchi et al.
5107
(CHOCO)
Hernández
2003;
A.Hernández
(com. pers).
Hernández
2003;
A.Hernández
(com. pers).
Killip 1938
Passiflora pilosissima Killip, 1931 * nr
and
ant vc
1500-2100
Lehmann 7630
(US)
Passiflora popayanensis Killip, 1930 *
and
cau
2400-2900
Passiflora punctata L., 1753
and pac
cau cun na vc
[Passiflora sandrae J. MacDougal, 2006]
pac
Passiflora tribolophylla Harms, 1922 *
NT
Tr
V.N.: Gulupa
(ant)
Killip 1938
CR
Tr
Lozano 6472
(COL)
Killip 1938
VU/EN
Tr
20-1750
Romero-C. 3150
(COL)
Croat 1978;
Holm-Nielsen et
al. 1988
LC/NT
Tr
cho
800-1100
Garwood 1178
(MO) - Type
MacDougal
2006
pac
ant cau cho
50-1820
Lehmann 5420
(foto, COL)
LC/NT
Passiflora tuberosa Jacq., 1804 nr
and
vc
1200
Cuatrecasas
15930 (VALLE)
Killip 1938;
Hno. Daniel
1968
Killip 1938
Tr
Collected in
the border of
Panama and
Colombia
(cho)
Tr
EN
Tr
Passiflora ursina Killip & Cuatrec., 1960
and
ant na vc
2100-3100
Roldán 2345
(HUA)
VU/EN
Tr
Passiflora vespertilio L.,1753
amz and ori
ama met na
150-500
Plowman 2425
(COL)
Killip 1960;
Holm-Nielsen et
al. 1988;
Hernández 2003
Holm-Nielsen et
al. 1988
LC/NT
Tr
48
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Section Hahniopathanthus
Passiflora guatemalensis S. Watson, 1887
and car
ant cl lg ma
qu ri to vc
0-1580
Uribe 2532
(COL)
Passiflora hahnii (Fourn.) Mast., 1872 nr
and car
ant cl lg ma
to vc
100-1250
Killip & Hazen
8670 (Y)
Section Pseudodysosmia
Passiflora adenopoda Moc. & Sessé ex DC., 1828
and car
ant by cl cun
cau ma qu ri
to vc
100-2100
Cuatrecasas
15703
(VALLE)
Passiflora lobata (Killip) Hutch. ex J.M.
MacDougal,1986
pac
ant cho
0-1200
Gentry 23791
(COL)
Passiflora morifolia Mast., 1872
and
na
500-1000
Karsten s.n. (W)
n.v.
Section Pseudogranadilla
Passiflora bicornis Mill., 1768
car
ant at bl lg ma
0-500
Saravia 3643
(COL)
Passiflora hirtiflora Jørgensen & Holm-Nielsen, 1987 nr
and
ns
2650
Escobar 3152
(HUA)
Passiflora kalbreyeri Mast., 1883 *
and car
ce ns snt
1100-3100
Killip 20284
(COL)
Killip 1938
LC/NT
Tr
Passiflora menispermacea Triana & Planch., 1873 *
and
to
1400-3000
Cuatrecasas
9247 (MA)
Killip 1938
LC
Tr
Section Xerogona
49
LC
Tr
CR
Tr
V.N.:
Granadilla
Abroquelada
(ant)
Holm-Nielsen et
al. 1988;
MacDougal
1994
LC
MacDougal
1994; Ulmer &
MacDougal
2004
Killip 1938;
MacDougal
1994.
NT
Tr
V.N: Pegajosa
(qu),
Granadilla
Culebra (vc),
Gulupo (cun).
Edible fruit
Tr
EN
Tr
Killip1938;
Holm-Nielsen
1974
LC
Tr
V.N.: Cachito
de Venado
(bl), Cinco
Llagas (at)
Tr
Killip 1938;
Ulmer &
MacDougal
2004
Uribe 1955b;
Holm-Nielsen
1974; Ulmer &
MacDougal
2004
CR
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora capsularis L., 1753
and car pac
ant cl cun cho
hu lg ma na
ns qu snt to
vc
100-2000
Uribe 2566
(COL)
Passiflora costaricensis Killip, 1922
pac
cho
20-1500
Croat 42591
(HUA)
Passiflora escobariana J.M. MacDougal, 1992
and
ant
1090-1100
MacDougal
3823 (HUA) Isotype
Passiflora rubra L., 1753
and car
ant cl cau
cun hu lg pu
na pu qu ri to
vc
500-2000
Garcia-B. 17279
(COL)
Subgenus Dysosmia (DC.) Killip, 1938
Passiflora foetida var. eliasiiKillip, 1938
car
at bl ma
0-500
Penell 12029
(N)
amz and car ori
pac
ama ant ara
at bl by cau
ce cor cq cs
cun cho gn gv
hu lg ma met
na ns qu snt
suc to va vc
0-1500
Schultes 22576
(COL)
Passiflora foetida var. gossypiifolia (Desv.) Mast. 1872
50
LC
Tr
NT
Tr
VU
Tr
LC
Tr
V.N.:
Chulupa de
Monte (cl)
Kiliip 1938
VU
Killip 1938;
Martin &
Nakasone 1970;
Romero-C.
1991; Ulmer &
MacDougal
2004; Ulmer &
Ulmer, 2005
LC
Tr
V.N.: Flor de
la Pasión,
Pasionaria
(at)
Tr
V.N.:
Granadilla
(cho), Flor de
la Pasión (at),
Gulupo (cun),
Bejuco
Canastilla
(met),
Chulupa de
Loma (ant
hu), Cinco
Llagas (cor).
I.N.: Iñanaleeg murulale
(ama). Edible
fruit
Killip 1938;
Holm-Nielsen
1974; HolmNielsen et al.
1988
Holm-Nielsen
1974; HolmNielsen et al.
1988
MacDougal
1992; Ulmer &
MacDougal
2004
Holm-Nielsen et
al. 1988
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora foetida var. hispida (DC.) Killip ex Gleason,
1931
and car
ant bl cun ns
to
0-1500
Killip & Smith
21000 (N)
Killip 1938;
Ulmer & Ulmer,
2005
LC
Passiflora foetida var. isthmina Killip, 1938
and pac
na snt vc
0-1200
Killip 5289 (N)
Killip 1938
VU
Passiflora foetida var. moritziana (Planch.) Killip ex
Pull, 1937
car
ma
0-500
Killip & Smith
21088 (N)
Killip 1938
VU
Tr
V.N.: Flor de
la Pasión (ma)
Passiflora foetida var. sanctae-martae Killip, 1938 * nr
car
ma
0-500
Smith 1532 (P)
Kiliip 1938
EN
Tr
Flor de la
Pasión (ma)
Passiflora vestita Killip, 1938
amz
pu
0-500
Betancourt 5164
(MO) n.v.
Killip 1938;
Holm-Nielsen et
al. 1988
Distephana (Juss.) Killip, 1938
Passiflora coccinea Aubl.,1775
amz ori
ama cs gn gv
met na va vch
150-1500
Davidse 5321
(COL)
Escobar 1988a
LC
Passiflora involucrata (Mast) A.H. Gentry, 1981
amz
ama cq va
150-350
Schultes 6923
(COL)
Escobar 1988a
LC
Tr
V.N.: Lluvia
Padie,
Granadillo de
Conga (ama),
Granadilla
colorada (cs).
Edible fruit
Tr
Passiflora glandulosa Cav., 1790
amz
va
150-500
Romero-C. 3668
(AAU) n.v.
Killip 1938;
Holm-Nielsen
1974
EN
Tr
Passiflora quadriglandulosa Rodschied, 1796
amz
ama gu
150-500
Lozano 604
(COL)
Escobar 1988a;
Holm-Nielsen et
al. 1988
LC/NT
Tr
51
VU/EN
Tr
V.N: flor de
la pasión
(ma), gulupo
(cun)
Tr
V.N.: Flor de
la Pasión (vc)
Tr
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora variolata Poepp. & Endl., 1838
amz
ama cq va
150-500
Zarucchi 2197
(COL)
Escobar 1988a
LC/NT
Passiflora vitifolia Kunth, 1817
amz and car ori
pac
ama ant bl by
cau ce cl cho
cor cq cun lg
gv ma met na
pu ri snt to
va vc vch
0-1800
Cuatrecasas
15740
(VALLE)
Killip 1938;
Romero C.
1956, 1991;
Martin &
Nakasone 1970;
Holm-Nielsen
1974; HolmNielsen et al.
1988
LC
and
by cau cl cun
na ns qu snt to
vc
1400-2700
Richter s.n.
(COL)
Jussieu 1805;
Holm-Nielsen
1974; Escobar
1988a
LC
Tr
V.N.: Tacso
(na), Curubo
de Monte (qu
ns).
amz and pac
ama cho ns
ant
0-1000
Renteria 3542
(COL)
Killip 1938;
Holm-Nielsen
1974; HolmNielsen et al.
1988; RomeroC. 1991
LC
Tr
V.N.:
Cocorilla
(cho).
Granadilla,
Naracujinha
(ama).
N.I.: CipoCipo
Naracujinha
(ama).
Subgenus Manicata (Harms) Escobar, 1988 (Syn.
Granadillastrum)
Passiflora manicata (Juss.) Pers., 1807
Tr
V.N.:
Granadilla,
Oncilla,
Parcha de
Culebra de
Agua (ama)
Tr
V.N.:
Chulupo (cq),
Granadilla de
Monte (cho),
Granadillo
(met cq),
Gulupa (to).
I.N.: Maloca
de Fisi (ama).
Edible fruit
Passiflora (Medik.) Mast., 1871 (Syn. Granadilla)
Series Digitatae
Passiflora serratodigitata L.,1753
Series Laurifoliae
52
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora ambigua Hemsl. ex Hook., 1902
amz and ori pac
ant by cl cho
cun hu ma
met pu snt vc
0-2000
Fuchs 21744
(COL)
Holm-Nielsen et
al. 1988
LC
Tr
Edible fruit
Passiflora gleasonii Killip, 1924
ori
gn
150-500
Madriñán 1014
(COL)
Killip 1938
EN
Tr
Passiflora guazumaefolia Juss., 1805
and car
ce cor bl ma
snt
0-500
Uribe 2405
(COL)
Killip 1938;
Coppens 2003
Passiflora killipiana Cuatrecasas, 1960 nr
amz
cq
250-500
Schultes 5875
(US)
Killip 1960
CR
Passiflora laurifolia L., 1753
and amz pac
0-1700
Zarucchi 1824
(COL)
Killip 1938
LC
Tr
Edible fruit
Passiflora nitida Kunth, 1817
amz and car ori
pac
ama cho cq
gv hu met snt
va
ama ant cho
cq cun cs gn
gv ma met na
pu va vc
0-1940
Triana 2931
(P)
Killip 1938;
Romero-C.
1956, 1991;
Holm-Nielsen
1974; García-B.
1975; Croat
1978
LC
[Passiflora phellos C. Feuillet, 2004]
amz
90-150
Wurdack &
Addeley 43479
(NY) - Holotype
Feuillet 2004
Tr
V.N.:
Granadilla
(cho met),
Granadilla
Babosa (na).
N.I.:
Burucuña,
Gemarundare,
Tuchica, JinoGojé (va).
Edible fruit
Tr
Reported in
the Amazon
of Brazil,
Peru and
Venezuela
53
LC/NT
Tr
V.N: La
Parcha (ce),
Cocorilla
(ma).
Edible fruit
Tr
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora popenovii Killip, 1922 nr
and
cau na vc
1200-2050
Escobar &
Escobar 1017
(HUA)
Killip 1938;
Holm-Nielsen et
al. 1988;
Romero-C.
1991; Ulmer &
MacDougal
2004
Passiflora riparia Mart. ex Mast., 1872
amz
cq pu va
300-400
Smith 3157
(US)
Killip 1960
LC/NT
Passiflora tolimana Harms, 1894 *
and
ant to vc
820-2000
Echeverry 3627
(TOLI)
Killip 1938
NT/VU
Series Incarnatae
Passiflora cincinnata Mast., 1868 nr
and
ns
1200
Killip & Smith
20879 (Y)
Killip 1938
CR
Passiflora edulis f. edulis Sims, 1818
amz and pac
ant cl cau cho
cun gv met
na qu ri snt
to vch vc
1100-2750
Idrobo 1637
(COL)
Holm-Nielsen et
al. 1988;
Vanderplank
2000; Ulmer &
MacDougal
2004
NE
Passiflora edulis f. flavicarpa Degener, 1932 nr
amz and car ori
pac
ant ara bl ce
cl cho cun gn
hu met pu qu
snt ri to vc
0-1800
Silvestone
14399 (CUVC)
Killip 1938;
Ulmer &
MacDougal
2004
NE
Series Kermesinae
54
EW
Tr
V.N:
Granadilla de
Quijos (na),
granadilla
caucana,
curubejo
(cau).
Cultivated.
Edible fruit
Tr
Edible fruit
Tr
Edible fruit
Tr
Ornamental
(qu).
Edible fruit
Tr
Introduced
from Brazil in
the 1950s.
V.N: Curuba
Redonda (ant
cl ri qu),
Gulupa (cun).
Cultivated or
feral. Edible
fruit
Tr
Introduced
from Brazil in
the 50s.
V.N.:
Maracuyá.
Cultivated.
Edible fruit
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora lehmanni Mast., 1885 *
amz and
Passiflora smithii Killip, 1930
and car
Passiflora trisulca Mast., 1887 *
Killip 1938;
Holm-Nielsen
1974
Killip 1938;
Holm-Nielsen
1974
LC
Tr
LC
Tr
V.N.: Curuba
Silvestre (to).
Marulanda 91
(HUA)
Killip 1938;
Hno. Daniel
1968
NT
Tr
1000-2700
Ocampo 83
(VALLE)
Deginani 2001
2450-2500
Cuatrecasas
1808
(COL)
Killip 1938
2600
Holm-Nielsen et
al. 6200 (AAU)
Holm-Nielsen et
al. 1988
450
Uribe 1334 (US)
ant cau cl
cun hu pu qu
ri snt vc
cun ma qu snt
to vc
1000-2000
Uribe 2588
(COL)
500-2000
Killip & Smithii
15015 (MO) Holotype
and
ant cl vc
1300-1800
Series Lobatae
Passiflora caerulea L., 1753 nr
and
cl cun qu
Passiflora gritensis H. Karst., 1859
and
by ns
[Passiflora montana Holm-Nielsen & Lawesson, 1987]
and
Passiflora picturata Ker, 1822 nr
LC/NT
Tr
Introduced
from
Argentina.
Ornamental.
Edible fruit
Tr
Killip 1938,
1960
DD
Tr
Collected on
the border of
Ecuador and
Colombia
(na)
Tr
Passiflora pennellii Killip, 1924 *
and
ant cun
1200-1600
Uribe 4827
(COL)
Killip 1938
NT/VU
Tr
Passiflora resticulata Mast. & André, 1884
amz and pac
cau gv na vc
0-2000
Marulanda &
Márquez 1665
(HUA)
NT/VU
Tr
Passiflora semiciliosa Planch & Linden, 1873 *
and car
ma ns
1850-3000
Garcia-B. 20749
(COL)
Killip 1938;
Holm-Nielsen
1974;HolmNielsen et al.
1988
Killip 1938
55
VU
Tr
V.N.: Gulupa,
Palcha (ns)
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora subpeltata Ortega, 1798
amz and car ori
ant bl cau ce
cor cq cs cun
ma suc to vc
0-2400
Ramírez et al.
11507 (CAUP)
Killip 1938;
Holm-Nielsen
1974
LC
Tr
V.N.:
Cocorilla (bl)
Series Quadrangulares
Passiflora alata Curtis, 1788 nr
amz
ama
200
Ocampo 82
(VALLE)
Killip 1938;
Ulmer &
MacDougal
2004
DD
amz and car ori
pac
ama ant bl
cau cl cho cq
cun gn hu ma
met na ns qu
va ri snt to vc
0-1500
Gentry 15371
(COL)
Killip 1938;
Romero-C.
1956, 1991;
HolmNielsen1974;
Holm-Nielsen et
al. 1988
LC
Tr
Introduced
(qu vc) from
Brazil in the
90s. V.N.:
Maracúa.
Cutivated.
Edible fruit
Tr
V.N.: Badea
(ant cl hu cun
met qu ri),
corvejo (snt),
Granadillo
Grande (cau),
curuba (vc),
Motorro (gn).
Cultivated.
Edible fruit
Series Menispermifoliae
Passiflora chocoensis G. Gerlach & T. Ulmer, 2000 * nr
pac
cho
0-100
Gerlach 434917
(COL) Holotype
CR
Tr
Passiflora menispermifolia Kunth, 1817
amz car and pac
ant bl by cho
cor cq cun
met na ns snt
to vc
0-2140
Cuatrecasas
15541
(VALLE)
Gerlach &
Ulmer, 2000;
Ulmer &
MacDougal
2004
Croat 1978;
Holm-Nielsen et
al. 1988
LC
Tr
V.N.:
Chulupe (cq)
Series Simplicifoliae
Passiflora danielii Killip, 1960 *
and
ant
1300-2600
Passiflora longipes Juss., 1805 *
and
cun by qu snt
to
2500-3500
Hno. Daniel
1536
(MEDEL) Isotype
Sanchez 17
(COL)
Passiflora quadrangularis L.,1759
56
Killip 1960;
Hno. Daniel
1968
VU/EN
Tr
Killip 1938
NT
Tr
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora longipes var. oxyphylla L. Uribe, 1977 *
and
by ns snt
2000-2600
Cadena 83
(UIS)
Uribe 1977
NT
Tr
Passiflora oerstedii Mast., 1872
and ori pac
ant cau cho
cun cau met
na qu ri vc
0-2000
Romero-C. 6141
(COL)
Killip 1938;
Holm-Nielsen
1974; HolmNielsen et al.
1988
LC
Tr
Series Tiliaefoliae
Passiflora ligularis f. lobata (Mast.) Killip, 1938 nr
and
ant
1800-2000
Archer 1498
(COL)
Killip 1938
NT
Passiflora ligularis Juss., 1805
and
ant cl cun cau
by cho hu
met na ns pu
qu ri snt to
vc
1550-2500
Dombey 739
(P) - Type
Killip 1938;
Romero-C.
1956, 1991;
Holm-Nielsen et
al. 1988
LC
Passiflora magnifica L.K. Escobar, 1990 *
and
ant
1250-1750
Callejas 6586
(HUA) n.v.
VU
Passiflora maliformis L.,1753
and car pac
ant by cl cau
cun cho hu
ma na qu snt
to vc
0-2200
Humboldt &
Bonpland 1804
(P) - Type
Escobar 1990;
Ulmer &
MacDougal
2004
Killip 1938;
Romero-C,
1956, 1991;
Holm-Nielsen,
1974; García-B.
1975; HolmNielsen et al.
1988
Tr
V.N:
Granadilla.
Cultivated.
Edible fruit
Tr
V.N:
Granadilla;
Granadilla
Pipo (na).
N.I.: Awapit
(na).
Cultivated.
Edible fruit
Tr
Edible fruit
Passiflora multiformis Jacq.,1809
and car
lg ma ns
0-1300
Romero-C. 8992
(COL)
57
Killip 1938
LC
NT/VU
Tr
V.N: Gulupa,
Granadilla de
Piedra, o de
Hueso (cu, na
vc), Gurapa
(snt), Chulupa
(hu).
Cultivated.
Edible fruit
Tr
V.N: Palchita
(ns). Edible
fruit
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
LC
Passiflora palenquensis Holm-Niels. & Lawesson, 1987
pac
ant cau cho na
vc
0-1200
Espina & Garcia
1951
(COL)
Holm-Nielsen &
Lawesson 1987;
Holm-Nielsen et
al. 1988
Passiflora platyloba Killip, 1922
pac
cho
0-1050
Gentry &
Juncosa 40946
(COL)
Gentry 1976
NT/VU
Passiflora seemannii Griseb.,1858
amz and pac ori
ant by cho
cun gn gv met
ns snt va vc
vch
0-1300
MacDougal
4144 (HUA)
Croat 1978
LC
Passiflora serrulata Jacq., 1767
car
at ma lg
0-500
Bunch 601
(FMB)
Killip 1938;
Coppens 2003
NT/VU
Passiflora tiliifolia L., 1753
and pac
ant cau cho cl
na qu vc to
1100-2500
González 1411
(CAUP)
Killip 1938;
Holm-Nielsen
1974; HolmNielsen et
al.1988;
Coppens 2003
LC/NT
Subgenus Porphyropathanthus L.K Escobar, 1989
Passiflora sierrae L.K. Escobar, 1989 *
car
ma
3000-3700
Cuatrecasas
24375 (COL)
Escobar 1989
EN/CR
Tr
Subgenus Psilanthus (DC.) Killip, 1938
Passiflora bicuspidata (H.Karst.) Mast.,1872 *
and
by cun ns snt
2500-3500
Rojas 138
(CDMB)
Uribe 1972;
Killip 1978
VU
Tr
58
Tr
V.N.:ranadilla
(cho),
"Camelo"
(vc).
Cultivated.
Edible fruit
Tr
Edible fruit
Tr
V.N.: Palcha,
Chulupa
(met),
Granadilla
Montañera
(cun). Edible
fruit
Tr
V.N.:
Guayabita
Cimarrona
(ma).
Edible fruit
Tr
V.N.:
Granadilla,
Machimbi
(Colombia).
Cultivada.
Fruto
comestible.
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora hyacinthiflora Planch. & Linden, 1873 *
and
by ma ns
2900-3300
Garcia-B. 20700
(COL)
Killip 1938
Passiflora trinervia (Juss.) Poir., 1811 *
and
cl qu to vc
2500-3700
Cuatrecasas
20241
(VALLE)
Subgenus Rathea (Karst.) Killip, 1938
Passiflora andina Killip, 1938 nr
and
na
2800
Karsten (V)
Passiflora colombiana L.K. Escobar, 1986 *
and
na pu
3000-3600
Mora 6175
(PSO) Paratype
and
cau
2300-2800
Tryon 6001
(COL)
Passiflora cumbalensis var. cumbalensis (H. Karst.)
Harms,1894
and
na pu
3000-3800
Fernandez 5834
(COL)
Passiflora cumbalensis var. goudotiana (Triana &
Planch.) L.K. Escobar, 1987
and car
ant by cl cq
cun hu ma na
pu qu ri snt
to vc
1800-3300
Uribe 2593
(COL)
[Passiflora sanctae-barbarae Holm-Nielsen &
Jørgensen, 1987]
and
2200-2700
Harling &
Andersson
12445 (AAU) Isotype
LC/NT
Tr
Jussieu 1805;
Killip 1938
VU
Tr
Killip 1938;
Holm-Nielsen et
al.1988
Escobar 1986,
1988
CR
Tr
CR
Tr
Escobar 1987,
1988b
LC
Romero-C.
1956; HolmNielsen 1974;
Escobar 1987,
1988; HolmNielsen et al.
1988
Escobar 1987,
1988; HolmNielsen et al.
1988
LC/NT
Tr
V.N.: Curuba
de Monte.
Edible fruit
Tr
V.N.: Curuba
Roja, Tauso
(na). Edible
fruit
Subgenus Tacsonia (Juss.) Tr. & Planch, 1873
Section Bracteogama
Passiflora cumbalensis var. caucana L.K. Escobar, 1987
*
59
Holm-Nielsen et
al. 1988
LC
Tr
V.N.: Curuba
bogotana
(cun), Curubo
mucura,
curuba
rosada, Tausa
(na).
Cultivated.
Edible fruit
Tr
Reported in
the northern
Andes of
Ecuador
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora tripartita var. azuayensis Holm-Nielsen &
Jørgensen, 1988 nr
and car
ant by cun
ma ns
2000-2610
Escobar 19999
(HUA)
Holm-Nielsen et
al. 1988
LC/NT
Tr
V.N.: Curuba.
Edible fruit
(by cun)
Passiflora tripartita var. mollissima Holm-Nielsen &
Jørgensen, 1988
and car
ant by cau cl
cun ma na ns
pu snt vc
2200-3500
Romero-C 8007
(PSO)
Holm-Nielsen et
al. 1988;
Romero-C.
1991; Ulmer &
MacDougal
2004
LC
Passiflora tarminiana Coppens & Barney, 2001
and
ant by cau cl
cun hu na qu
snt to vc
2000-2900
Coppens 72
(COL) - Type
Coppens et al.
2001; Campos
2001
LC
Tr
V.N.: Curuba
de Castilla
(ant by cu cl);
Tauxso (na).
Cultivated.
Edible fruit
Tr
V.N.: Curuba
India.
Cultivated.
Edible fruit.
Section Colombiana
Series Colombianae
Passiflora adulterina L.f.,1781 *
and
by cun snt to
2600-3600
Barclay 4517
(COL)
Escobar 1988a
NT
Tr
Passiflora crispolanata L.Uribe, 1954 *
and
by cun
2500-3500
Uribe 6773
(COL)
Uribe 1954;
Escobar 1988a
NT
Passiflora cuatrecasasii Killip, 1960 *
and
by cun met
snt
2200-3500
Killip 1960;
Escobar 1988a
VU
Passiflora formosa T. Ulmer, 1999 *
and
by
3000-3100
Cuatrecasas
9479 (foto,
MEDEL)
Uribe 5945
(COL)
Tr
V.N.: Curuba
Paramera
(cun)
Tr
Ulmer 1999
EN
Tr
Passiflora lanata (Juss.) Poir., 1811 *
and
cun by snt to
2200-3500
Uribe 2587
(COL)
Jussieu 1805;
Holm-Nielsen
1974; Escobar
1988a
NT/VU
Tr
V.N.:
Granadilla
(cun)
Passiflora pamplonensis Planch.& Linden ex Triana &
Planch., 1873 *
and
snt
2000-3000
Funck & Schlim
1385 (foto,
VALLE)
Escobar 1988a
EN/CR
Tr
Curubita de
Piñuela (snt)
60
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora rigidifolia Killip, 1960 * nr
and
ant
3750
Burke 185 (K) Type
Killip 1960
Passiflora rugosa var. rugosa (Mast.) Triana & Planch.,
1873
and
cun met ns
3000-3500
Peñuela 008
(COL)
Escobar 1988a
LC/NT
Tr
Passiflora rugosa var. venezolana L.K. Escobar, 1986
and
ns snt
2500-3500
Garcia-B. 20001
(COL)
Escobar 1988a
LC/NT
Tr
Passiflora trianae Killip, 1938 *
and
ns snt
3000-3500
Escobar 569
(COL)
Escobar 1988a
VU/EN
Tr
Passiflora truxillensis Planch. & Linden, 1873 nr
and
ns
1800-3000
V. Barney & G.
Coppens (foto),
com. Personal
Escobar 1988a;
Ulmer & Ulmer
2005
EN
Tr
Series Leptomischae
Passiflora antioquiensis H. Karst., 1859 *
and
ant cau cl cun
hu pu qu ri to
vc
1800-2700
Escobar 2133
(HUA)
Hno. Daniel
1968; Uribe
1972; Garcia- B.
1975; Escobar
1988a
LC/NT
Passiflora cremastantha Harms, 1922 *
and
cau
2000-2500
Lehmann 5421
(F) - Type
Escobar 1988a
EX
Passiflora flexipes Triana & Planch., 1873 *
and
cl qu ri
2500-3380
Vargas 626
(FAUC)
Escobar 1988a
NT/VU
Passiflora leptomischa Harms, 1922 *
and
ant cau qu vc
2000-2800
Escobar et al.
4421 (PSO)
Escobar 1988a
LC/NT
Passiflora tenerifensis L.K. Escobar, 1988 *
and
vc
2800-3100
Escobar 4853
(COL)
Escobar 1988a,
1989b; Campos
2001
EN/CR
61
DD
Tr
Known only
from the type.
Tr
V.N.:
Granadilla
(vc), Curuba
Antioqueña
(ant). Wild or
cultivated in
home
gardens.
Edible fruit
Tr
Known only
from the type.
Tr
V.N.: Curuba
de Monte (cl
qu ri).
Edible fruit
Tr
Edible fruit
Tr
V.N.: Curuba
de Monte
(vc).
Edible fruit
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Series Quindiensae
Passiflora linearistipula L.K. Escobar, 1988 *
and
cl
2650-3170
Ocampo et al.
56 (HUA)
Escobar 1988a
EN/CR
Tr
Not collected
since 1984.
Passiflora quindiensis Killip, 1938 *
and
to
2900-3100
Uribe 3320
(COL)
Escobar 1988;
Campos 2001
VU/EN
Tr
Section Fimbriatistipula
Passiflora fimbriatistipula Harms, 1894 *
and
cau hu
2130-3240
Fernandez et al.
30182 (AFP)
Escobar 1988a
NT/VU
Passiflora uribei L.K. Escobar, 1988 *
and
na pu
2500-2700
Escobar et al.
2896
(HUA)
Uribe 1958;
Escobar 1988a
EN
Tr
I.N.:
Pachuaca
(cau)
Tr
Section Parritana
Passiflora jardinensis L.K. Escobar, 1988 *
and
ant
2750-3000
Zarucchi 6963
(COL)
Escobar 1988b
VU/EN
Tr
Passiflora parritae (Mast.) L.H. Bailey, 1916 *
and
cl qu ri to
2500-3020
Sánchez 15
(FAUC)
Escobar 1988a
VU/EN
Tr
Curuba de
Monte (to).
Edible fruit
Section Poggendorffia
Passiflora pinnatistipula Cav.,1799
and
ant by cun na
ns
2000-3600
Uribe 6643
(COL)
Escobar 1988b;
Holm-Nielsen
1974; HolmNielsen et al.
1988; Campos
2001
LC/NT
Tr
V.N.: Curuba
Redonda,
Gulupa (cun)
Cultivated.
Edible fruit
Passiflora x rosea (H. Karst.) Killip, 1938
and
by cun
2500-3500
Uribe 3941
(COL)
Escobar 1988a
VU
Tr
Natural
hybrid of P.
pinnatistipula
x P. tripartita
var.
mollissima.
Edible fruit
(when fertile)
Section Tacsonia
62
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora mixta L. f., 1781
and
ant cl by cau
cun na ns qu
ri snt to vc
1700-3700
Humboldt &
Bonpland
(P) - Type
Escobar 1988a;
Holm-Nielsen
1974; HolmNielsen et al.
1988; García-B.
1975
LC
Passiflora schlimiana Triana & Planch., 1873 *
car
ce lg ma
2400-3220
Romero-C. 7407
(COL)
Holm-Nielsen
1974; Escobar
1988a; RomeroC. 1991;
Coppens 2003
VU/EN
Section Tacsoniopsis
Passiflora bracteosa Planch. & Linden, 1873
and
ns snt
2200-3000
Garcia-B. 20745
(COL)
Escobar 1988a
EN
Passiflora purdiei Killip, 1938 *
and
cun ma
Purdie s.n. (K)
n.v.
Escobar 1988a
EX
Subgenus Tryphostemmatoides (Harms) Killip), 1938
Passiflora tryphostemmatoides Harms, 1894
and
ant cau hu qu
ri vc
1000-2700
Lehmann 5662
(K) - Isotype
NT
Tr
Passiflora gracillima Killip, 1924
and
ant cau cl hu
na qu to
2000-3150
Penell 9393
(MO) - Isotype
Killip 1938;
Holm-Nielsen et
al. 1988
Killip 1924,
1938
LC
Passiflora arbelaezii L. Uribe, 1957
and pac
ant cau cho
cun na vc
0-2300
Roldán 1162
(COL)
Uribe 1957
LC/NT
Tr
V.N.:
Golondrina
(cho)
Tr
Passiflora pacifica L.K. Escobar, 1988 *
pac
cho na vc
0-1800
Escobar 2143
(HUA)
Escobar 1988b
LC/NT
Tr
63
Tr
V.N.: Curuba
de Monte
(vc), Curubo
de Páramo
(cun),
Palchuaca
(cau),
Curubito de
Indio (cl).
Edible fruit
Tr
V.N.: Curuba.
Edible fruit
Tr
V.N.:
Palchoaca (ns,
snt)
Tr
Known only
from the type.
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
II.1.6. Conclusions
With 167 reported species, Colombia is the country with the highest Passifloraceae
richness. This richness is concentrated in the Andean region, particularly in the
departments of Antioquia, Valle del Cauca and Cundinamarca. Comparing with other
countries indicates that the northern Andes of Colombia and Ecuador constitute the center
of diversity for the most important genus, Passiflora. The low level of exploration in
parts of the Andes, the Amazonian and the Orinoquian raises expectations that Colombia
may still harbor many unknown species. Future studies should encompass new regions,
including protected areas and conflict zones. Indeed, a better knowledge of this diversity,
and its distribution, is urgent for the in situ conservation of this threatened richness,
targeting the conservation of these resources as well as their habitat. Both aspects may
even be combined if the genus Passiflora can be used as an indicator of biodiversity in
the Andean region, as was the objective of a project in the coffee growing zone. Another
important aspect is its direct valorization as a germplasm resource for crop diversification
programs, implying the need for a better understanding of its morphological and genetic
diversity.
II.1.7 Acknowledgements
The authors wish to thank the herbaria that provided specimens or collection data, and
particularly Francisco J. Roldán (HUA) and Alexandra Hernández (COL), as well as
Colciencias, the Colombian Ministry for Environment (MMA) and the Research Center
of the Colombian Coffee Grower Federation (Cenicafé) for funding the collecting
missions. The first author gratefully acknowledges financial support from the Gines-Mera
Fellowship Foundation (CIAT-CBN). We are indebted also to José O. Velásquez (Casa
Mutis, Mariquita), Hernando Criollo (U.Nariño), Mauricio Villegas (Cenicafé), Vicky
Barney (Bioversity International), Alvinxon Castro (U.Chocó), Robinson Galindo (PNN
Catatumbo) and Carolina Alcazar (Proselva) for assistance in obtaining plant data for this
study. We are extremely grateful to Colombian farmers contacted in the fieldwork for
64
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
their continuous help and availability in localizing a great part of the observed plant
material.
II.1.8. Appendix 1. Synonymy = valid name
Cieca auriculata M. Roemer, 1846 = Passiflora auriculata Kunth, 1817
Cieca coriacea (Juss) M. Roemer, 1846 = Pasiflora coriacea Juss., 1805
Cieca discolor M. Roemer, 1846 = Passiflora misera Kunth., 1817
Cieca pallida (L.) M. Roemer, 1846 = Passiflora pallida L., 1753
Decaloba alnifolia M. Roemer, 1846 = Pasiflora alnifolia Kunth, 1817
Decaloba biflora (Lam.) M. Roemer., 1846 = Pasiflora biflora Lam., 1789
Decaloba bogotensis (Benth.) M. Roemer, 1846 = Passiflora bogotensis Benth., 1845
Decaloba cuneata M. Roemer., 1846 = Passiflora cuneata Willd., 1809
Decaloba filipes M. Roemer, 1846 = Pasiflora filipes Benth, 1843
Decaloba holosericea M. Roemer, 1846 = Passiflora holosericea L., 1753
Decaloba jacquini M. Roemer, 1846 = Passiflora pulchella Kunth, 1817
Dilkea johannesii var. parvifolia Hoehne, 1915 = Dilkea johannesii Barb.Rodr., 1885
Dilkea acuminata Mast., 1871 = Dilkea retusa Mast., 1871
Dilkea magnifica Steyerm., 1968 = Dilkea retusa Mast., 1871
Dilkea wallisii Mast., 1872 = Dilkea retusa Mast., 1871
Disemma hahnii E. Fourn., 1869 = Passiflora hahnii Mast., 1872
Disemma hahnii Fourn., 1869 = Passiflora hahnii (Fourn.) Mast., 1872
Distephana cuneata M. Roemer, 1846 = Passiflora bicuspidata (H.Karst.) Mast., 1872
Distephana spinosa (Poeep. & Endl.) M. Roemer, 1835 = Passiflora spinosa (Poepp. & Ende.) Mast., 1871
Granadilla rubra Moench, 1802 = Passiflora rubra L., 1753
Grandilla vespertilio Moench, 1802 = Passiflora vespertilio L., 1753
Passiflora erubescens Triana & Planch., 1873 = Passiflora erytrophylla Mast., 1872
Passiflora velata Mast., 1872 = Passiflora serrulata Jacq., 1767
Passiflora williamsii Killip, 1922 = Passiflora platyloba var. williamsii (Killip) A.H. Gentry, 1976
Passiflora adenophylla Mast., 1872 = Passiflora subpeltata Ortega, 1798
Passiflora alba Link & Otto, 1798 = Passiflora subpeltata Ortega, 1798
Passiflora albicans L. Uribe, 1958 = Passiflora uribei L. K. Escobar, 1988
Passiflora angustifolia Swartz, 1788 = Passiflora suberosa L., 1753
Passiflora appendiculata G.F.W. Mey., 1818 = Pasiflora auriculata Kunth, 1817
Passiflora bauhinifolia Kunth, 1817 = Passiflora alnifolia Kunth, 1817
Passiflora bifurca Mast., 1873 = Passiflora cuneata Willd., 1809
Passiflora bilobata Vell., 1827 = Passiflora rubra L., 1735
Passiflora boyacana Killip, 1960 = Passiflora crispolanata L. Uribe, 1954
Passiflora capsularis var. geminifolia DC., 1828 = Passiflora sexflora Juss., 1805
Passiflora caucaense Holm-Niels., 1974 = Passiflora emarginata Humb. & Bonpl., 1813
Passiflora chilensis Miers, 1826 = Passiflora pinnatistipula Cav., 1799
Passiflora cisnana Harms, 1894 = Passiflora rubra L., 1753
Passiflora corumbaensis Barb., 1898 = Passiflora cincinnata Mast., 1868
Passiflora cualiflora Harms, 1906 = Passiflora citrifolia (Juss.) Mast., 1871
Passiflora difformis Kunth, 1817 = Passiflora coriaceae Juss., 1805
Passiflora digitata L., 1763 = Passiflora serratodigitata L., 1753
Passiflora elegans Triana & Planch., 1873 = Passiflora quindiensis Killip, 1938
Passiflora emiliae Sacco, 1966 = Passiflora ambigua Hemsl. ex Hook., 1902
Passiflora eminula Mast., 1883 = Passiflora costata Mast., 1872
Passiflora eriocaula Harms, 1922 = Passiflora rugosa (Mast.) Triana & Planch.var. rugosa,1873
Passiflora erosa Rusby, 1907 = Passiflora morifolia Mast., 1872
Passiflora erosa Rusby, 1906 = Passiflora morifolia Mast., 1872
Passiflora fulgens Wallis ex Morren, 1866 = Passiflora coccinea Aubl., 1775
Passiflora gigantifolia Harms, 1894 = Passiflora macrophylla Spruce ex Mast., 1883
Passiflora glauca Humb. & Bonpl., 1813 = Passiflora arborea Spreng., 1826
65
Chapter II. Biogeography and an updated list for conservation
________________________________________________________________________
Passiflora goudotiana Triana & Planch., 1873 = Passiflora cumbalensis (H. Karst.) Harms var. goudotiana (Triana &
Planch.) L. K. Escobar, 1987
Passiflora heydei Killip, 1922 = Passiflora morifolia Mast., 1872
Passiflora hydrophila Barb Rodr., 1891 = Passiflora costata Mast., 1872
Passiflora incana Seemann ex Mast., 1883 = Passiflora seemanni Griseb., 1858
Passiflora inundata Ducke, 1925 = Passiflora costata Mast., 1872
Passiflora laticualis Killip, 1924 = Passiflora misera Kunth., 1817
Passiflora longipes var. retusa Triana & Planch., 1873 = Passiflora longipes Juss., 1805
Passiflora lorifera Mast, & André, 1883 = Passiflora macrophylla Spruce ex Mast., 1883
Passiflora lunata J.E. Smith., 1790 = Passiflora biflora Lam., 1879
Passiflora macrocaropa Mast., 1869 = Passiflora quadrangularis l., 1759
Passiflora micrantha Killip, 1938 = Passiflora erythrophylla Mast., 1872
Passiflora miraflorensis Killip, 1924 = Passiflora sexflora Juss., 1805
Passiflora mollis var. integrifolia Planch. ex Mast., 1872 = Passiflora cuspidifolia Harms, 1893
Passiflora nympheoides Karst., 1859 = Passiflora nitida Kunth, 1817
Passiflora oblongifolia Pulle, 1906 = Passiflora laurifolia L., 1753
Passiflora ocanensis Planch. & Linden, 1873 = Passiflora lindeniana Planch. ex Triana & Planch., 1873
Passiflora ornata Kunth, 1817 = Passiflora maliformis L., 1753
Passiflora pala Planch. & Linden, 1873 = Passiflora bogotensis Benth., 1845
Passiflora paraguayensis Chad., 1899 = Passiflora capsularis L., 1753
Passiflora pennipes Sm., 1819 = Passiflora pinnatistipula Cav., 1799
Passiflora praeacuta Mast., 1887 = Passiflora oerstedii Mast., 1872
Passiflora pubera Planch. & Linden, 1873 = Passiflora sphaerocarpa Triana & Planch., 1873
Passiflora pulchella Kunth,1817 = Passiflora bicornis Mill., 1768
Passiflora quadriglandulosa var. involucrata (Mast.) Killip, 1938 = Passiflora involucrata (Mast.) A.H. Gentry, 1981
Passiflora reticulata Sauv., 1873 = Passiflora holosericea L.,1753
Passiflora salmonea Harms, 1894 = Passiflora parritae (Mast.) Bailey, 1916
Passiflora sanguinea J.E. Smithi, 1819 = Passiflora vitifolia Kunth, 1817
Passiflora schultzei Harms, 1929 = Passiflora arborea Spreng., 1826
Passiflora spherocarpa var. pilosula Mast., 1883 = Passiflora pubera Planch. & Linden, 1873
Passiflora stipulata Aubl., 1858 = Passiflora subpeltata Ortega, 1798
Passiflora suberosa var. pallida (L.) Mast. = Passiflora pallida L., 1753
Passiflora tomentosa Lam. var. mollissima Triana & Planch., 1873 = Passiflora mollissima (Kunth) L.H. Bailey, 1916
Passiflora trisecta Planch. & Linden ex Triana & Planch., 1873 = Passiflora trianae Killip, 1938
Passiflora Van-Volxemii Triana & Planch., 1893 = Passiflora antioquiensis Karst., 1859
Passiflora var. cuellensis Goudot ex Triana & Planch., 1873 = Passiflora menispermifolia Kunth, 1817
Passiflora vesicaria L., 1753 = Passiflora foetida L., 1753
Passiflora vitifolia var. involucrata Mast., 1872 = Passiflora involucrata (Mast.) A.H. Gentry, 1981
Passiflora weberiana André, 1885 = Passiflora morifolia Mast., 1872
Passiiflora acerifolia Schlecht. & Cham., 1830 = Passiflora adenopoda Moc. & Sessé ex DC., 1828
Rathea floribunda Karst., 1859 = Passiflora andina Killip, 1938
Tacsonia adulterina Juss., 1805 = Passiflora adulterina L. f., 1781
Tacsonia bicuspidata H. Karst., 1859 = Passiflora bicuspidata (H. Karst.) Mast., 1872
Tacsonia cumbalensis H. Karst., 1859 = Passiflora cumbalensis var. cumbalensis (H. Karst.) Harms, 1894
Tacsonia cuneata Benth, 1845 = Passiflora bicuspidata (H. Karst.) Mast., 1872
Tacsonia flexipes (Triana & Planch) Mast., 1883 = Passiflora flexipes Triana & Planch., 1873
Tacsonia glandulosa Juss., 1805 = Passiflora glandulosa Cav.,1790
Tacsonia infundibularis Mast., 1883 = Passiflora bracteosa Planch. & Linden, 1873
Tacsonia lanata Juss., 1811= Passiflora lanata (Juss.) Poir., 1811
Tacsonia mixta (L.f.) Juss., 1805 = Passiflora mixta L.f., 1781
Tacsonia mollissima Kunth var. glabrescens Mast.,1872 = Passiflora mollissima (Kunth) L.H. Bailey, 1916
Tacsonia mollissima Kunth, 1817 = Passiflora mollissima (Kunth) L.H. Bailey, 1916
Tacsonia parritae Mast., 1882 = Passiflora parritae (Mast.) L.H. Bailey, 1916
Tacsonia pinnatistipula var. pennipes (Sm.) DC., 1828 = Passiflora pinnatistipula Cav., 1799
Tacsonia pinnatistipula (Cav.) Juss., 1805 = Passiflora pinnatistipula Cav.,1799
Tacsonia quadriglandulosa (Rodschied) DC., 1828 = Passiflora quadriglandulosa Rodschied, 1796
Tacsonia rosea (H. Karst.) Sodiro, 1903 = Passiflora x rosea (H. Karst.) Killip, 1938
Tacsonia rugosa Mast., 1872 = Passiflora rugosa (Mast.) Triana & Planch, 1873 var. rugosa
Tacsonia spinescens Klotsch in Schomb., 1848 = Passiflora securiclata Mast., 1893
Tacsonia spinosa Poepp. & Endl., 1835 = Passiflora spinosa (Poepp. & Ende.) Mast., 1871
Tacsonia trinervia Juss., 1805 = Passiflora trinervia (Juss.) Poir., 1811
Tetrastylis lobata Killip, 1926 = Passiflora lobata (Killip) Hutch. ex J.M. MacDougal, 1986
66
CHAPTER III
____________________________________
Distribution, diversity and in situ conservation of
Colombian Passifloraceae
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
III.1. Distribution, diversity and in situ conservation of Colombian
Passifloraceae
John Ocampo Pérez1, Geo Coppens d'Eeckenbrugge2, Mike Salazar1 and Andy Jarvis1, 3.
1
Bioversity International (formerly IPGRI), Regional Office for the Americas, A.A. 6713,
Cali, Colombia.
2
CIRAD/FLHOR, UPR ‘Gestion des ressources génétiques et dynamiques sociales’,
Campus CNRS/Cefe, 1919 route de Mende, 34293 Montpellier, France.
3
International Center for Tropical Agriculture (CIAT), A.A. 6713, Cali, Colombia.
III.1.1 Abstract
Analysis was made of 3,930 records of 165 wild Passifloraceae to assess the distribution of
their diversity in Colombia, identify collection gaps, and explore their potential as indicator
species. Despite variable collecting density among and within biogeographic regions, the
Andean region clearly presents a higher species richness, particularly in the central coffee
growing zone and the departments of Antioquia, Cundinamarca and Valle del Cauca. The
elevational distribution of diversity shows a small peak below 500 m, and two higher ones
between 1000-2,000 and 2,500-3,000 m. This pattern corresponds to divergent adaptive
trends among genera and infrageneric divisions. The analysis on 19 climatic variables
showed that the two principal variance components, explaining 77% of the total, are
respectively associated with temperature and precipitation, without influence of seasonality.
Distribution parameters allow recognizing more than 36 narrow endemics. Prediction of
species distribution showed nine areas with very high richness (predicted sympatry of 41 to
54 species) in the Andean region, three of which correspond to collection gaps. Endemics
were not particularly frequent there, so a prioritization of protected areas based on species
richness would not favor their conservation. The sites with high Passiflora diversity are
poorly represented in the current system of protected areas. Instead their striking
correspondence with ecotopes of the coffee growing zone imposes a conservation strategy
integrating agricultural and environmental management at the landscape level.
Reciprocally, several traits of Passiflora species make them particularly suited as indicators
for any effort of conservation or restoration in this region of importance for the country.
Keywords: Andes, Colombia, coffee growing zone, distribution, geographic information
systems, Passifloraceae diversity, endemism.
68
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
III.1.2. Introduction
New World Passifloraceae include four tropical genera (Ancistrothyrsus, Dilkea,
Mitostemma and Passiflora) and about 550 species (Ulmer & MacDougal, 2004). The
largest genus of the family is Passiflora L., with ca. 525 species distributed in a wide range
of habitats, essentially tropical, ranging from humid rain forests to semi-arid subtropics.
Most of them are herbaceous or woody vines, usually climbing by tendrils, while a few are
trees or shrubs. More than 80 species produce an edible fruit, the most interesting ones
belonging to subgenera Passiflora and Tacsonia (Coppens d’Eeckenbrugge, 2003). Among
them, is the maracuja, P. edulis Sims, and its yellow-fruited form, P. edulis f. flavicarpa
Degener, with a world production estimated at more than 640.000 tons (Passionfruit, 2006).
In Colombia, ten more species/forms are cultivated. The fruits of the purple maracuja
(P. edulis f. edulis Sims.), the sweet granadilla (P. ligularis Juss.), the giant granadilla
(P. quadrangularis L.), the stone granadilla (P. maliformis L.), the granadilla de Quijos
(P. popenovii Killip), the sweet maracuja (P. alata Curtis), and the banana passion fruits,
‘curuba de Castilla’ (P. tripartita var. mollissima (Kunth) Holm-Nielsen & Jørgensen),
‘curuba quiteña’ (P. tarminiana Coppens & Barney) are common on the national or
regional markets. Those of the ‘granadilla de clima frío’ (P. pinnatistipula Cav.), the
‘curuba roja’ or rosy passion fruit (P. cumbalensis Karst.), the ‘curuba antioqueña’
(P. antioquiensis (Karst) Harms.), P. nitida Kunth and P. palenquensis Holm-Niels. &
Lawesson are also cultivated, at a very local scale. These and other Passiflora species may
be used for other purposes, such as garden ornamentals or pharmaceuticals (Coppens
d’Eeckenbrugge et al., 2001). Killip in 1938 classified the genus Passiflora in 22
subgenera, including 355 American species (Annex 1). This classification has been
amended by Escobar (1988, 1989, 1994) who merged subgenera Tacsioniopsis and
Tacsonia, and proposed a new subgenus, Porphyropathanthus, and MacDougal (1994),
who reinstated the name Decaloba for Killip’s subgenus Plectostemma. More recently,
Feuillet & MacDougal (2003; Annex 2) proposed a much deeper revision, reducing the
number of subgenera to only four (Astrophea, Decaloba, Deidamioides and Passiflora),
and downgrading most of Killip’s divisions to lower levels. This proposal has been partially
justified by recent molecular data (Muschner et al., 2003; Yockteng, 2003, Annexe Ib;
69
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Yockteng & Nadot, 2004; Hansen et al., 2006, see Annex 4a-d), however it still needs to be
more clearly supported by further studies.
Colombia’s geographic location and its variety of ecosystems place it as the second most
biodiverse country in the world (McNeely et al., 1990). It is divided into five main
biogeographic regions: Amazonian, Andean, Caribbean, Orinoquian, and Pacific
(Hernández et al., 1991). The Andean region presents a highly diverse topography (1005,400 m), with three long mountain ranges, the Eastern, Central and Western Cordilleras,
which separate two main inter-Andean valleys from the other regions. The uplift of the
Andes created new habitats and increased local isolation, favoring high speciation rates in
many taxa (Gentry, 1986). The continuously humid climate of the Amazonian and
Orinoquian lowlands and the extremely wet climate of the Pacific region contrast with the
dryer and more seasonal climate of the Caribbean. As a result, the Colombian flora includes
some of the world most diverse groups of vascular plants, with 51,220 documented species
(May, 1992; UNEP-WCMC, 2004). It is hoped that most of the floristic richness is located
in the protected areas that cover 365,120 km2, approximately 32% of the territory (Parques
Naturales de Colombia, 2006), falling under different categories of protection, including
Natural National Parks, Flora and Fauna Sanctuary, Natural National Reserves, Unique
Natural Area, Park Way and Indigenous Areas. Smaller forest reserves have also been
created to protect river basins for water supply. On the other hand, destruction of many
natural habitats has drastically affected species distributions, often reducing their historical
ranges to a set of small, fragmented populations (Brooks et al., 2002). Such habitat
alteration is predicted to lead to substantial extinction in the near future.
Many conservation biologists have focused their attention on areas presenting high levels of
endemism and diversity, and that are also experiencing a high rate of loss of ecosystems.
Such regions, characterized by localized concentrations of biodiversity under threat, and
representing priorities for conservation actions, are defined as biodiversity hotspots (Myers
et al., 2000; Sechrest et al., 2002). The application of this concept in the case of Colombia
implies very wide studies to investigate the distribution of biodiversity and endemism
across the country. Complete inventories are not realistic at that scale, so other approaches
70
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
have been taken to exploit incomplete biodiversity data, combining remote sensing and
field sampling/inventories of indicator taxa at different scales (Villareal et al., 2006). We
proposed here to use climatic niche modeling and test the potential of Passifloraceae as an
indicator of biodiversity in Colombia, as this family represents several interesting traits in
terms of diversity, adaptation and evolution.
Indeed, Colombia is particularly rich in Passifloraceae. The family is represented by 167
species (about 27% of the total), grouped in three genera, Ancistrothyrsus, Dilkea, and
Passiflora, with greatest diversity in the Andean region (123 species). The country has 58
endemic Passifloraceae species, 95% of them exclusively Andean, implying a high
extinction risk, as the Andes region is the most densely populated and hence disturbed of
the country (Ocampo et al., 2007). According to the categories and criteria of IUCN Red
List of threatened species (IUCN, 2003) and Ocampo et al. (2007) have established that
more than 100 Colombian species of Passifloraceae are threatened to some degree, while
three species are considered extinct.
As vines, most Passiflora species have adapted to many different habitats, particularly for
their support. They are medium-lived organisms depending on longer-lived trees and
shrubs, which makes them responsive to both medium and long-term changes of their
ecosystems. They also show high levels of co-evolution with their herbivores, the bestknown example being that of Heliconius butterflies (Gilbert, 1982), and some species even
exhibit elements of the carnivory syndrome (Radhamani et al., 1995). They have developed
mutualism with protector insects as nectar-feeding ants (Apple & Feener, 2001), and with a
wide range of pollinators, including small and large insects, birds, and even bats (Büchert
& Mogens, 2001; Sazima & Sazima, 1978).
Finally, given the economic importance of several of its representatives, the genus
Passiflora constitutes an important genetic resource, and the characterization and
evaluation of wild and cultivated populations is seen as a high priority for Andean countries
because of its clear potential for development and crop diversification (Debouck &
71
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Libreros, 1995). Strategies for conservation and improvement are needed to optimize the
use and conservation of this resource.
Biodiversity data have been traditionally produced through a variety of complementary
approaches using field survey and sampling, museum records, botanical collections, and, in
recent times, spatial analysis of digital data integrated within a Geographical Information
System (GIS). In each area, the combination of geological, edaphic, climatic, ecological,
historical and anthropic factors produce a unique range of constraints defining unique
patterns of genetic diversity (Maxted et al., 1995). GIS makes it possible to build maps of
species richness, potential distribution and endemism, to prioritize areas for conservation
based on principles such as complementarity, and to assess the completeness of existing
protected areas networks (Peterson, 2001). Applications have been developed in recent
years that offer new possibilities for understanding biological diversity. Several of these
methods use climatic variables as the principal drivers of herbarium or collecting data,
often in combination with spatial environmental information, and are generally
acknowledged to generate valuable additional information for diversity studies and
conservation actions (Franklin, 1995; Skov, 2000; Lehmann et al., 2002). Such modeling
tools have been applied to problems of plant biogeography (Hijmans et al., 2001; Hijmans
and Spooner, 2001; Midgley et al., 2002; Vargas et al., 2004; Ferguson et al., 2005),
conservation (Jones et al., 1997; Hijmans et al., 2000; Jarvis et al., 2002, 2003, 2005;
Kingston & Waldren, 2005), evolutionary ecology (Zaharieva et al., 2004), invasive or
endemic species management (Peterson & Robins, 2003; Peterson, 2004; Leimberck et al.,
2004), and potential areas for plant collection (Rodríguez et al., 2005). In Passiflora,
Segura et al. (2003) reported a first mapping of the potential distribution of five species of
subgenus Tacsonia in the Andean countries using the Floramap software of Jones et al.
(2002). Their results showed evidence of intra-specific variation in climatic adaptation
along the Andes, from Colombia to Peru. Recently, Scheldeman et al. (2006) used the
DIVA-GIS computer program to determine the distribution and environmental adaptation
of highland papayas (Vasconcellea spp.) based on germplasm and herbarium data. The
combination of observed diversity maps and potential diversity maps permitted identifying
important collection gaps, mainly in Colombia and Ecuador.
72
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
The present study of the distribution and diversity of Passifloraceae in Colombia was
conducted through sequentially: (1) assessing the geographic distribution of collections, (2)
analyzing it in terms of species richness across the Colombian territory, (3) inferring the
potential distribution of each species with predictive distribution models, (4) summing
these spatial predictions to produce a map of potential diversity, and (5) locating collecting
gaps by detecting those areas where Passifloraceae species are likely to occur but have no
yet been collected. Combining these results permits an analysis of the current status of in
situ and ex situ conservation of Passifloraceae in Colombia. It also provides elements to
evaluate the potential of this group as an indicator for the detection of biodiversity hotspots
and monitoring of conservation/restoration efforts.
III.1.3. Materials and methods
III.1.3.1. Geography and climate
Colombia is located in the north of South America, between 12° 26’ 46” N and 4° 13’ 30”
S and between 66° 50’ 54” W and 79° 02’ 33” W, covering an area of 1,141,748 km2, with
altitudes ranging from the sea level to 5,775 m (IGAC, 2006). It is divided in 32
departments. Its natural habitat diversity is distributed among five main biogeographic
regions (Amazonian, Andean, Caribbean, Orinoquian and Pacific). Figure 1 shows the
distribution of the 32 departments and the five biogeographic regions of the country.
Colombian climates are tropical, with relatively uniform temperatures throughout the year.
Precipitations vary greatly, with some of the wettest parts of the world in the Pacific
lowlands (average annual rainfall reaching 10,000 mm) contrasting with extremely dry
areas in the coast (< 500 mm per year), and show a tendency to increase with altitude in the
Andean region.
73
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Figure 1. Colombia’s geopolitical division in 32 departments and biogeographic division in five regions.
Amazonas (ama), Antioquia (ant), Arauca (ara), Atlántico (at), Bolívar (bl), Boyacá (by), Cauca (cau),
Cesar (ce), Caldas (cl), Córdoba (cor), Caquetá (cq), Casanare (cs), Cundinamarca (cun), Chocó (cho),
Guainia (gn), Guaviare (gv), Huila (hu), La Guajira (lg), Magdalena (ma), Meta (met), Nariño (na), Norte de
Santander (ns), Putumayo (pu), Quindío (qu), Risaralda (ri), Santander (snt), San Andrés y Providencia (sp Island), Sucre (suc), Tolima (to), Vaupés (va), Valle del Cauca (vc), Vichada (vch).
74
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
III.1.3.2. Herbarium, field and literature data
The dataset used in this analysis consists of the information gathered and georeferenced by
Ocampo et al. (2007), from 3,930 individuals belonging to 167 Passifloraceae species,
consisting of 3,330 specimens from 23 herbaria (AFP, CAUP, CDMB, CHOCO, COL,
COAH, CUVC, FAUC, FMB, HUA, HUQ, JAUM, K, MA, MEDEL, MO, NY, P, PSO,
SURCO, TOLI, VALLE, UIS), 555 records from explorations in localities of 17
departments, and 45 records of specimens mentioned by Killip (1938, 1960), Uribe
(1955a,b) and Escobar (1988a,b, 1989, 1994).
III.1.3.3. Species distribution and richness
The number of observations and their elevation range were tabulated for each species. The
distribution of species was plotted on dot-maps using the DIVA-GIS software. To quantify
the area over which each species is distributed, the maximum distance (MaxD) and circular
area (CAr) were calculated, following the method of Hijmans et al. (2001), and compared.
MaxD is the longest distance between any pair of observations of one species. CAr is
calculated by assigning a circle of radius r to each observation and calculating the area
covered by all circles for each species. In this case, a radius of 50 km was used (CA50).
Species richness was calculated as the number of species within a defined area,
superimposing species location maps, using the point-to-grid richness analysis tool in
DIVA-GIS, with a 0.1 x 0.1° grid (corresponding to 12 x 12 km at the equator). The
circular neighborhood option was applied with a 2° radius (Hijmans & Spooner, 2001) to
eliminate border effects due to the assignation of the grid origin. Species richness was used
as a measure of taxonomic diversity because it is a simple, useful, and widely used
parameter (Gaston, 1996).
III.1.3.4. Description of climatic preferences
In order to highlight the factors that may influence species geographic distribution, and
provide an indication of their tolerance to abiotic stress, climate data were extracted for
each collection point. DIVA-GIS was used to develop climatic models for predicting the
occurrence/diversity of Passifloraceae species in the study area. This package uses
75
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
WorldClim climate data (Hijmans et al., 2005), which consists of global surfaces of climate
with a 30” grid resolution (corresponding to 1 x 1 km at the equator) derived from a
network of over 12,500 meteorological stations across Latin America, 1,479 of which are
located in Colombia. For each collection site, 19 bioclimatic variables (derived from 12
monthly means for temperature, rainfall and diurnal temperature range according to Busby
(1991) were extracted from the WorldClim data set. Principal components analysis (PCA)
was performed on climatic variables from the collection sites where at least one specimen
was present. A varimax normalized rotation was applied with the STATISTICA® 6.0
software. To promote readability, the centroid, i.e. the arithmetic average of the factor
scores, was used to represent each species general climatic preferences.
III.1.3.5. Potential species distribution
Maps of potential species distribution were produced using the BioClim method inside
DIVA-GIS for those species with more than 10 observations. Predictive distribution
modeling (also termed ecological niche modeling) attempts to predict the geographic
distribution of a species using sites of known existence to understand the environmental
adaptation of the species, and extrapolating this knowledge to other regions with similar
conditions, where no collection has been made. BioClim was chosen because it is a robust
methodology, requiring presence-only data (Hijmans & Graham, 2006). It was run using
the 19 bioclimatic variables from WorldClim (Hijmans et al., 2005) with 30” resolution
(equivalent to 1 km at the equator). Eighty-six species with fewer than 10 observations
were omitted from the analysis, as the number of points was too low for reliable results.
Unfortunately, this excluded many of the endemic and rare species. Whilst soils and habitat
are likely to be also important in determining the geographic distribution of these species,
they were not included in the model due to a lack of consistent, continuous data at the
country level. Finally, an analysis of complementarity (Rebelo, 1994) was applied to
identify the lowest number of protected areas needed for the conservation of native
Passifloraceae.
76
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
III.1.4. Results and discussion
III.1.4.1. Distribution of observations
Figure 2 and Table 1 show the distribution of collection/observation points for all
specimens. The Andean region of Central Colombia is by far the most densely explored,
particularly in the central coffee growing zone (Quindío, Caldas and Risaralda; 18.93 to
77.20 observations/1000 km²) and the three large departments of Antioquia, Valle del
Cauca and Cundinamarca (12.49 to 19.82 observations/1000 km²). By comparison, the
northeastern Andes (Boyacá, Santander, Norte de Santander) and the central department of
Tolima appear less well explored (3.59 to 9.39 observation/1000 km²). The situation is
more difficult to appreciate in the southern Andes, as the departments of Cauca and Nariño
also belong in good part to the Pacific region. However, they show a collection density only
slightly superior to that of Chocó, which indicates that they have also been less explored
than the central departments of the Andean region. The situation is heterogeneous in the
Caribbean region, with only two of its seven departments exhibiting more than
3 observations/1000 km² (excluding the atypical case of the small San Andrés and
Providencia islands). Finally, the Amazonian and the Orinoquian are by far the less
explored biogeographic regions of the country, although they cover half of its area.
The mean number of observations per species also reflects variation in exploration among
departments (Table 1), confirming the much denser exploration in the Andes of Antioquia,
Cundinamarca and Valle del Cauca (more than 7 observations per species) and in the
Pacific region, while this ratio takes much lower values in the other biogeographic regions.
However, the relation between exploration density and this indicator is not simple, as the
numerous observations in the central coffee growing zone (Caldas, Quindío and Risaralda)
are distributed among a very wide diversity of species, so the mean number of observations
per species is not as high as could be expected for such densely explored areas.
77
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Figure 2. Collection localities of Passifloraceae specimens used in this study, among Colombian departments.
Points on the maps represent sites of collection.
78
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
This variation in exploration of the Colombian territory can be related in good part to
human geography. The superposition of roads and rivers on the map (not presented here)
shows that collections were mostly made along rivers in the Orinoquian and Amazonian
regions, and roads in other regions, which is a very common and self-explained observation
(Hijmans et al. 2000). This is likely why regions with poor infrastructure (e.g. Amazonian,
Orinoquian and Pacific) have been little explored. An aggravating factor is collector safety.
Most of these areas also suffer from chronic political conflict. However other areas are
affected although better equipped in infrastructure.
Table 1. Number of observations, species, rare and endemic Passifloraceae species by Colombian division.
Source for department areas: IGAC (2005, http://www.igac.gov.co).
Area
(km2)
Nb.
observ.
Nb.
observ.
/1000km2
Total
species
Total
species
/1000km2
Total
species/Log.
area
Observ.
/ species
Rare
species
Endemic
species
Andean
Antioquia
Boyacá
Caldas
Cundinamarca
Huila
Quindío
Norte de Santander
Risaralda
Santander
Tolima
62.869
23.012
7.291
23.942
18.331
1.943
22.007
3.592
30.537
22.672
785
145
245
419
62
150
79
68
207
213
12.49
6.30
33.60
17.50
3.38
77.20
3.59
18.93
6.78
9.39
70
36
36
53
22
38
36
24
48
44
1.11
1.56
4.94
2.21
1.20
19.56
1.64
6.68
1.57
1.94
14.588
7.502
7.502
11.045
4.585
7.919
7.502
5.002
10.003
9.169
11.21
4.03
6.81
7.91
2.82
3.95
2.19
2.83
4.31
4.94
28
14
14
23
18
25
25
20
31
27
6
1
1
0
0
0
0
0
3
4
Andean and Pacific
Cauca
Nariño
Valle del Cauca
30.985
32.046
21.195
161
170
420
5.20
5.30
19.82
42
44
56
1.36
1.40
2.69
8.753
9.170
11.670
3.83
3.79
7.38
24
27
28
1
0
1
Pacific
Chocó
46.530
211
4.53
40
0.86
8.336
5.28
23
1
Caribbbean
Atlántico
Bolívar
Cesar
Córdoba
La Guajira
Magdalena
S. Andrés y Providencia
Sucre
3.319
26.469
22.213
25.020
20.848
22.742
53
10.917
18
33
13
33
21
84
4
6
5.42
1.25
0.59
1.32
1.01
3.69
75.47
0.55
7
15
10
9
12
31
2
3
2.11
0.57
0.45
0.36
0.58
1.36
37.74
0.27
1.459
3.126
2.084
1.876
2.501
6.460
0.417
0.625
2.57
2.20
1.30
3.67
1.75
2.71
2.00
2.00
5
9
9
6
9
19
2
2
0
1
0
0
0
1
0
0
Orinoquian
Arauca
Casanare
Meta
Vichada
23.393
44.428
85.286
100.242
10
4
85
16
0.43
0.09
1.00
0.16
6
4
24
9
0.26
0.09
0.28
0.09
1.250
0.834
4.930
1.876
1.67
1.00
3.56
1.78
3
4
14
6
0
0
0
0
Amazonian
Amazonas
Caquetá
Guainía
Guaviare
Putumayo
Vaupés
109.665
91.725
70.691
55.391
24.885
54.135
87
47
16
27
56
35
0.79
0.51
0.23
0.49
2.25
0.65
19
18
10
14
26
20
0.17
0.20
0.14
0.25
1.04
0.37
3.751
3.751
2.084
5.418
2.918
4.168
4.05
2.61
1.60
1.93
2.15
1.75
16
13
9
11
20
10
0
0
0
0
0
0
Predominant region /
department
79
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
This is particularly obvious in the less explored Andean departments (Tolima, Santander,
Norte de Santander and part of Boyacá) and in most departments of the Caribbean.
Conversely, most densely populated areas, particularly those around main cities and their
universities (Bogotá, Medellín, Cali, central coffee growing zone), have been much more
densely explored. This high heterogeneity in exploration must be taken into account to
avoid a sampling bias when comparing the situation among departments (Table 1).
However, even doing so, Figure 2 and Table 1 show a clear concentration of collection
density in the Andean departments. Furthermore, a closer look at the heterogeneity of
collection density within these departments, as shown in Figure 1, confirms the association
between topography and the density of botanical observations. The most likely explanation
is simply that the higher Passiflora diversity in the Andes made their exploration more
rewarding and stimulating for its students. The distribution of species richness clearly
confirms this view.
III.1.4.2. Species richness
Departments of the Andean region present a higher species richness (Table 1). The only
non-Andean department showing a comparable value for this parameter is Chocó. Among
the Andean departments, Antioquia has by far the highest number of species (70 species;
43% of the total), followed by Valle del Cauca and Cundinamarca. These three departments
also show the three highest mean numbers of observations per species, which confirms that
they are much better explored than the others. Concerning rare species (≤ 5 observations, ≤
100 km of MaxD and ≤ 20,000 km2 of CA50), Santander (northeast) occupies the first place,
with 31 species, followed by Valle del Cauca (28), and Antioquia (28), Nariño (27), and
Tolima (27). Thus, there is little doubt that a more thorough exploration north of the
Eastern Cordillera (Santander) and south of the Central Cordillera (Tolima) would discover
more specimens per observed species and/or more species. This problem is similar in the
departments of the Amazonian, Orinoquian and Pacific regions, where species richness is
very poor in relation to their surface. The number of rare species is abnormally high as
compared to the total number, indicating a sampling bias due to very poor exploration in
these regions. In the case of Chocó (Pacific region) collecting efforts have been practically
limited to the borders of the main access road.
80
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
III.1.4.3. Species diversity
When species richness is related to department size, the most diverse area clearly
corresponds to the central coffee growing zone, as this ratio appears to be several times
higher in Caldas, Risaralda and Quindío than in the other Andean departments. A precise
comparison with departments of other regions is only possible if the species are equally
sampled, i.e. if the number of observations per species is equivalent. This is the case for
Chocó, Amazonas, and Córdoba, all of them showing a much lower diversity. The map of
Passifloraceae diversity, as produced by the GIS analysis (Figure 3), confirms the
importance of the Andes and the special contribution of the central coffee growing zone.
This correspondence is so striking that one can wonder if, in addition to the highly variable
topography, the soil fertility of this area has particularly contributed to its biodiversity.
III.1.4.4. Distribution by altitude
The genera Ancistrothyrsus and Dilkea reach altitudes of 800 m, mostly in the Amazonian
region. In contrast, the genus Passiflora is distributed between sea level and 3,700 m.
Figure 4 presents the relationship between elevation range and species diversity (species
number related to diversity) for genus Passiflora, which appears trimodal, with maximal
values below 500 m and in the ranges 1000-1,500 and 2,500-3,000 m. The species number
decreases sharply after 3,500 m until the limit of 4,000 m. With the aim of understanding
better this very particular altitudinal distribution, we have taken into account the
complexity of this genus, gathering the sixteen subgenera of Passiflora present in Colombia
into five groups defined on morphological and molecular grounds (cfr. chapters IV and V),
and analyzed altitudinal distribution of richness in these species subsets. This grouping is
similar to that of the proposal of Feuillet & McDougal (2003), except that Killip’s
subgenera Rathea and Tacsonia are maintained as a distinct fifth group, because of its large
and elongated flowers, mostly with red or pink color and very reduced crown, specifically
adapted to pollination by the sword-hummingbird. The four others correspond to (1)
subgenus Astrophea (trees and shrubs), (2) a Decaloba-like group, equivalent to subgenus
Decaloba sensu Feuillet & McDougal (subgenera Apodogyne, Decaloba, Murucuja,
Porphyropathanthus, Pseudomurucuja and Psilanthus; mostly species with laminar
nectaries, small apetalous flowers, pollinated by bees and small insects, and small fruits),
81
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
(3) a Deidamioides-like group, equivalent to subgenus Deidamioides sensu Feuillet &
MacDougal (subgenera Deidamioides and Tryphostemmatoides), and (4) a Passiflora-like
group gathering subgenera Calopathanthus, Distephana, Dysosmia, Dysosmioides,
Passiflora, and Manicata, i.e. species with large flowers, pollinated by large bees or
hummingbirds, and large fruits.
Figure 3. Species richness observed for Passifloraceae in 1x1 km grid cells in Colombia (165 species). Points
on the maps represent sites of collection.
82
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
The comparison between partial curves clearly shows three distinct patterns in the adaptive
potential of these five groups. Astrophea and the Passiflora-like group present a bimodal
distribution with a first cohort of species adapted to tropical lowlands below 500 m, with 16
and 28 species respectively, and a second one adapted to medium elevations (from 1000 to
2,000 m). Very few species can be found at higher altitudes, with only one record of
P. lindeniana Planch. ex. Triana & Planch. just below 2,700 m for Astrophea, and seven
species for Passiflora. The opposite is true for subgenus Tacsonia, which shows exclusive
adaptation to cold highland climates, as it is typically concentrated between 2,500 and
4,000 m, with a peak between 2,500 and 3,000 m. The fast radiation of this subgenus is
clearly the cause of the peak of the global curve around 2,500-3,000 m. The third pattern is
that of the Decaloba-like group, whose range of adaptation extends spectacularly from 0 to
more than 3,000 m, with no lowland peak and a slight peak around 1000-1,500 m, and the
small Deidamioides-like group, which shows a similar quite uniform distribution of species
richness from 0 to 3,150 m, mostly in the Pacific and Andean regions. An interrogation
remains concerning the first inflexion of the global curve and those of Astrophea and
Passiflora-like groups in the range 500-1000 m. Interestingly, Jørgensen & León-Yánez
(1999) report a bimodal altitudinal distribution of vines in the Ecuadorian flora, with
maximal diversity below 500 m and in the 2,000-3,000 m range, and a maximal diversity
for Passiflora between 2,500 and 3,000 m. Taking latitudinal variation into account (the
same Tacsonia species usually show a higher distribution in Ecuador, with a difference of
about 300-500 m), this corresponds very well with our observations in Colombia.
Considering all Passifloraceae, the variation in number of Ecuadorian species with altitude
(Kessler, 2002) follows the same pattern as in Colombia. The Ecuadorian richness and high
endemism level for subgenus Tacsonia is another strong point of convergence with the
Colombian situation. According to Jørgensen & León-Yánez (1999) bimodality in
altitudinal vine diversity distribution might be due to differential collecting intensity.
However, there is no reason to expect a more continuous pattern. Indeed, Kessler (2002)
showed that there is no common elevational pattern for diversity, but a wide variety of
independent patterns at all taxonomic levels, with maxima at different elevations, and that
endemism appeared highest in the narrowest and most fragmented elevational belts. “The
degree to which these influences become visible along the elevational gradient are
83
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
determined by which combination of species is analysed”. Strikingly, the same conclusion
may be drawn within Passifloraceae, and particularly within Passiflora, taking into account
infrageneric divisions. This result restricts the potential use of Passifloraceae as an indicator
group to the Andean region, where they have developed most of their diversity.
Figure 4. Distribution of total species richness (within circles) and species relative diversity in relation to
altitude in Colombia (3,930 observations), for Passiflora and five infrageneric groups.
III.1.4.5. Climatic requirements
A PCA was carried out on the 19 climatic variables for the 3,930 records of our dataset.
The first two principal components explain 77% of the total variance (Table 2). The first
one accounts for half of the variation (49%) and is strongly correlated (r > 0.97) with
variables associated to temperature (maximum, mean and minimal, but not seasonality in
temperature). The second one explains 28% of the total variation and is related principally
with precipitation in the whole year and in particular seasons (but again, not for seasonality
in precipitation). Figure 5 shows the distribution of species in the principal plane. The first
axis differentiates Andean species adapted to temperatures below 15 °C (i.e. >2,000 m), on
the left side from those growing below 2,000 m, on the right side. Characteristically, these
rightmost species originate from the Amazonian and Orinoquian. The second axis separates
84
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
the species in respect to precipitation. Thus P. arbelaezii L. Uribe, P. costaricensis Killip,
P. chocoensis Gerlach & Ulmer, P. lobata (Killip) Hutch. ex MacDougal, P. occidentalis
Hernández, P. pacifica L.K. Escobar, P. palenquensis Holm-Niels. & Lawesson and P. tica
Gómez-Laur. & L.D. Gómez logically shows preferences for high precipitation, a
predominant condition in the Pacific region, and all are predicted to exist sympatrically. At
the other extreme of the second axis, are species adapted to lower precipitation levels,
specifically to the marked dry season of the Caribbean region, such as P. bicornis Mill.,
P. serrulata Jacq., P. guazumaefolia Juss. and P. pallida L. Amazonian species tend to take
an intermediate position.
Table 2. Factor loadings, eigenvalues and percentages of variance for the first four components, resulting
from the PCA analysis on 19 bioclimatic parameters for the 3,930 collection points (Colombian
Passifloraceae).
Bioclim Parameters
Principal component
Annual Mean Temperature
Mean Monthly Temperature Range
Isothermality
Temperature Seasonality
Max, Temperature of Warmest Month
Min, Temperature of Coldest Month
Temp, Annual Range
Mean Temperature of Wettest Quarter
Mean Temperature of Driest Quarter
Mean Temperature of Warmest Quarter
Mean Temperature of Coldest Quarter
Annual Precipitation
Precipitation of Wettest Month
Precipitation of Driest Month
Precipitation Seasonality
Precipitation of Wettest Quarter
Precipitation of Driest Quarter
Precipitation of Warmest Quarter
Precipitation of Coldest Quarter
Eigenvalue
Percentage of variance
1
2
3
4
0.98
0.08
0.00
0.45
0.97
0.98
0.08
0.98
0.98
0.98
0.98
0.24
0.29
0.09
0.23
0.28
0.09
0.10
0.29
9.26
0.17
-0.21
0.06
0.03
0.16
0.20
-0.22
0.17
0.18
0.17
0.17
0.96
0.91
0.91
-0.55
0.91
0.93
0.87
0.89
5.37
0.09
-0.16
-0.95
0.77
0.12
0.06
0.37
0.09
0.10
0.11
0.07
0.04
0.15
-0.28
0.60
0.17
-0.25
-0.20
0.05
1.74
-0.03
-0.96
-0.01
-0.18
-0.12
0.04
-0.89
-0.02
-0.04
-0.04
-0.03
0.10
0.10
0.13
0.00
0.09
0.13
0.12
0.02
1.51
48.74
28.26
9.15
7.94
The species repartition in the principal plane consistently reflects the potential for climatic
adaptation of the groups that were defined for the analysis of altitudinal distribution. Thus,
subgenus Tacsonia shows a marked adaptation potential to cool conditions, while subgenus
85
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Astrophea and the Passiflora-like group show much higher potential in hot and mild
climates. The genera Ancistrothyrsus and Dilkea are even more clearly adapted to lowlands,
mainly in the Amazonian region. The Decaloba-like group shows a much broader
adaptation range, explaining its quite constant presence across the different biogeographic
regions.
Figure 5. Distribution of Passifloraceae species in the PCA principal plane for climatic variables, with
indication of genera (Ancistrothyrsus and Dilkea) and subgenera of genus Passiflora.
III.1.4.6. Areas of distribution - endemic species
Table 3 gives MaxD and CA50 for the 165 Passifloraceae native species of Colombia.
Figure 6 shows a good correspondence (R² = 0.77) between these distribution parameters,
whose comparison also provides information on species dispersion. Species with a high
MaxD but relatively low CA50 indicate low densities, as a result of biological rarity and/or
under-collection, whereas high CA50 relative to MaxD indicates high density due to
86
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
population biology and/or intense collection. The species with the widest distributions in
Colombia (more than 1,100 km MaxD) are those that show a very wide distribution
including most neotropical regions, such as the very common P. foetida L., P. auriculata,
P. quadrangularis and P. laurifolia L. (wild and cultivated) P. suberosa L, and
P. serratodigitata L., P. capsularis L., P. rubra L., and P. misera Kunth, and others of still
considerable regional distribution, such as P. vitifolia Kunth, P. coccinea Aubl., P. spinosa
(Poepp. & Endl.) Mast., P. nitida, P. subpeltata Ortega, P. maliformis, P. menispermifolia
Kunth, P. biflora Lam., and Dilkea parviflora Killip. Only P. arborea Spreng (Panamá to
Ecuador) and P. cumbalensis (Colombia to Peru) show a more restricted distribution. These
high-MaxD species are concentrated at low to medium elevations, the only exception being
P. cumbalensis. According to IUCN criteria, they are not threatened (Least Concern
category), except for P. arborea (Near Threatened; Ocampo et al., 2007). Between 200 and
1,100 km of MaxD, there are species of regional importance, such as P. mixta L.f.,
P. ligularis, and endemics with a relatively wide distribution, such as P. sphaerocarpa
Triana & Planch. (805 km²), P. lehmanni Mast. (878 km²), P. antioquiensis and P. mollis
Kunth. The latter displays a relatively high CA50 in its group, as its 17 observations are
quite scattered along the Cordillera Occidental. The position of P. coriacea Juss. in this
group of medium dispersion is very surprising, as it is found in all neotropical countries.
The 71 species with MaxD values below 225 km include 36 narrow endemics, 21 of which
are exclusive to nine departments, particularly Antioquia (six species), Tolima (four) and
Santander (three). The 14 others show similar MaxD and CA50 but live across
administrative divisions. Only four of these 36 narrow endemics are represented by ten or
more observations while ten species are only known from the type collection. Ocampo et al.
(2007) considered a MaxD under 100 km as a criterion of rarity and threat, which places 26
endemic species in a very critical situation. The situation of 33 non-endemic species with a
MaxD less than 100 km must be examined in relation to their distribution in neighbor
countries. P. truxilliensis Planch. & Linden ex Triana & Planch. is shared with Venezuela,
but it is a narrow endemic living around the border. The distribution of 14 species extends
to farther places in neighboring countries, and 18 species present a wide distribution,
extending to non-neighboring countries. For example, P. tricuspis is only reported once, in
the Andean foothill, so it has a null MaxD, however its distribution extends south to
87
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Bolivia. Sixteen of these 30 species are adapted to lowland conditions, which suggests that
their apparent rarity is in fact due to the poor collecting in the corresponding regions.
Figure 6. Extent of Passifloraceae species distribution in Colombia: circular area (CA50) vs. maximum
distance (MaxD).
88
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Table 3. Total number of Passifloraceae present in Colombia. Number of observations, maximum distance (MaxD) and circular area (CA) for each species.
Endemic species are highlighted by an asterisk (*). RC: species rare for Colombia; Roc: species rare in other countries; Rne: rare narrow endemic, Ne: narrow
endemic; Re: rare endemic; Ce: common endemic. Presence in a richness spot is indicated by Rs and the identification number used in Figure 7.
Species
Ancistrothyrsus antioquiensis L.K. Escobar (ined.)*
Ancistrothyrsus tessmannii Harms
Dilkea johannesii Barb. Rodr.
Dilkea parviflora Killip
Dilkea retusa Mast.
Passiflora adenopoda Moc. & Sessé ex DC.
Passiflora adulterina L.f. *
Passiflora alnifolia Kunth
Passiflora alata Curtis
Passiflora ambigua Hemsl. ex Hook.
Passiflora andina Killip
Passiflora andreana Mast.
Passiflora antioquiensis H. Karst. *
Passiflora apoda Harms
Passiflora arbelaezii L. Uribe
Passiflora arborea Spreng.
Passiflora auriculata Kunth
Passiflora azeroana L. Uribe *
Passiflora bicornis Mill.,
Passiflora bicuspidata (H.Karst.) Mast. *
Passiflora biflora Lam.
Passiflora bogotensis Benth. *
Passiflora bracteosa Planch. & Linden
Passiflora bucaramangensis Killip *
Passiflora callistema L.K. Escobar *
Passiflora candollei Tr. & Planch.
Nb.
observ.
MaxD
(km)
CA
(km2)
Rare
species
Endemics and distribution
2
1
1
22
5
51
43
121
1
48
1
3
55
43
48
67
128
10
11
16
40
56
7
8
1
4
41
0
0
1,185
952
383
234
1,244
0
929
0
45
667
678
746
1,204
1,635
574
675
438
1,326
1,057
122
70
0
854
11,762
7,814
7,814
40,688
106,159
82,650
39,072
170,761
7,814
137,261
7,814
12,214
99,064
83,615
113,491
144,115
334,952
34,734
52,098
61,674
122,047
89,250
23,180
15,032
7,814
26,294
RC
RC / Roc
RC / Roc
Rne (Antioquia)
89
RC
Ce
RC
RC /Roc
RC
Colombia and Ecuador
Ce
Ce
Ce
RC /Roc
RC
RC
RC
Ce
Colombia and Venezuela
Ne (Santander)
Rne (Bolivar)
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Passiflora capsularis L.
Passiflora chelidonea Mast.
Passiflora chocoensis G. Gerlach & T. Ulmer *
Passiflora cincinnata Mast.
Passiflora citrifolia (Juss.) Mast.
Passiflora coccinea Aubl.
Passiflora colombiana L.K. Escobar *
Passiflora coriacea Juss.
Passiflora costaricensis Killip
Passiflora cremastantha Harms *
Passiflora crispolanata L.Uribe *
Passiflora cuatrecasasii Killip *
Passiflora cumbalensis (Karst.) Harms
Passiflora cuneata Willd.
Passiflora cuspidifolia Harms,
Passiflora danielii Killip *
Passiflora dawei Killip *
Passiflora emarginata Humb. & Bonpl. *
Passiflora engleriana Harms *
Passiflora erytrophylla Mast. *
Passiflora escobariana J.M. MacDougal
Passiflora filipes Benth.
Passiflora fimbriatistipula Harms *
Passiflora flexipes Triana & Planch. *
Passiflora foetida L.
Passiflora formosa T. Ulmer *
Passiflora glandulosa Cav.
Passiflora gleasonii Killip
Passiflora gracillima Killip
Passiflora grandis Killip *
Passiflora gritensis H. Karst.
Passiflora guatemalensis S. Watson
Passiflora guazumaefolia Juss.
Passiflora hahnii (Fourn.) Mast.
Passiflora haughtii Killip *
Passiflora hirtiflora Jorgensen & Holm-Nielsen
64
18
1
1
3
21
2
59
1
1
11
9
156
9
33
5
4
46
2
6
2
3
18
24
143
1
1
2
29
2
8
11
8
1
1
1
1,437
1,024
0
0
68
1,285
42
741
0
0
246
181
1,196
877
812
180
208
654
110
225
3
48
198
322
1,830
0
0
3
684
14
346
971
349
0
0
0
90
159,962
94,209
7,814
7,814
14,049
107,128
11,910
136,372
7,814
7,814
29,720
21,312
199,941
50,607
86,640
20,590
23,702
78,393
8,902
27,643
8,136
13,227
33,664
36,121
420,440
7,814
7,814
8,075
74,546
9,161
26,115
59,505
41,192
7,814
7,814
7,814
RC
RC
RC
Rne (Choco)
RC
Rne
RC
RC
RC
RC
RC
RC / Roc
RC / Roc
Rne (Cauca)
Ce
Ne (Rs 2)
Rne (Antioquia, Rs 2)
Rne (Rs 3)
Ce
Rne (Antioquia, Rs 2)
Ne (Rs 2,3)
(Antioquia, Rs 2)
México to Ecuador
Ne
Ce
RC
RC
RC
Rne (Boyacá)
RC
Rne
RC
RC
RC
Rne (Santander)
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Passiflora holosericea L.
Passiflora holtii Killip
Passiflora hyacinthiflora Planch. & Linden *
Passiflora involucrata (Mast) A.H. Gentry
Passiflora jardinensis L.K. Escobar *
Passiflora kalbreyeri Mast. *
Passiflora killipiana Cuatrecasas
Passiflora lanata (Juss.) Poir. *
Passiflora laurifolia L.
Passiflora lehmanni Mast. *
Passiflora leptomischa Harms *
Passiflora ligularis Juss.
Passiflora lindeniana Planch. ex Triana & Planch.
Passiflora linearistipula L.K. Escobar *
Passiflora lobata (Killip) Hutch. ex J.M. MacDougal
Passiflora longipes Juss. *
Passiflora lyra Planch. & Lind. ex Killip
Passiflora macrophylla Spruce ex Mast.
Passiflora magdalenae Triana & Planch. *
Passiflora magnifica L.K. Escobar*
Passiflora maliformis L.
Passiflora manicata (Juss.) Pers.
Passiflora mariquitensis Mutis ex Uribe *
Passiflora megacoriacea Porter-Utley (ined.)
Passiflora menispermacea Triana & Planch. *
Passiflora menispermifolia Kunth
Passiflora micropetala Mast.
Passiflora misera Kunth
Passiflora mixta L. f.
Passiflora mollis Kunth
Passiflora monadelpha Jorgensen & Holm-Nielsen
Passiflora morifolia Mast.
Passiflora multiformis Jacq.
Passiflora munchiquensis Hernandez (ined.)*
Passiflora mutisii Killip *
Passiflora nitida Kunth
7
1
3
8
8
19
1
32
11
17
21
101
2
4
3
21
4
20
21
6
122
62
3
1
2
43
11
54
162
17
7
1
4
4
1
72
238
0
305
1,197
35
426
0
284
1,350
805
449
914
395
8
194
334
69
716
129
33
1,208
889
10
0
18
1,410
1,318
1,148
966
554
67
0
147
200
0
1,452
91
25,632
7,814
17,746
48,827
11,335
41,237
7,814
45,476
84,672
91,156
46,331
170,123
15,628
8,695
23,115
45,557
14,716
90,432
31,127
12,215
212,270
114,036
8,436
7,814
9,610
167,659
68,015
145,398
191,787
208,941
33,665
7,814
17,652
22,441
7,814
279,511
RC / Roc
RC
RC
RC / Roc
Re
Ne (Antioquia)
Ce
Colombia to Peru
Ce
Ce
Ce
RC / Roc
RC
RC
Colombia and Venezuela
Rne (Caldas)
Ce
RC / Roc
RC
RC
RC
RC
RC / Roc
RC
RC
RC
RC
Ne
Ne (Antioquia, Rs 2)
Rne (Tolima)
Rne (Tolima, Rs 4)
Colombia and Ecuador
Rne
Rne (Tolima)
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Passiflora occidentalis Hernandez (ined.)*
Passiflora oerstedii Mast.
Passiflora pacifica L.K. Escobar *
Passiflora palenquensis Holm-Niels. & Lawesson
Passiflora pamplonensis Planch.& Linden ex Tr. & Planch. *
Passiflora pallida L.
Passiflora panamensis Killip
Passiflora parritae (Mast.) L.H. Bailey *
Passiflora pennellii Killip *
Passiflora phaeocaula Killip
Passiflora picturata Ker
Passiflora pilosissima Killip *
Passiflora pinnatistipula Cav.
Passiflora pittieri Mast.
Passiflora platyloba Killip
Passiflora popayanensis Killip *
Passiflora popenovii Killip
Passiflora punctata L.
Passiflora purdiei Killip *
Passiflora putumayensis Killip
Passiflora pyrrhantha Harms
Passiflora quadrangularis L.
Passiflora quadriglandulosa Rodschied
Passiflora quindiensis Killip *
Passiflora resticulata Mast. & André
Passiflora rigidifolia Killip *
Passiflora riparia Mart. ex Mast.
Passiflora rubra L.
Passiflora rugosa (Mast.) Triana & Planch
Passiflora schlimiana Triana & Planch. *
Passiflora securiclata Mast
Passiflora seemannii Griseb.
Passiflora semiciliosa Planch & Linden *
Passiflora serratodigitata L.
Passiflora serrulata Jacq.
Passiflora sexflora Juss.
10
41
9
20
1
6
15
14
6
5
1
2
21
1
4
6
17
8
1
1
1
112
4
8
4
1
3
90
12
7
4
40
4
18
10
14
474
728
510
1,181
0
898
295
100
313
498
0
270
750
0
201
64
636
592
0
0
0
1,676
414
225
414
0
716
1,351
421
181
849
1,341
578
1,566
331
353
92
42,350
148,975
39,585
100,769
7,814
50,078
41,614
20,357
24,413
28,305
7,814
15,628
57,114
12,661
16,471
15,078
31,075
40,022
7,814
7,814
7,814
314,317
21,256
24,711
18,938
7,814
23,442
117,934
35,549
27,852
30,708
129,777
26,175
67,105
29,354
43,143
Ce
Ce
RC
Rne (N. de Santander)
RC
Ne
Ce
RC / Roc
RC
Re
RC
RC
RC
RC
RC / Roc
RC / Roc
Ne (Cauca)
Rne
RC
Ne (Tolima, Rs 4)
RC
RC
Rne (Antioquia)
Roc
RC / Roc
Ne
Colombia and Venezuela
RC
Re
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Passiflora sierrae L.K. Escobar *
Passiflora skiantha Huber
Passiflora smithii Killip
Passiflora sodiroi Harms
Passiflora sphaerocarpa Triana & Planch. *
Passiflora spicata Mast.
Passiflora spinosa (Poepp. & Endl.) Mast.
Passiflora suberosa L.
Passiflora subpeltata Ortega
Passiflora tarminiana Coppens & Barney
Passiflora tenerifensis L.K. Escobar *
Passiflora tica Gómez-Laur. & L.D. Gómez
Passiflora tiliifolia L.
Passiflora tolimana Harms *
Passiflora trianae Killip *
Passiflora tribolophylla Harms *
Passiflora tricuspis Mast.
Passiflora trinervia (Juss.) Poir.*
Passiflora tripartita (Juss.) Poir.
Passiflora trisulca Mast. *
Passiflora truxillensis Planch. & Linden ex Triana & Planch.
Passiflora tryphostemmatoides Harms
Passiflora tuberosa Jacq.
Passiflora uribei L.K. Escobar *
Passiflora ursina Killip & Cuatrec.
Passiflora variolata Poepp. & Endl.
Passiflora venosa Rusby
Passiflora vespertilio L.
Passiflora vestita Killip
Passiflora vitifolia Kunth
Passiflora x rosea (H. Karst.) Killip
2
1
28
1
35
1
20
66
35
28
4
5
48
12
2
1
1
27
56
8
1
25
1
3
2
6
1
3
1
359
7
46
0
827
0
878
0
1,521
1,497
1,344
832
71
319
1,010
426
39
0
0
220
1,210
441
0
557
0
54
7
412
0
292
0
1,729
161
93
12,194
7,814
72,555
7,814
96,244
7,814
118,197
158,860
89,527
103,373
15,195
23,119
97,205
33,711
11,594
7,814
7,814
36,932
145,398
25,258
15,628
77,831
7,814
12,960
8,503
27,059
7,814
20,887
7,814
456,229
20,988
RC
RC / Roc
Rne (Magdalena)
Colombia and Peru
RC / Roc
Colombia and Ecuador
Ce
Colombia and Brazil
RC / Roc
RC
RC
RC
RC
RC
Rne (Valle del Cauca)
Ce
Rne
Rne
Ne (Rs 5)
RC / Roc
Ce
Colombia and Venezuela
RC
RC
RC / Roc
Rne
Colombia and Ecuador
RC / Roc
RC
RC / Roc
Colombia and Ecuador
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
III.1.4.7. Potential distribution of species and species assemblages
For each of the 80 species with more than ten observations, the BioClim model provided a
probabilistic prediction of its geographic distribution using climate data as the driver of
species distribution. The results cannot be presented here for each individual species, but
are available upon request from the authors. The potential distribution of richness was
calculated by summing the 80 spatial predictions of species occurrence (Figure 7). The
areas of highest predicted species richness (41 to 54 predicted sympatric species) are
mostly located in the center of the country, on the slopes of the three cordilleras, between
1000 and 2,000 m of elevation. Collection has been intense in many of them, showing a
high diversity, however the correspondence is not perfect between observed and modeled
distribution. While the species-rich areas of Antioquia and Caldas, Quindío, Cundinamarca
and eastern Boyacá, and even the poorly explored but promising Santander, are well
represented on the map (areas 2, 5, 3, 4 and 1 respectively), only very small richness spots
are drawn for Valle del Cauca (area 7), or no richness spot indicated for Cauca and
southern Huila. Conversely, predicted richness spots 6, 8 and 9 (eastern Tolima-northern
Huila- southern Cundinamarca, western Caquetá, Nariño) were not detected in the analysis
of observed diversity, indicating potential collecting gaps. The model predicts a very poor
representation of Passiflora in the lowlands of the Caribbean and Orinoquian and part of
the Pacific, as well as in the Sierra Nevada de Santa Marta, an isolated mountain range on
the Caribbean Coast, reputed for its high level of endemism. In both cases, this may be
attributed to the poor exploration of these areas (low densities of observation) and to poor
representation of their species (few observations per species) resulting in them not having
sufficient observations to be used in the predictive modeling. This bias can be corrected by
further collecting in these regions. Alternatively, materials of Colombian species collected
in border regions of neighboring countries, belonging to the same biogeographic entities
(e.g. the Venezuelan Llanos for the Orinoquian, Costa Rican and Ecuadorian Pacific,
Brazilian, Ecuadorian and Peruvian Upper Amazonian) might be used to refine these
models and increase the number of observations per species under analysis.
The biodiversity hotspot concept not only considers species richness and diversity but also
endemism. In an analysis of New Zealand fern diversity, Lehmann et al. (2002) observed a
94
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Figure 7. Modeled distribution of Colombian Passifloraceae diversity based on data from 80 species
presenting more than 10 observations. Ellipses individualize high richness spots mentioned in the text.
poor correlation between both parameters. For Colombian Passifloraceae, we could not
establish rigorously their correspondence, as the analysis was not designed for rare species,
95
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
however we compared their distributions, distinguishing four categories among the 58
endemics: those with a relatively wide distribution (MaxD > 100 km, 19 species), the
narrow endemics (MaxD under 100 km, three species), the rare endemics (less than six
observations, three species), and the rare narrow endemics (both criteria, twenty species).
Six of the 10 narrow endemics and seven of the 25 rare narrow endemics live in one of the
nine hotspots defined by the diversity analysis. None of the three rare endemics are found
in those sites. Of the 27 rare and/or narrow endemics living out of those hotspots, three are
considered extinct (Ocampo et al., 2007), seven are only adapted to lowlands, which easily
explains their absence in Andean hotspots; two species are endemic to the Sierra Nevada de
Santa Marta, an area of endemism that could not be sufficiently taken into account for
reasons explained previously. Finally, only 11 of the 27 living Andean rare/narrow endemic
species, i.e. less than 50%, live in one of the hotspots. This proportion must be compared
with more than 54 out of 80 in the case of the non-rare species whose distribution
determined those hotspots. Thus, preserving these nine areas should have a less positive
impact on the conservation of narrow endemics than on the general Passifloraceae
diversity. Using an analysis of complementarity reserve selection developed by Rebelo
(1994), 52 sites of 25km x 25km were selected to represent all 165 native species
throughout the country. The best five sites, in Caldas, Risaralda, Norte de Santander,
southern Antioquia and Boyacá, capture a total of 64 species. In just seven sites, 50% of all
species could be conserved, though many of the endemic/rare species are not captured in
these sites.
III.1.4.8. Conservation of Passifloraceae and their habitat
Figure 8 combines the estimated distribution of Passifloraceae diversity with that of
protected areas in the Andes of Colombia, showing a general lack of correspondence. Most
Andean protected areas are concentrated around the summits, obviously targeting páramo
ecosystems. Very few small protected areas harbor a high diversity of Passifloraceae: the
watershed forest reserves of Sierra del Peligro (Boyacá, 16.5 km²), Río Nare (Antioquia,
118.8 km²), Río San Francisco, Cuchillas Peñas Blancas, Cerro Quininí (Cundinamarca,
28.8, 16.3 and 18.0 km²), whose responsibility has recently been upgraded to the
department level. The Parque Nacional Farallones (Valle del Cauca) is the only reserve of
96
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
national importance to protect part of a small Passifloraceae hotspot, on the eastern fringes
of the nature reserve. This poor coverage is not good news, neither for a family including
71% threatened species, nor for the habitats where these species have developed numerous
interactions with many other organisms.
Figure 8. Distribution of protected areas in Colombia, showing poor correspondence with areas of high
Passifloraceae diversity.
Figure 9 shows a striking general superposition of areas of high Passifloraceae diversity on
certain Colombian coffee growing zone ecotopes (Cenicafé, 2005) whose conservation is of
97
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
the utmost importance for the country. This is not surprising, as the corresponding elevation
belts include or enclose those of major diversity (1000-2,500). It clearly shows that efforts
for the conservation of Passifloraceae habitats and genetic resources must be integrated in
the more general management of the coffee growing zone environment at the landscape
level. The latter can be ensured by coordinating existing actions for watershed protection,
management of private and low-level public reserves, as well as recreational areas for
public education (rural and peripheral urban parks, arboreta, etc.), creation of
environmental corridors, and improvement of agricultural practices, particularly in coffee
farms (crop association, shade tree diversity, and guadua bamboo growing for building
materials and water management at farm level).
III.1.4.9. Passifloraceae as indicators of biodiversity
According to Pearson (1994), an ideal indicator taxon should cumulate seven criteria: (i) a
well-known and stable taxonomy, (ii) well known natural history, (iii) readily surveyed and
manipulated, (iv) higher taxa broadly distributed geographically and over a breadth of
habitats, (v) lower taxa specialized and sensitive to habitat changes, (vi) patterns of
diversity in other taxa, and (vii) potential economic importance. Passifloraceae clearly fill
the fifth and seventh criteria, taking into account that several common species are indicators
of more or less perturbated habitats. Concerning the fourth criterion, our analyses have
repeatedly underlined that Colombian Passifloraceae distribution is concentrated in the
Andean region, so their use as indicators should be restricted to the corresponding elevation
belts. Lianas growing in high trees are not always easily surveyed (third criterion), however
their typical structures, showy flowers and interesting fruits make them easy to identify as a
group, catching the attention of local populations and specialists (e.g. protected area staff),
who can thus help in localizing the different species in particular places. The application of
molecular techniques should produce important progress in the complex taxonomy of this
group and further the understanding of its natural history. The sixth criterion is particularly
important. The numerous interactions of Passifloraceae with other organisms (surrounding
vegetation, pollinators, and herbivores) constitute a first indication that their diversity is
necessarily related to that of other components of the ecosystem. Another indication came
from a preliminary study, where we found an excellent correspondence between the
98
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
distributions of diversity of Passiflora and Vasconcellea (mountain papayas), another plant
group whose radiation is also clearly related to the rise of the Andes (Scheldeman et al.,
2006). Similar results must be obtained with more plant taxa before considering
unequivocally Passifloraceae as a reliable surrogate for floral diversity in Andean
ecosystems. However, given the excellent correspondence between Passifloraceae diversity
maps and coffee growing zone ecotope maps, we may already recommend them as useful
indicators of habitat degradation or restoration in this environmentally and economically
very important region.
Figure 9. Correspondence between Passifloraceae high richness spots and coffee growing zone ecotopes.
99
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
III.1.5. Conclusions
Collections of Passifloraceae have not been uniform as a consequence of difficulty of
access and/or chronic social conflict in many areas. It has been much denser in the central
coffee growing zone, Antioquia, Valle del Cauca and Cundinamarca. The southern and
northeastern Andes, and the Caribbean have been little explored. For the lowland forests of
the Pacific, the Orinoquian and the Amazonian, data are so poor that they are misleading.
Despite the resulting sampling bias, collecting parameters clearly point to the concentration
of observed Passifloraceae diversity in the Andes, and more particularly the central coffee
growing zone. This is further highlighted by the elevational pattern of diversity, showing
three peaks, a small one for lowlands, and two higher ones at intermediate and high
elevations (1000-2,000 and 2,500-3,000 m). A more thorough analysis shows that this
trimodal pattern corresponds to different adaptive evolutions among genera and
infrageneric divisions of Passiflora. This is also reflected in the analysis of the climatic
preferences of these infrageneric groups.
The analysis of species distribution areas shows a trend for more extent dispersion of
species occurring at low and intermediate elevations. On the contrary, narrow endemics are
more frequent among highland species.
The modeled species richness map allowed to identify nine richness spots of variable size,
three of which, located in the southern and southeastern Andes of Colombia, correspond to
collection gaps, as they were not detected in the analysis of observed diversity. Another
probable collection gap, not detected by diversity modeling, corresponds to the Sierra
Nevada de Santa Marta, an isolated mountain range with both high diversity and endemism.
The proportion of endemics living in high richness spots is lower than the proportion of all
species used for modeling, confirming the lack of relation between diversity concentration
and endemism reported in other studies. If this is further substantiated in different groups of
organisms, it could limit the application of the biodiversity hotspot concept, as the best
protected areas for diversity would not necessarily provide protection to a high proportion
of narrow endemics.
100
Chapter III. Diversity and in situ conservation
_________________________________________________________________________
Passifloraceae diversity is not currently conserved by the current network of Colombian
protected areas. On the contrary, it is particularly concentrated on certain ecotopes of the
coffee growing zone, i.e. highly disturbed habitats, so any conservation effort must be
integrated in local management strategies at the landscape level. Passifloraceae may
provide an interesting indicator to evaluate the outcome of such efforts.
III.1.6. Acknowledgements
The first author gratefully acknowledges financial support from the Gines-Mera Fellowship
Foundation (CIAT-CBN). Part of this research has been funded by the Colombian Ministry
for Environment and the Research Center of the Colombian Coffee Grower Federation
(Cenicafé) through of the collaborative project “Estudio de la Diversidad de las
Passifloraceae y Caricaceae en la zona cafetera de Colombia”. Finally, we thank Dr.
Xavier Scheldeman (Bioversity International) for his scientific advice regarding data
analysis.
101
CHAPTER IV
___________________________________
A phenetic analysis of morphological diversity in
the genus Passiflora L.
Chapter IV. Morphological diversity
______________________________________________________________________
IV.1. A phenetic analysis of morphological diversity in the genus
Passiflora L.
John Ocampo Pérez1* and Geo Coppens d’Eeckenbrugge2
1
Bioversity International (formerly IPGRI), Regional Office for the Americas, A.A. 6713,
Cali, Colombia.
2
CIRAD/FLHOR, UPR ‘Gestion des ressources génétiques et dynamiques sociales’,
Campus CNRS/Cefe, 1919 route de Mende, 34293 Montpellier, France.
IV.1.1. Abstract
Morphological variation was studied in 124 accessions from eight subgenera and 60
species among the most common cultivated and wild species of genus Passiflora, using
the analysis of variance components and principal component analysis (PCA) on 43
quantitative traits, and neighbor joining cluster analysis on 84 qualitative traits. The
coefficients of variation appear generally superior for subgenus Decaloba, indicating
stronger differentiation, as compared with subgenera Passiflora and Tacsonia. Twentyfour quantitative descriptors showing high variation at the subgenus level, were selected
for the PCA. The five principal components retained represent 84% of the total variation.
The first one (32%) is closely associated with flower length (hypanthium, nectary
chamber, androgynophore) and secondarily with the constriction of the floral cup above
the nectary chamber. The second one (27%) is associated with flower width (length of
bracts and length of corolla and corona elements) and bract shape. The third one (14%)
is associated with peduncle branching, stem width and leaf length, which relates it
clearly with variation between subgenus Astrophea, and secondarily subgenus
Tryphostemmatoides, and all other subgenera. The projection of accessions in the
resulting tridimensional space consistently separates subgenera. A selection of 32
qualitative traits and four categorized quantitative variables, whose segregation follows
divisions among Killip’s subgenera, allowed classifying our 60-species sample
consistently, using a strictly phenetic approach. Most discriminating characters include
size of stems and leaves, presence of tendrils, number and distribution of extrafloral
nectaries, dimensions and general shape of bracts, width and length of flowers, corona
complexity, ovary shape, and, although they could not be systematically analyzed, fruit
size and color. Eight of the nine Killip’s subgenera represented in our sample are
supported by the morphological analysis, although subgenus Tryphostemmatoides is
only supported in the quantitative analysis. In a second analysis, 74 qualitative
descriptors were incorporated in the cluster analysis, which increased distances and
improved bootstrap values, without affecting the general structure of the dendrogram,
neither at the subgeneric nor at the interspecific level. Our results support seven of the
eight Killip’s subgenera of our sample, but no infrasubgeneric classifications. However,
the new classification of subgenus Decaloba by Feuillet & MacDougal was partly
supported. They converge on many points with previous phylogenetic results obtained
with DNA sequences, although the latter group subgenera Tacsonia and Distephana
with subgenus Passiflora.
Key words: Passiflora L, systematics, morphological descriptors, principal components,
variation.
103
Chapter IV. Morphological diversity
______________________________________________________________________
IV.1.2. Introduction
With more than 525 species, Passiflora L. is the most important genus of the
Passifloraceae, a family including nearly 630 species. It is essentially neotropical, with
only 22 species native to the Old World, in Southeast Asia, Australia and Oceania.
Passionflowers are herbaceous or woody vines, usually climbing by tendrils, but a few
are trees or shrubs. Other typical vegetative traits include alternate leaves, axillary
stipules, and petiolar and/or laminar nectary glands. In addition, the genus exhibits
several unique floral features, such as an androgynophore, a complex corona,
constituted of one or several concentric rows of filaments, and a limen-operculum
system limiting access to the nectary chamber, with impressive interspecific variation in
size, shape and colors (see Figure 3 of Chapter I and Figure 1 of present chapter).
Many species exhibit interesting fruits, pharmacological properties (e.g. sedative effect)
and/or ornamental potential. More than 80 Passiflora species produce an edible fruit,
the most interesting ones belonging to subgenera Passiflora and Tacsonia (Coppens
d’Eeckenbrugge, 2003). The two botanicals forms of P. edulis Sims, flavicarpa Degener
(yellow maracuja) and edulis (purple maracuja) are by far the most important crops in
the
family,
with
a
world
production
estimated
at
ca.
640.000
tons
(http://www.passionfruitjuice.com). Other cultivated passion fruits are P. tripartita
var. mollissima (Kunth) Holm-Nielsen & Jørgensen (curuba de Castilla), P. tarminiana
Coppens & Barney (curuba India), P. ligularis Juss. (sweet granadilla), P. maliformis L.
(granadilla de piedra), P. quadrangularis L. (giant granadilla), P. popenovii Killip
(granadilla de Quijos), P. alata Curtis (fragrant granadilla) and P. laurifolia L. (golden
apple). These eight species are mainly commercialized on South American local and
national markets, principally in Colombia and Brazil, with incursions on the
international market. Passion fruits are consumed fresh or processed into juices,
sherbets, ice cream, and components of industrial pastry and candies. The most
important commercial species are susceptible to a large number of pests and diseases,
with considerable negative effects on production. The high potential of Passiflora for
crop diversification and economic development induced research institutions in the
Andean countries to prioritize their characterization and the evaluation of wild and
cultivated populations (Debouck & Libreros, 1995), and develop strategies for
conservation and improvement of these genetic resources.
104
Chapter IV. Morphological diversity
______________________________________________________________________
Figure 1. Schema of a flowering branch of Passiflora vitifolia Kunth (drawing by Jesus Salcedo).
105
Chapter IV. Morphological diversity
______________________________________________________________________
The high diversity of combinations of characters related to the nectary glands, stipules,
corona, operculum and limen (Figure 1 and 2), have been heavily relied upon for
delimiting infrageneric divisions in Passiflora taxonomy, dividing it into subgenera,
sections, and series. Although most species appear well delimited, there are many cases,
in low level taxa (sections and series), where two or more species are very difficult to
distinguish. The reference work is that of Killip (1938), who described 355 American
species, plus 20 species in 1960, placing them in 22 subgenera (Annex 1). This
classification was amended and completed with the renaming of several subgenera, the
suppression of subgenus Tacsoniopsis, and the creation of subgenus Porphyropathantus
(Escobar, 1988a,b, 1989; McDougal, 1994). In his description of the general
Passifloraceae morphology, Killip (1938) underlined the particular contribution of
certain organs to taxonomy. Among those of primary importance, he included the
peduncle, the bracts (shape and position), the operculum, the presence or absence of
petiolar glands and the grooved or reticulate seed surface, the two latter being almost
perfectly correlated within subgenus Decaloba. He also mentioned habit (vines vs.
trees), stipule shape and margin, leaf nervation (subgenus Astrophea), shape and
arrangement of corona elements, petal absence, and fruit size. However, from the
identification keys given by Killip (1938), no clear correlation emerges between main
divisions and a particular character hierarchy. The presence of uncinate trichomes on
the leaf epidermis was given particular importance by MacDougal (1994) in his revision
of section Pseudodysosmia of subgenus Decaloba, which he defined as the “hooked
trichome group”. In the same work, MacDougal mentions the taxonomic importance of
the posture of apex in relation to the main shoot axis. However, a cernuous shoot tip is
found in parts of subgenera Decaloba, Astrophea, and Tacsonia (Ulmer & MacDougal,
2004), which suggests instead that this trait is not determinant for delimiting subgenera.
Recently, Feuillet & MacDougal (2003; Annex 2) have proposed a new general
classification, taking into account the Old World species and recognizing only four
subgenera, Astrophea (unchanged; trees and shrubs, rarely lianas), Decaloba (vines with
small flowers and fruits, the latter usually black), Deidamioides (vines, with twoflowered peduncles, not clearly defined as a morphological group) and Passiflora (vines
to lianas, with large flowers and fruits). However, this classification has not been fully
developed yet, the placement of most species must be deduced from earlier literature,
and its morphological basis is not clear. In the book of Ulmer & MacDougal (2004),
106
Chapter IV. Morphological diversity
______________________________________________________________________
where about 200 species are described and assigned a place in the new classification,
the revision of Passiflora morphology underlines taxonomic importance for shoot apex
posture, extrafloral nectaries, bract size and shape, operculum, and seed coats, but only
bract and seed traits are specifically mentioned as markers of subgenera, sections, and
series. The new classification was presented as the “last premodern” classification,
because it does not take into account the results obtained from molecular biology
techniques. However, even on strictly morphogical grounds, it lacks the support of a
systematic evaluation of morphological diversity in the genus and its subtending
structure. The situation is further complicated as many new species have been described
during the elaboration of this classification, most often by splitting well-known species
on the basis of slight variations from their type, observed in small populations (e.g.
P. miniata Vanderplank, P. longicuspis Vanderplank and P. aimae Anonay & Feuillet,
from P. coccinea Aubl., and P. formosa Ulmer, from P. lanata Juss.).
On the other hand, the first molecular studies carried out on significant Passiflora
species samples have consistently validated the three of the four major subdivisions
proposed by Feuillet & MacDougal (2003). The results of Muschner et al. (2003; Annex
4a), on nuclear ribosomal internal transcribed spacers, (ITS-1 and ITS-2) and plastid
trnL-trnF intergenic spacer, Yockteng (2003; Annex 4b), on chloroplast matK
Yockteng & Nadot (2004; Annex 4c), on chloroplast-expressed glutamine synthetase
(ncpGS), and Hansen et al. (2006; Annex 4d), on sequences analysis of the chloroplast
(rpoC1 and trnL-trnT), confirm the clear separation of three clades corresponding to the
new contours of subgenera Decaloba, Astrophea, and Passiflora. These three major
clades correspond to cytogenetic groups, as they appear characterized by chromosome
numbers of 2n = 12, 24, and 18, respectively. The results of Hansen et al. (2006) also
support the small subgenus Deidamioides, whereas the two species that represent it in
the study of Yockteng & Nadot (2004) are split in two widely divergent branches. In
addition, the latter study indicated that four other small subgenera, Dysosmia (DC.)
Killip, Tryphostemmatoides (Harms) Killip, Polyanthea (DC.) Killip, and Tetrapathea
(DC.) Rchb., should also be recognized. More general problematic points, in these three
genus-wide studies, are the monophyly of genus Passiflora, the position of subgenus
Astrophea, and the low resolution at infrasubgeneric levels, particularly in the
Passiflora clade, questioning the validity of supersections and series, where many
ambiguities and inconsistencies still persist. In fact, such phylogenetic molecular studies
107
Chapter IV. Morphological diversity
______________________________________________________________________
cannot give us a definitive answer on the relationship between morphologically illdefined species and species groups. Actually, if two or more species within a series are
only distinguished by the exact number of filament whorls in the corona, the color of
one of these whorls, or the number of nectary glands on the petiole, we must first
question the discontinuity in the variation described, at both morphological and genetic
levels, and assess objectively the morphological basis of the classification.
Despite the impressive morphological diversity described among Passiflora species,
few studies have compared intra- and intersubgeneric, and intra- and interspecific
variation with statistical tools. A first study was conducted by Villacís et al. (1998) on
the most common species of subgenera Tacsonia and Manicata, on Colombian and
Ecuadorian accessions. Floral traits were mostly represented in their set of 33
qualitative descriptors, and vegetative traits in the set of 28 quantitative descriptors. The
former showed limited intraspecific variation and a consistent picture of interspecific
relations, while the latter provided more information on intraspecific variation but a less
consistent picture of the differences between species. The descriptor list was corrected,
to take into account traits specific to subgenus Tacsonia, and augmented to 62
qualitative and 67 quantitative descriptors, giving a better balance between floral and
vegetative traits. The sample was also amplified, to include more species and more
origins, including similar materials from Venezuela, Colombia, Ecuador, Peru, and
Bolivia (Coppens d’Eeckenbrugge et al., 2002). This study showed wide variation at
both intra and interspecific levels, with a geographic component in the species best
represented, and the overall picture was consistent with results obtained in the same
subgenus with biochemical and molecular data (Segura et al., 2002, 2003, 2005). These
convergent data led to the formal description of the cultigen P. tarminiana Coppens &
Barney as a distinct species (Coppens d’Eeckenbrugge et al., 2001). The same
descriptor list was used to study morphological variation in the three most common
cultivated banana passion fruits and their hybrids, showing maternal effects in the
hybrid phenotypes and confirmed the spontaneous introgression occurring between the
wild P. mixta L. and the cultivated P. tripartita var. mollissima (Primot et al., 2005).
With a descriptor list similar to the one used by Villacís et al. (1998), Medina et al.
(2000) evaluated 25 accessions of the subgenera Passiflora, Manicata, Tacsonia,
Dysosmia, and Decaloba. The qualitative descriptors allowed a clear separation between
subgenera Passiflora, Tacsonia and Decaloba, with two groups in the latter. P. foetida
108
Chapter IV. Morphological diversity
______________________________________________________________________
L. (subg. Dysosmia) and P. manicata (Juss.) clustered with subgenera Passiflora and
Tacsonia respectively. The quantitative descriptors allowed distinction of subgenera
Tacsonia and Manicata, but showed no structure among species of the three other
subgenera. A very detailed descriptor list was also used by Porter-Utley (2003) to study
section (Killip) or supersection (Feuillet & MacDougal) Cieca of subgenus Decaloba,
and more particularly the species complex around P. suberosa L. and P. coriacea Juss.
Seventy quantitative traits were measured, 33 to 39 of which could be categorized and
gathered with qualitative traits for neighbor joining cluster analyses of the different
subsamples. Morphological data appeared consistent with molecular data in confirming
monophyly of section/supersection Cieca, recognizing P. tridactylites Hook. f. and
P. pallida L. as distinct species, and detecting occasional introgression of the latter with
its close relative P. suberosa. On the other hand, there was considerable incongruence
between molecular (ITS sequences) and morphological phylogenies, which was mostly
attributed to smaller sample and intraspecific variation in the molecular data. A
Brazilian Passiflora collection including ten species was characterized by Crochemore
et al. (2003) with 22 qualitative and quantitative descriptors. The results showed clear
differentiation between the two botanical forms of P. edulis. More recently, De Oliveira
et al. (2005) tested a new morphometric method, based on leaf structures in a sample of
ten Passiflora species. The method was very accurate in correctly differentiating among
species, but two species were not consistently classified, P. foetida, a problematic
species in all classifications, and P. miersii Mast. Although its potential must be further
assessed on wider samples, the method is very promising for species identification, as it
can be applied on sterile specimens. On the other hand, this advantage becomes a
limitation for any comparison with classical taxonomy works, where flower structures
play an essential role, so this method must be seen as a powerful additional tool. In
conclusion, despite the obvious interest of an objective classification in a plant family of
such morphological richness and complexity, only a few teams have developed and
applied the necessary methodology. Except for the studies targeting specifically
subgenus Tacsonia or section/supersection Cieca of subgenus Decaloba, their success
has been limited by the size and diversity of their species sample. This limitation is not
specific to morphological characterization. However, it tends to be much more severe
than in molecular characterization, because of the need of field germplasm collections
gathering species with variable climatic adaptations in one or very few places where
they can develop until flowering. In the case of wild Passiflora accessions, a possible
109
Chapter IV. Morphological diversity
______________________________________________________________________
solution is in situ characterization, which then must be taken into account to avoid or
reduce environmental bias in the analysis.
The present study benefited from projects on diversity of Colombian Passifloraceae,
including a component of collecting and establishing germplasm in field collections. As
Colombia is the country with the highest Passiflora species diversity, both for wild and
cultivated materials (Ocampo et al., 2007; Chapter II), a wide species sample could be
studied, although practical limitations allowed describing accessions of only 51 species
from eight subgenera, out of the 164 reported for the genus in the country. A few nonnative species were added, extending the sample to 60 species. We revised the
descriptor lists used by Medina et al. (2000) and Coppens d’Eeckenbrugge et al. (2002),
aiming at a better description of variation in subgenera other than Tacsonia (particularly
Distephana, Decaloba, Astrophea and Tryphostemmatoides). Our goal was twofold, to
test the utility of the revised set of descriptors over a wide range of Passiflora species,
and to study morphological divergence among subgenera, species and accessions.
IV.1.3. Materials and methods
IV.1.3.1. Plant materials
The morphological study was carried out in six experimental stations maintained by
several institutions in Colombia (Andes region) (University of Caldas, the coffee
grower federation research institute - Cenicafé, CIRAD/IPGRI, Passicol S.A). These
germplasm collections were situated in La Italia (Caldas; 1,100 m), Tesorito (Caldas;
2,400 m), Paraguacito (Quindío; 1,250 m), El Moral (Valle del Cauca; 2,200 m),
Tenerife (Valle del Cauca; 2,750 m) and El Tambo (Cauca; 1,800 m).
The germplasm sample consisted of 261 individuals, representing 124 accessions
and 60 species belonging to the genus Passiflora and nine Killip’s subgenera (Table 1
and Figure 2). The Colombian accessions were collected by Ocampo et al. (2007) in
2001-2004. Accessions from other countries were obtained from previously established
collections. Geographic distribution was taken into account in the selection of
accessions of a same species. Narrow endemics (e.g P. trinervia (Juss.) Poir.) are
110
Chapter IV. Morphological diversity
______________________________________________________________________
represented by one accession, and widespread species (e.g. P. edulis f. flavicarpa) by
one accession per region.
Three plants were grown from seeds for each accession, at a distance of 3 m between
rows and 3 to 5 m within rows, according to adult plant size. The taxonomic treatment
of Killip (1938), with emends by Escobar (1988a,b, 1989, 1994) and MacDougal (1994)
is followed in this text.
P. foetida L.
P. arborea Spreng.
P. coriacea Juss.
ASTROPHEA
DECALOBA
DYSOSMIA
P. vitifolia Kunth
P. maliformis L.
P. manicata (Juss.) Pers.
DISTEPHANA
PASSIFLORA
MANICATA
P. arbelaezii L. Uribe
TRYPHOSTEMMATOIDES
P. trinervia (Juss.) Poir.
PSILANTHUS
P. tarminiana Coppens & Barney
TACSONIA
Figure 2. Variation in shape and color among species from nine of Killip’s subgenera.
111
Chapter IV. Morphological diversity
______________________________________________________________________
IV.1.3.2. Data collection
The descriptor list (Annex 5) was developed, in various stages, mostly by experts of
Bioversity International (formerly IPGRI), CIRAD and CORPOICA. It was adapted to
take into account the wide diversity in our collections. It included 43 quantitative and 84
qualitative descriptors, presented synthetically in Table 2. They were assessed on three
individuals per accession, and five measures were taken for quantitative characters for
each individual. The color characters were recorded with the Royal Colour Chart (Royal
Horticultural Society, 2002). Quantitative fruit traits were not taken into account, as
they are too often submitted to convergent selection processes, both in the wild and in
cultivated species.
IV.1.3.3. Analyses of quantitative variation
Shape descriptors were computed as ratios of crude ones. Quantitative data were
submitted to an analysis of variance to compare variation among and within subgenera,
species, accessions, and individuals. To identify characteristics that mostly contributed
to the variation among subgenera, we selected traits for which more than half of the
variance was caused by variation at this level. When shape descriptors showed similar
discriminating power, they were preferred over crude descriptors, to avoid giving too
much importance to size components of variation. The selected descriptor set was
submitted to a principal component analysis (PCA), carried out with the varimax
normalized rotation option using the STATISTICA 6.0 software (Hill & Lewicki, 2006),
retaining those factors with an eigenvalue superior to one, and the individuals were
projected on the first three PCA axes.
112
Chapter IV. Morphological diversity
______________________________________________________________________
Table 1. List of accessions used in the present study. Taxonomy according to Killip (1938) and emends by Escobar (1988a,b, 1989) and MacDougal (1994).
Subgenus / section / serie / specie
Subgenus Astrophea (DC.) Masters, 1871
Section Euastrophea (Harms.) Killip
Passiflora arborea Spreng.
Passiflora emarginata Humb. & Bonpl.
Passiflora sphaerocarpa Triana & Planch.
Subgenus Decaloba(DC.) Rchb., 1828
Section Cieca (Medic.) Mast
Passiflora coriacea Juss.
Passiflora suberosa L.
Section Decaloba (DC.) Mast
Series Auriculatae
Passiflora auriculata Kunth
Series Lutae
Passiflora filipes Benth.
Series Miserae
Passiflora misera Kunth
Passiflora trifasciata Lemaire
Series Punctatae
Passiflora alnifolia Kunth
Code
No. ind.
Country
Locality
Status
arbCA
emaCA
emaVA
sphVA
2
2
2
2
Colombia
Colombia
Colombia
Colombia
Manizales (Caldas)
Manizales (Caldas)
Yotoco (Valle Del Cauca)
Cali (Valle Del Cauca)
Wild, edible fruit
Wild, edible fruit
Wild
Wild
corCA
corVA
corVL
cotTO
subAN
subCA
subCAL
subSA
subVA
2
2
2
2
2
2
2
2
3
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Palestina (Caldas)
Cartago (Valle Del Cauca)
Palmira (Valle Del Cauca)
Ibagué (Tolima)
Jerica (Antioquia)
Manizales (Caldas)
Manizales (Caldas)
Barichara (Santander)
Palmira (Valle Del Cauca)
Wild
Wild
Wild
Wild
Wild
Wild
Wild
Wild
Wild
aucCA
2
Colombia
Victoria (Caldas)
Wild
filRI
2
Colombia
Pereira (Risaralda)
Wild
misVA
triEC
2
2
Colombia
Ecuador
Jamundí (Valle Del Cauca)
Misahuallí (Napo)
Wild
Wild
alnCA
2
Colombia
Manizales (Caldas)
Wild
113
Chapter IV. Morphological diversity
______________________________________________________________________
Passiflora bogotensis Benth.
Passiflora biflora Lam.
Passiflora cuspidifolia Harms
Passiflora erythrophylla Mast.
Passiflora magdalenae Triana & Planch.
Series Sexflorae
Passiflora sexflora Juss.
Section Xerogona (Raf.) Killip
Passiflora capsularis L.
Passiflora rubra L.
Section Pseudodysosmia (Harms.) Killip
Passiflora adenopoda Moc. & Sessé ex D.C
Section Hahniopathanthus (Harms.) Killip
Passiflora guatemalensis S. Wats.
Subgenus Dysosmia (DC.) Killip, 1938
Passiflora foetida var. gossypiifolia (Desv.) Mast.
Passiflora foetida var. hispida (DC.) Killip ex Gleason
Subgenus Distephana (Juss.) Killip, 1938
Passiflora vitifolia Kunth
Subgenus Manicata (Harms) Escobar, 1988
Passiflora manicata (Juss.) Pers.
alnNA
alnSA
bogCU
bifCU
bifTO
cusBO
2
2
2
2
2
1
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Chachagui (Nariño)
Tona (Santander)
Bogotá (Cundinamarca)
Mariquita (Tolima)
La Mesa (Cundinamarca)
Duitama (Boyacá)
Wild
Wild
Wild
Wild
Wild
Wild
eryBO
magCO
2
2
Colombia
Colombia
Duitama (Boyacá)
Victoria (Caldas)
Wild
Wild
sexQU
2
Colombia
Calarca (Quindío)
Wild
capAN
capSA
capVA
rubCA
rubQU
2
2
2
2
2
Colombia
Colombia
Colombia
Colombia
Colombia
Jerico (Antioquia)
Barichara (Santander)
Cartago (Valle Del Cauca)
Manizales (Caldas)
Buenavista (Quindío)
Wild
Wild
Wild
Wild
Wild
adeQU
adeTO
2
2
Colombia
Colombia
Buenavista (Quindío)
Ibagué (Tolima)
Wild, edible fruit
Wild, edible fruit
guaCA
2
Colombia
Filadelfia (Caldas)
Wild
fotCH
fotTO
2
3
Colombia
Colombia
Quibdó (Chocó)
Armero (Tolima)
Wild, edible fruit
Wild, edible fruit
vitCA
vitTO
2
2
Colombia
Colombia
Victoria (Caldas)
Ibagué (Tolima)
Wild, edible fruit
Cultivated, edible fruit
manEC
manSA
manSA
1
2
2
Ecuador
Colombia
Colombia
Baños (Tungurahua)
Santander
Santander
Wild
Wild
Wild
114
Chapter IV. Morphological diversity
______________________________________________________________________
Subgenus Passiflora [= Granadilla (Medic.) Mast. 1871]
Series Quadrangulares
Passiflora alata Curtis
Passiflora quadrangularis L.
Series Digitatae
Passiflora serratodigitata L.
Series Tiliaefoliae
Passiflora maliformis L.
Passiflora ligularis Juss.
Passiflora serrulata Jacq.
Passiflora tiliifolia L.
Series Laurifoliae
Passiflora guazumaefolia Juss.
Passiflora nitida Kunth
Passiflora popenovii Killip
manSA
manVA
2
1
Colombia
Colombia
alaBR
alaVA
quaHU
quaVA
3
3
3
2
Brasil
Brasil
Colombia
Colombia
serrBR
2
Brasil
malAN
malCA
malVA
malQU
malHU
malVA
malVA
malTO
ligCA
ligEC
ligQU
ligRI
ligQU
serrMA
tilVA
2
3
2
2
2
3
1
2
3
1
3
1
2
2
2
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Colombia
Fredonia (Antioquia)
Palestina (Caldas)
Buga (Valle Del Cauca)
Calarca (Quindio)
Rivera (Huila)
Tulua (Valle Del Cauca)
La Unión (Valle Del Cauca)
Ibagué (Tolima)
Anserma (Caldas)
Cuenca (Azuay)
Salento (Quindio)
Santa Rosa (Risaralda)
Genova (Quindio)
Plato (Magdalena)
El Cerrito (Valle Del Cauca)
Cultivated, edible fruit
Wild, edible fruit
Wild, edible fruit
Wild, edible fruit
Cultivated, edible fruit
Wild, edible fruit
Wild, edible fruit
Wild, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Wild, edible fruit
Wild, edible fruit
Cultivated, edible fruit
Wild, edible fruit
Wild, edible fruit
guzMA
nitCH
popCA
popNA
2
1
2
2
Colombia
Colombia
Colombia
Colombia
Plato (Magdalena)
Quibdó (Chocó)
Timbío (Cauca)
Chachagui (Nariño)
Wild, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Series Serratifoliae
115
Santander
El Cerrito (Valle Del Cauca)
Wild
Wild
Paicol (Huila)
Palmira (Valle Del Cauca)
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Wild, edible fruit
Chapter IV. Morphological diversity
______________________________________________________________________
Passiflora bahiensis Klotzsch
Series Incarnatae
Passiflora cincinnata Mast.
Passiflora edulis f. edulis Sims
Passiflora edulis f. flavicarapa Degener
Passiflora incarnata L.
Series Kermesinae
Passiflora lehmanni Mast.
Passiflora smithii Killip
Series Lobatae
Passiflora caerulea L.
Passiflora subpeltata Ortega
Passiflora gibertii N.E. Brown
Subgenus Psilanthus (DC.) Killip, 1938
Passiflora trinervia (Juss.) Poir.
Subgenus Tacsonia (Juss.) Tr. & Planch, 1873
Section Colombiana
Series Leptomischae
Passiflora antioquiensis Karst.
Passiflora flexipes Triana & Planch.
Passiflora tenerifensis L.K. Escobar
bahBR
3
Brasil
cinBR
cinBS
edeCA
edeCAL
edeCC
edePE
edeQU
edf-BR
edfCA
edfHU
edfPE
edfPR
edfVA
incUS
3
3
3
2
2
1
1
4
3
3
3
3
3
1
Brasil
Brasil
Colombia
Colombia
Colombia
Peru
Colombia
Brasil
Colombia
Colombia
Peru
Peru
Colombia
U.S.A
lehQU
smiTO
2
2
caeFR
supAN
gibBR
Salvador (Bahia)
Cultivated, Edible fruit
Anserma (Caldas)
Miami (Florida)
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Colombia
Colombia
Buenavista (Quindío)
Ibagué (Tolima)
Wild
Wild
3
3
2
France
Colombia
Brasil
Montpellier (Herault)
Santa Marta (Magdalena)
Cultivated, edible fruit
Wild
Wild
triCO
2
Colombia
Salento (Quindío)
Wild
ant-AN
antCA
fleQU
tenVA
2
2
2
2
Colombia
Colombia
Colombia
Colombia
Sta. Rosa Osos (Antioquia)
Manizales (Caldas)
Salento (Quindío)
El Cerrito (Valle Del Cauca)
Wild, edible fruit
Wild, edible fruit
Wild, edible fruit
Wild, Edible fruit
116
Salamina (Caldas)
Manizales (Caldas)
Puracé (Cauca)
Genova (Quindío)
La Unión (Valle Del Cauca)
Rivera (Huila)
Chapter IV. Morphological diversity
______________________________________________________________________
Series Colombianae
Passiflora adulterina L.f.
Passiflora lanata (Juss.) Poir.
Series Quindiensae
Passiflora linearistipula L.K. Escobar
Section Bracteogama
Passiflora cumbalensis var. cumbalensis (H. Karst.) Harms
Passiflora luzmarina Jorgensen
Passiflora tarminiana Coppens & Barney
Passiflora tripartita var. mollissima Holm-Nielsen & Jørgensen
Passiflora tripartita var. tripartita Holm-Nielsen & Jørgensen
Section Parritana
Passiflora jardinensis L.K. Escobar
Passiflora parirtae (Mast.) L.H Bailey
Section Poggendorffia
Passiflora pinnatistipula Cav.
Passiflora x rosea (H.Karst.) Killip
Section Tacsonia
Passiflora mathewsii (Mast.) Killip
tenVA
1
Colombia
El Cerrito (Valle Del Cauca)
Wild, Edible fruit
aduBO
lanBO
2
2
Colombia
Colombia
Duitama (Boyacá)
Duitama (Boyacá)
Wild
Wild
linCA
2
Colombia
Manizales (Caldas)
Wild
cumEC
cumNA
luzEC
luzEC
tarAR
tarBO
tarCC
tarEC
tarPE
tarVA
tarVE
tvmCU
tvmEC
tvmNA
tvmPE
tvmVA
tvmVE
tvtEC
2
1
1
1
3
2
2
4
1
3
3
4
1
2
1
3
1
2
Ecuador
Colombia
Ecuador
Ecuador
Argentina
Colombia
Colombia
Ecuador
Peru
Colombia
Venezuela
Colombia
Ecuador
Colombia
Peru
Colombia
Venezuela
Ecuador
Tulcán (Carchi)
Pasto (Nariño)
Loja (Loja)
Loja (Loja)
Castellar
Boyacá
Silvia (Cauca)
Baños (Tungurahua)
El Cerrito (Valle Del Cauca)
Tachira
Ambato (Tungurahua)
Wild, edible fruit
Wild, edible fruit
Wild, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Cultivated, edible fruit
Wild, Edible fruit
jarAN
parTO
2
2
Colombia
Colombia
Jardin (Antioquia)
Herveo (Tolima)
Wild
Wild, edible fruit
pinBO
xroBO
1
1
Colombia
Ecuador
Boyacá
Tuta (Boyacá)
Wild, edible fruit
Wild, Edible fruit
matEC
1
Ecuador
Cuenca (Azuay)
Wild
117
El Cerrito (Valle Del Cauca)
Tachira
Cundinamarca
Baños (Tungurahua)
Pasto (Nariño)
Chapter IV. Morphological diversity
______________________________________________________________________
Passiflora mixta L.f.
Subgenus Trryphostemmatoides (Harms) Killip, 1938
Passiflora arbelaezii L. Uribe
Passiflora gracillima Killip
mixVA
mixVA
mixVA
mixVA
mixVA
2
3
3
1
1
Colombia
Colombia
Colombia
Colombia
Colombia
El Cerrito (Valle Del Cauca)
El Cerrito (Valle Del Cauca)
El Cerrito (Valle Del Cauca)
El Cerrito (Valle Del Cauca)
El Cerrito (Valle Del Cauca)
Wild, edible fruit
Wild, edible fruit
Wild, edible fruit
Wild, edible fruit
Wild, edible fruit
arbCH
graQU
1
1
Colombia
Colombia
Quibdó (Chocó)
Salento (Quindío)
Wild
Wild
118
Chapter IV. Morphological diversity
______________________________________________________________________
IV.1.3.4 Cluster analyses on qualitative data
According to the PCA results, certain quantitative variables were selected on the basis
of their contribution to the axes, categorized and added to the qualitative dataset,
provided that the corresponding information was not yet included in a purely qualitative
descriptor. The resulting dataset was treated in two steps. A first set of qualitative
variables was selected on the basis of their contribution to differentiation between
subgenera, discarding those that show frequent variation at lower levels. A second set
included all qualitative descriptors. Both sets were submitted to a neighbor joining
cluster analysis (Saitou & Nei, 1987), using the coefficient of dissimilarity of Sokal &
Michener and calculating bootstrap values from 100 replicates, with the DARwin 5.2
software (Perrier et al., 2003). This phenetic approach was preferred because of the
relatively poor information on morphological evolution in Passiflora.
Table 2. List of 127 descriptors used in the morphological characterization study.
Organ
Stem
Tendril
Stipule
Petiole
Leaf
Qualitative characters (84)
Habit
Shape
Pubescence
Anthocyanin
Presence
Shape
Pubescence
Anthocyanin
Presence
Permanence
Color
Pubescence
Shape
Margin
Anthocyanin
Anthocyanin
Pubescence
Color
Nectary shape
Nectary stipe
Heterophylly
Lobe number
Margin
Base shape
Apex shape
Presence of acumen
Code
STHA
STSH
STPU
STAN
TEPR
TESH
TEPU
TEAN
SPPR
SPPE
SPCO
SPPU
SPSH
SPMA
SPAN
PEAN
PEPU
PECO
PENS
PESN
LEPO
LELN
LEMA
LEBS
LEAS
LEPA
Pubescence – adaxial
Pubescente – abaxial
Anthocyanin – lamina
Anthocyanine – nerves
Color – adaxial
Heteroblasty
Presence of laminar nectaries
LEAX
LEPB
LEAL
LEAN
LECA
LEPH
LENL
119
Quantitative characters (43)
Diameters
Internode length
Code
STDI
STIN
Length
Width
Terminal arista length
SPLE
SPWI
SPTA
Length
Distance from base to first gland
Nectary number
PELE
PEDG
PENM
Margin serration density
Angle between lateral lobes
Central lobe length
Right lobe length
Central lobe width
Distance between leaf sinus and
petiole insertion
Nectary number on lamina
Nectary number on leaf margin
LEMS
LEAB
LELC
LERL
LECL
LESS
LELA
LENN
Chapter IV. Morphological diversity
______________________________________________________________________
Peduncle
Bract
Flower
Fruit
Distribution of laminar nectaries
Presence of marginal nectaries
Distribution of margin nectaries
Nectary shape
Pubescence
Anthocyanin
Bifurcation
LEDN
LEPN
LELM
LENS
PDPU
PDAN
PDBN
Union
Presence
Permanence
Pubescence
Position of pubescence
Color
Anthocyanin
Shape
Margin
Apex shape
Marginal nectaries
Corona type
Corolla type
Orientation
Pubescence on corolla
Hypanthium pubescence
Dominant petal color
Chlorophylla on exterior of sepal
BRUN
BRPR
BRPE
BRPU
BRPP
BRCO
BRAN
BRSH
BRMA
BRAS
BRNM
FLCY
FLCT
FLOR
FLPU
FLHP
FLCP
FLCS
Keel-shaped sepals
Sepal awn
Presence of petals
Union of sepals
Color of filaments at base
Color of filaments at apex
Distribution of anthers
Ovary pubescence
Color of ovary
Color of style
Color distribution on styles
Color of stigmas
Color of androgynophore
Color distribution on
androgynophore
Pubescence of androgynophore
Limen margin
Nectary chamber ring
Hypantium type
Internal color of hypanthium
Chlorophylla on exterior of
hypanthium
Anthocyanin on exterior of
hypanthium
Nectaries on sepals
Dominant sepal color
Anthocyanin on exterior of sepals
FLKS
FLSA
FLPP
FLUS
FLCB
FLCA
FLDA
FLOP
FLCO
FLCS
FLDS
FLCG
FLCN
FLDN
Type
Shape
FLPN
FLML
FLNR
FLHY
FLCI
FLCE
FLAE
FLNS
FLCP
FLAP
FRTY
FRSH
120
Length
Diameter
Pedicel length
Length to first bifurcation
Length to second bifurcation
Length
Width
PDLE
PDDI
PDPL
PDLF
PDBS
BRLE
BRWI
Orientation (in degrees to vertical)
Petal length
Petal width
Sepal length
Sepal width
Diameter of nectary chamber
Hypanthium diameter above
nectary chamber
Hypanthium diameter -distal
Flower length
Hypanthium length
Length of nectary chamber
Number of corona series
Filament length
Staminal filaments length
Ovary length
Style length
Gynophore length
FLOG
FLPL
FLPW
FLSL
FLSW
FLNC
FLHD
Androgynophore length
Operculum length
Limen length
FLAL
FLOP
FLLL
FLHS
FLLE
FLHL
FLCN
FLNS
FLFL
FLSF
FLOL
FLSL
FLGL
Chapter IV. Morphological diversity
______________________________________________________________________
IV.1.4. Results and Discussion
IV.1.4.1. Quantitative variation
As expected, a very high variability was observed among the 124 accessions in the field.
Table 3 gives the mean values and coefficients of variation for the whole sample and for
the different subgenera. Coefficients of variation appear generally superior for subgenus
Decaloba, as compared with subgenera Passiflora and Tacsonia that have comparable
representation in number of species. This higher relative variation is much more
obvious for inflorescences (18 out of 20 traits), and shape ratios (10 out of 13 ratios), of
higher taxonomical importance, than for vegetative parts (9 of 21 descriptors),
suggesting a higher interspecific differentiation in this subgenus.
Figure 3 shows the relative variance components for 57 quantitative descriptors. All the
descriptors present a residual variance under 25% and then a high repeatability. Many
descriptors appear to be efficient in discriminating among subgenera. Thus, the
proportion of variance at this level exceeds 50% for 26 of them, including stem
diameter, leaf margin indentation, leaf length, numbers of nectar glands on leaf margins
and petiole, diameter of peduncle, length of first and second order peduncle segments,
dimensions and shape of bracts, length of flower, hypanthium, sepals and petals, nectary
chamber, crown longest series, androgynophore, stamens and ovary, relative
constriction above nectary chamber, and bract/hypanthium length ratio. At the species
level, 28 characters are more important. They are related to dimensions of stipule,
lobation (angle between lateral nerves, shape of central lobe, length of lateral lobe,
distance between leaf sinus and petiole insertion), number of laminar nectary glands,
position of petiolar nectary glands, length of peduncle, diameter of hypanthium, length
of gynophore, shape of petals and sepals, androgynophore/hypanthium length ratio
(defining protrusion of gynoecium and androecium) and pedicel/peduncle ratio. At
lower levels, variance between accessions and plants rarely contributes more than 20%
of the total.
121
Chapter IV. Morphological diversity
______________________________________________________________________
Table 3. Mean values and coefficients of variation for the whole sample and for the different subgenera.
LELC
LERL
LECL
LESS
LELA
LENN
PELE
PEDG
PENM
BRLE
BRWI
PDLE
PDDI
PDPL
PDLF
PDBS
FLNS
FLOG
FLPL
FLPW
FLSL
FLSW
FLNC
FLHD
DHS
FLHS
FLLE
FLCN
FLFL
FLSF
FLOL
FLSL
FLGL
FLAL
FLOP
FLLL
PEDG/PELE
LECL/LELC
BRWI/BRLE
FLPW/FLPL
FLSW/FLSL
FLNC/FLHD
FLLE/FLHL
FLCN/FLHL
FLAL/FLHL
PDDI/PDPL
PDPL/PDLE
BRLE/FLHL
FLGL/FLAL
LECL/LECL
340
30.00
0.00
285.54
0.52
8.69
0.55
7.48
0.87
2.72
0.94
6.81
0.29
105.42
0.40
96.95
0.21
83.91
0.41
37.59
0.39
21.72
0.55
0.00
0.00
22.57
0.34
9.25
0.57
6.09
0.37
36.30
0.29
15.70
0.36
86.71
1.43
2.50
0.00
8.93
0.40
0.00
0.00
1.40
0.80
10.99
2.02
44.28
0.25
19.39
0.23
46.91
0.24
18.65
0.28
15.07
0.17
10.21
0.23
12.00
0.21
103.36
0.17
71.01
0.28
10.17
0.25
2.10
0.80
14.86
0.32
11.60
0.17
13.61
0.24
4.37
0.67
82.69
0.21
6.61
0.24
1.53
0.36
0.41
0.45
0.39
0.30
0.43
0.25
0.44
0.16
0.41
0.23
1.53
0.23
1.54
0.26
0.15
0.34
1.23
0.32
0.08
0.70
0.29
0.77
0.54
0.39
0.06
1.49
0.39
0.30
310
10.81
0.24
98.50
1.05
5.34
0.47
2.46
1.88
0.18
3.40
0.21
3.50
80.38
0.39
59.80
0.38
65.31
0.35
33.50
0.53
45.76
0.38
4.09
1.10
0.00
25.26
0.69
15.25
1.18
0.81
1.22
2.63
1.68
1.28
2.73
19.74
0.63
1.51
0.02
4.42
0.71
0.00
0.00
2.00
1.10
72.73
0.96
7.98
0.76
2.76
0.93
14.62
0.38
5.62
0.39
8.55
0.34
8.73
0.36
9.59
0.36
11.52
0.31
6.32
2.29
3.64
0.34
7.68
0.40
4.26
0.34
3.45
0.40
4.62
0.31
0.93
0.26
6.25
0.36
2.12
0.49
0.87
0.70
0.60
0.67
0.58
0.44
0.18
1.41
0.24
0.82
0.42
0.38
0.99
0.08
5.85
0.71
1.89
0.78
3.36
0.79
0.11
0.65
0.40
1.36
0.92
1.28
0.17
0.41
0.59
0.37
50
230.00
0.25
54.67
0.32
4.45
0.16
2.86
0.29
0.00
0.00
0.00
247.27
0.26
0.00
123.49
0.31
0.00
0.00
0.00
26.20
0.36
26.38
0.37
2.00
0.00
2.10
0.10
1.30
0.00
17.88
0.35
2.92
0.11
13.44
0.42
23.30
0.21
23.30
0.21
1.00
0.00
135.00
0.20
24.08
0.19
8.44
0.35
24.47
0.16
9.36
0.27
7.59
0.17
7.55
0.17
8.28
0.21
23.24
0.18
5.82
0.25
4.92
0.11
13.39
0.19
6.36
0.14
4.98
0.14
6.76
0.33
1.00
0.00
12.40
0.30
1.84
0.17
0.60
0.34
1.01
0.03
0.50
0.12
0.62
0.07
0.34
0.23
0.38
0.25
1.01
0.03
4.25
0.30
0.89
0.24
2.21
0.31
0.18
0.25
0.76
0.43
0.38
0.21
0.09
0.39
0.76
0.26
35
30.00
0.00
321.63
0.31
11.65
0.12
18.64
0.12
3.87
0.19
8.63
0.06
116.74
0.07
112.82
0.24
117.99
0.27
52.39
0.23
41.35
0.22
0.00
0.00
28.63
0.21
7.89
0.36
9.20
0.29
42.97
0.10
26.04
0.09
65.70
0.25
2.50
0.00
13.70
0.12
0.00
0.00
5.00
0.00
158.00
0.14
42.27
0.08
14.65
0.09
44.36
0.08
14.17
0.13
15.85
0.09
10.93
0.07
14.64
0.11
42.44
0.07
13.03
0.24
8.58
0.08
5.82
0.23
12.19
0.12
8.20
0.10
12.14
0.12
2.70
0.19
29.11
0.09
10.37
0.08
1.25
0.16
0.28
0.32
0.47
0.14
0.61
0.07
0.35
0.10
0.32
0.16
1.46
0.10
3.42
0.21
0.69
0.22
2.35
0.24
0.04
0.27
0.22
0.27
3.52
0.28
0.09
0.23
0.50
0.12
25
15.00
0.00
40.04
0.14
2.70
0.26
5.35
0.10
6.22
0.04
5.00
0.00
92.24
0.03
73.22
0.04
45.47
0.04
42.16
0.05
41.05
0.08
0.00
0.00
21.92
0.16
0.00
0.00
20.92
0.11
16.20
0.27
49.39
0.16
1.50
0.00
2.60
0.00
0.00
0.00
4.20
0.84
142.00
0.23
18.90
0.14
7.92
0.13
19.47
0.11
7.22
0.12
8.32
0.05
8.40
0.05
8.40
0.05
12.88
0.08
8.35
0.10
2.48
0.06
11.20
0.33
4.60
0.13
3.71
0.20
6.44
0.09
1.00
0.02
8.46
0.11
2.32
0.07
1.40
0.07
0.00
0.58
0.04
0.78
0.28
0.42
0.08
0.37
0.09
0.99
0.03
1.56
0.17
0.30
0.10
1.03
0.19
0.03
0.16
0.05
0.16
2.55
0.22
0.12
0.12
0.58
0.04
122
20
35.00
0.00
61.30
0.14
6.20
0.17
1.00
0.00
0.00
6.10
0.05
77.90
0.10
115.10
0.09
96.30
0.13
49.30
0.11
44.00
0.02
0.00
7.00
0.15
30.85
0.18
0.50
0.00
2.00
0.00
26.70
0.19
8.70
0.18
46.85
0.13
2.50
0.00
4.92
0.07
0.00
0.00
2.00
0.00
110.00
0.42
56.25
0.06
13.58
0.20
59.60
0.05
16.60
0.07
13.90
0.13
11.90
0.15
12.75
0.06
55.75
0.07
12.07
0.09
6.10
0.26
24.10
0.05
14.20
0.12
10.76
0.06
13.20
0.06
2.72
0.09
42.08
0.08
10.00
0.06
3.55
0.14
0.02
0.22
0.43
0.11
0.34
0.32
0.24
0.14
0.28
0.11
1.17
0.05
4.64
0.08
0.50
0.22
3.50
0.08
0.05
0.13
0.11
0.15
2.25
0.27
0.06
0.06
0.47
0.14
10
5.00
0.00
22.91
0.10
0.46
0.61
0.25
0.21
0.00
0.00
0.00
32.11
0.38
0.00
22.38
0.15
0.00
0.00
0.00
19.09
0.14
19.09
0.14
2.00
0.00
1.00
0.00
0.50
0.00
23.86
0.10
0.80
0.00
8.24
0.05
26.50
0.06
16.50
0.10
2.00
0.00
0.00
13.54
0.19
3.56
0.27
14.51
0.29
5.51
0.34
7.10
0.14
6.81
0.14
6.81
0.14
8.30
0.11
4.96
0.16
3.80
0.18
5.83
0.11
2.46
0.15
2.85
0.07
3.63
0.07
0.51
0.06
4.86
0.04
0.84
0.06
0.65
0.08
1.00
0.00
0.76
0.26
0.50
0.00
0.26
0.16
0.38
0.17
1.05
0.13
1.70
0.16
0.77
0.12
1.00
0.12
0.03
0.11
0.35
0.10
0.21
0.15
0.11
0.08
0.16
0.08
10
9.60
0.05
50.80
0.04
8.70
0.08
1.00
0.00
0.00
0.00
33.10
0.08
116.50
0.01
92.70
0.03
18.70
0.08
89.40
0.01
6.00
0.00
0.00
24.70
0.05
0.00
0.00
8.60
0.10
2.70
0.18
99.30
0.01
1.11
0.05
3.30
0.15
0.00
0.00
1.00
0.00
172.50
0.03
11.20
0.08
4.90
0.06
39.10
0.02
7.00
0.13
10.00
0.08
8.70
0.06
7.50
0.07
162.00
0.02
112.90
0.02
29.30
0.07
2.90
0.25
8.30
0.06
11.90
0.06
5.10
0.14
9.20
0.11
150.50
0.01
4.20
0.10
0.00
0.00
0.16
0.08
0.32
0.24
0.44
0.11
0.18
0.13
1.16
0.14
1.44
0.03
0.26
0.08
1.33
0.02
0.01
0.05
0.03
0.15
0.08
0.10
0.06
0.12
0.43
0.11
1304
37.96
1.11
235.09
0.76
9.82
0.83
5.59
1.22
1.03
1.96
3.89
0.92
75.40
0.75
103.05
0.46
63.42
0.72
54.96
0.69
25.74
0.83
1.02
2.79
0.11
8.10
31.99
0.66
16.65
0.98
3.27
0.85
25.31
0.75
13.93
0.81
49.56
1.41
2.30
0.24
7.81
0.54
1.10
4.64
1.02
4.68
3.10
0.91
73.71
0.94
30.28
0.55
10.86
0.66
33.51
0.46
13.11
0.50
12.33
0.32
10.82
0.34
13.30
0.39
44.56
0.88
26.47
1.16
6.48
0.57
17.37
1.02
10.13
0.50
7.81
0.47
9.92
0.43
2.34
0.94
30.69
1.14
3.88
0.67
1.38
0.65
0.52
0.64
0.52
0.42
0.47
0.53
0.33
0.42
0.40
0.28
1.17
0.26
3.07
0.87
0.74
1.32
1.75
0.95
0.09
0.75
0.30
1.10
1.67
0.92
0.14
0.80
0.52
0.42
Organ
Stem
504
44.01
0.15
321.40
0.55
14.45
0.73
5.98
1.24
0.13
5.77
4.26
0.81
58.09
1.17
120.92
0.26
51.21
1.06
75.49
0.56
15.37
1.16
0.00
0.00
44.24
0.58
23.91
0.75
2.90
0.58
33.91
0.50
21.50
0.47
44.50
0.45
2.66
0.12
8.63
0.41
0.00
0.00
5.00
0.00
114.94
0.45
34.58
0.28
10.36
0.43
36.15
0.26
14.71
0.34
13.32
0.26
13.04
0.30
17.39
0.29
27.00
0.23
11.97
0.31
5.56
0.32
35.39
0.44
11.08
0.25
8.30
0.30
11.11
0.25
1.91
0.58
11.36
0.29
2.75
0.46
1.65
0.64
0.56
0.53
0.59
0.37
0.64
0.22
0.30
0.24
0.41
0.24
1.04
0.13
2.33
0.21
0.47
0.22
1.01
0.35
0.08
0.77
0.23
0.63
2.89
0.47
0.18
0.68
0.58
0.44
Leaf
LEMS
LEAB
Passiflora Tacsonia Decaloba Astrophea Manicata Dysosmia Distephana Thryphostemmatoides Psilanthus Total sample
Bract
SPWI
SPTA
n
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Mean
CV
Flower
STIN
SPLE
Level
Shape relation
Factor
Descriptors
STEDI
Chapter IV. Morphological diversity
______________________________________________________________________
Figure 3. Relative variance components for 57 quantitative descriptors. Bold characters are used for traits
displaying more than 50% of variance among subgenera.
IV.1.4.2. Correlations and principal components analysis (PCA)
A Pearson correlation (r) matrix shows high coefficients, between 0.71 and 0.97, among
dimensions of certain floral components, such as petals and sepals, hypanthium, nectary
chamber, operculum, androgynophore, stamens and ovary. The length of the corona is
correlated with the upper hypanthium diameter (r = 0.75). Bract dimensions are
moderately correlated with these traits (0.33 to 0.65, but 0.75 with ovary length). Stem
diameter is correlated with peduncle branching (r = 0.73), which is due to the
association of these traits in representatives of subgenus Astrophea.
From the 26 quantitative descriptors showing high variation at the subgenus level, 24
were selected for the PCA, discarding two of them to avoid redundancy between shape
ratios and the original traits. Five principal components were retained, that represent
84% of total variation (Table 4). The first one (32%) is very clearly associated with
flower length (hypanthium, nectary chamber, androgynophore) and secondarily with the
123
Chapter IV. Morphological diversity
______________________________________________________________________
constriction of the floral cup above the nectary chamber. The second one (27%) is
associated with flower width (length of bracts and length of corolla and corona elements)
and bract shape. The third one (14%) is associated with peduncle branching, stem width
and leaf length, which relates it clearly with variation between subgenus Astrophea, and
secondarily subgenus Tryphostemmatoides, and all other subgenera. The fourth one (5%)
is associated with the number of nectaries on leaf margin, which essentially relates it to
the differentiation of subgenus Distephana. The last one (5%) is correlated only with
leaf serration.
Figure 4 presents the accessions in the three first axes, showing a clear grouping by
subgenus. The representatives of subgenus Tacsonia are placed on the right along the
first axis, in relation to their long and wide flowers. A few accessions, with shorter
flowers (P. luzmarina Jørgensen and P. pinnatistipula Cav.), appear closer to the origin
of this axis, together with P. manicata individuals. P. trinervia, of subgenus Psilanthus,
is placed even further on the right, thanks to its very long floral tube, however it is
clearly separated on the second axis by its much narrower flowers and minute setaceous
bracts. On the left side, subgenera Passiflora and Decaloba are not differentiated by the
flower length axis, but by the second, flower width related, axis. At the extremes of this
second axis, we find the accessions of the large-flowered P. alata Curtis and P.
quadrangularis (section Quadrangulares of subgenus Passiflora), on one side, and
those of the small-flowered P. arbelaezii and P. gracillima Killip of subgenus
Tryphostemmatoides, on the other side. As expected, the third axis clearly differentiates
subgenera Astrophea and Tryphostemmatoides. On the whole, subgenera Passiflora,
Tacsonia, Decaloba, Psilanthus, Astrophea and Tryphostemmatoides are clearly
separated in the main tridimensional space. As expected, P. manicata (subgenus
Manicata) takes an intermediate position between subgenera Passiflora and Tacsonia.
This species not only combines morphological traits typical of both subgenera, but
intermediate ecoclimatic requirements as well, as it may be found at lower elevations
than tacsos, but higher elevations than representatives of subgenus Passiflora. The
representatives of subgenus Tacsonia that come closest to P. manicata are
P. pinnatistipula and P. luzmarina, two tacsos with relatively shorter floral tubes. The
former is also differentiated by a filamentous corona, instead of the typical reduced
tacso coronas. Another species taking a particular position is P. foetida, of Killip’s
subgenus Dysosmia, placed near both subgenera Passiflora and Decaloba, but closer to
124
Chapter IV. Morphological diversity
______________________________________________________________________
the former. Subgenus Dysosmia is considered a supersection of subgenus Passiflora in
the Feuillet & MacDougal classification.
Table 4. Factor loadings from principal component analysis (varimax normalized rotation) on 24
quantitative descriptors.
Principal components
Descriptors
STDI
LEMS
LELC
LENN
PENM
PDDI
PDLF
PDBS
BRLR
FLPL
FLPW
FLSL
FLSW
FLLE
FLHL
FLCN
FLFL
FLSF
FLOL
FLSL
FLAL
FLOP
BRWI/BRLE
FLNC/FLHD
Expl.Var
Prp.Totl
1
2
3
4
5
-0.088
0.353
0.019
-0.017
0.648
0.135
-0.054
-0.054
0.296
0.501
0.713
0.577
0.520
0.960
0.904
0.824
-0.521
0.529
0.677
0.454
0.964
0.711
-0.160
0.700
0.215
0.365
0.571
0.040
0.407
0.847
-0.136
-0.113
0.723
0.774
0.595
0.716
0.754
0.179
0.063
0.208
0.723
0.651
0.658
0.693
0.064
0.236
0.718
0.056
-0.892
0.152
-0.686
0.011
0.013
-0.287
-0.943
-0.973
0.186
-0.021
-0.024
0.036
0.034
0.045
0.086
0.026
0.034
0.082
0.076
0.067
0.047
0.099
-0.221
0.042
-0.023
0.041
0.009
0.962
-0.055
0.010
-0.007
-0.006
-0.091
0.235
0.095
0.243
0.074
0.017
-0.100
-0.080
-0.007
0.097
0.061
0.073
0.033
0.363
-0.131
0.043
0.041
0.759
-0.195
0.014
0.158
0.145
-0.044
-0.046
0.240
0.042
0.028
-0.028
-0.108
0.090
0.119
-0.137
-0.329
0.171
-0.049
0.263
0.099
0.214
0.241
0.368
7.610
0.317
6.496
0.271
3.336
0.139
1.256
0.052
1.238
0.052
125
Chapter IV. Morphological diversity
______________________________________________________________________
Figure 4. Tridimensional plot of the scores of Passiflora accessions for the first three components of
quantitative variation. Colors refer to subgeneric classification.
IV.1.4.3. Qualitative variation among and within subgenera
Our first attempt to reduce the number of qualitative descriptors led us to retain 32 of
them on the basis of their potential to discriminate among subgenera. The criterion was
that the descriptor appears monomorphic or shows a highly dominant condition at least
in one subgenus, while polymorphic among subgenera. Four quantitative descriptors
were categorized and added because of their high correlations with the principal
components of quantitative variation. Thus the first component was represented by
androgynophore length, the second one by sepal length, the third one by stem diameter
and leaf length. The fourth and fifth ones were not included to avoid redundancy with
very similar qualitative descriptors. Table 5 synthesizes the observations for these
descriptors. The species of subgenus Astrophea exhibit the highest number of
unique/rare traits, including tree habit, wide stems of irregular section, very long leaves,
absence of tendrils, short triangular stipules, dorsal scar-like nectaries (appressed
against or near petiole), branched peduncles, bright-yellow sickle-sword-shaped corona
filaments, and tricostate ovaries. Unique and rare traits of the two species of subgenus
126
Chapter IV. Morphological diversity
______________________________________________________________________
Tryphostemmatoides include peduncle branching (shared with species of subgenus
Astrophea and P. sexflora Juss. of subgenus Decaloba), the presence of tendrils at the
axil of the peduncles, and the retuse leaf apex (unique in our sample, although this trait
can be observed in individuals of P. emarginata Humb. & Bonpl.). P. vitifolia Kunth
(subgenus Distephana) is differentiated by its tubular corona, formed by the partial
fusion of its elements and conspicuous nectary glands on leaf sinus and bracts,
P. foetida (subgenus Dysosmia) only by its pinnatisect bracts, and P. trinervia
(subgenus Psilanthus) by the absence of a limen. Most of these traits are typical for each
of these subgenera, ensuring that they will not bias the cluster analysis in terms of
subgeneric classification. Subgenera Decaloba, Passiflora, Tacsonia and Manicata do
not show unique traits, however they can be separated by clear segregations in nonexclusive traits. Thus, in subgenus Decaloba, the presence of nectary glands in the
lamina is only shared with subgenus Psilanthus, the flat hypanthium with subgenus
Tryphostemmatoides, the relatively small flower size with subgenera Astrophea,
Dysosmia, and Tryphostemmatoides. On the other side, subgenera Passiflora,
Distephana, Manicata and Tacsonia share wide flowers and the general presence of
petiolar nectaries. The last three and subgenus Psilanthus also share large red or pink
corollas and long tubular flowers (long androgynophores), typical of hummingbirdpollinated species. Floral tube length reaches extreme values in Psilanthus and Tacsonia,
with the exceptions of P. pinnatistipula and P. luzmarina, in relation to their specific
pollinator Ensifera ensifera Boissoneau, the sword-billed hummingbird. In addition they
present reduced coronas of short filaments or tubercles, generally in one row only, while
two-row coronas are most common in subgenera Decaloba, Tryphostemmatoides,
Astrophea, and Distephana, and highly complex coronas (more than three rows) are
typical in subgenera Passiflora, Manicata and Dysosmia. Bracts are foliaceous in
subgenera Passiflora and Tacsonia. Fruit shape is generally globose to short ovate in
subgenera Astrophea, Decaloba, Tryphostemmatoides, Dysosmia, Passiflora and
Distephana, and oval to fusiform in subgenera Tacsonia, Manicata, and Psilanthus.
Fruit color would also be interesting, with a particular frequency of blackish fruits in
subgenus Decaloba, however this descriptor could not be observed in all species.
127
Chapter IV. Morphological diversity
______________________________________________________________________
Table 5. Variation for 32 qualitative and four categorized quantitative descriptors in the different subgenera sampled.
Astrophea
Feuillet & MacDougal (2003)
Killip (1938), Escobar (1988), MacDougal (1994)
Species/accession number
Main pollinators
Decaloba
Astrophea
n= 3/4
medium (honey) bees
Decaloba
n= 17/31
small to large bees - wasps
2n = 24
1.000 - 2.200 m
Chromosome number
Altitudinal range
Deidamioides
Passiflora
Passiflora
n= 19/44
large (carpenter) bees
2n = 12, 22, 24, 36
Psilanthus
n= 1/1
sword-billed
hummingbird
2n = 12
Distephana
n= 1/2
hummingbirds
Dysosmia
n= 1/2
bees
60 - 2.700 m
2.600 - 3.200
Manicata
n= 1/1
hummingbirds
2n = 18, 20, 22
Tacsonia
n= 16/36
sword-billed
hummingbird
2n = 18
2n = 18
2n = 18
20 - 2.400 m
50 - 1.200 m
Tryphostemmatoides
n= 2/2
bees
2n = 18
2n = 12
30 - 1.200 m
2.100 - 3.700 m
1.900 - 2.500 m
50 - 2.000 m
Descriptors
Habit
Stem section
Secondery xylema
Tendril position
Stipule
Stipule nectaries (conspicuous)
Leaf lobation
Leaf base
tree
irregular
present
absent
short triangular
absent
one
cuneate-rounded
Leaf apex
Leaf margin
Laminar nectaries
Leaf margin nectaries (conspicuous)
obtuse/acute
entire
absent
absent
Petiolar nectaries
absent
Dorsal nectaries
Peduncle branching
Bract shape
present
present
linear
Nectary on bract
Flower orientation
Corolla shape
Dominant corolla color
Corona type
Corona filaments
Number of corona series
absent
erect
reflex
white
filamentous
sickle-sword-shaped
free
uniseriate
Corona color-clear (longest row)
Corona color-darkest (longest row)
white
yellow
Petals
present
Sepal awn
Nectar chamber ring
Hypanthium
Limen
Ovary shape
Fruit shape
absent
present
campanulate
present
tricostate
globose
Categorized (from quantitatives traits)
Stem diameter
Leaf length
vine
terete/angular
absent
axillary
setaceous/linear/foliaceous-aristate
absent
three
cuneate-rounded/cordate/
peltate (P. coriacea and P. guatemalensis )
rounded/obtuse/acute
entire/serrate (P. adenopoda )
present/absent (five species)b
absent/sinus/
all margin (P. adenopoda )
absent/orbicular (P. adenopoda )/
cylindrical (P. coriaceae and P. suberosa )
/auriculate (P. auriculata )
absent
absent/present (P. sexflora )
setaceous/linear/
foliaceous (P. adenopoda , P. guatemalensis )
/absent (four species)*
absent
erect/intermediate/pendular
intermediate/campanulate
white
filamentous
linear
free
biseriate/
uniseriate (P. guatemalensis and P. adenopoda )/
triseriate (P. filipes and P. magadalenae )
white/purple/
yellow (P. guatemalensis )
present/
absent (P. coriaceae and P. suberosa )
absent/present
absent
flat
present
globose
globose/
elongate (P. rubra and P. capsularis )
≥ 120 mm
Androgynophore length
b
≤ 20 mm
< 23 mm
vine
terete
absent
axillary
linear
present
three
cordate
vine
terete
absent
axillary
foliaceous-aristate
present
three
cordate
vine
terete/angular
absent
axillary
setaceous/linear/foliaceous-aristate
absent
one/three
cuneate-rounded/cordate
vine
angular
absent
axillary
foliaceous-aristate
absent
three
cordate
vine
terete
absent
axillary and peduncle
setaceous
absent
one
rounded
acute/very acute
entire
present
absent
acute
serrate
absent
sinus
acute
serrate
absent
absent
acute/very acute
serrate
absent
absent/sinus (P. jardinensis )
obtuse to very acute
serrate
absent
absent
retuse
entire
absent
leaf base
absent
rounded to very acute
entire/serrate
absent
absent/sinus
lateral lobe (P. edulis f. edulis )
all except auriculate
orbicular
absent
linear/orbicular
linear
absent
absent
absent
linear
absent
absent
foliaceous
absent
absent
linear
absent
absent
pinnatisect
absent
absent
foliaceous
absent
absent
foliaceous
absent
present
linear
absent
pendular
campanulate
red (pink)
filamentous
linear
free
uniseriate
absent/present
erect/intermediate/pendular
intermediate/reflex
white/red
filamentous
linear
free
pentaseriate
present
erect
reflex
red
filamentous
linear
fused at base
biseriate
absent
erect
intermediate/reflex
white
filamentous
linear
free
3- to 5-seriate
absent
erect/intermediate/pendular
campanulate/intermediate/reflex
red (including pink or orange)
tuberculous/
filamentous, linear
free
uni- to 5-seriate
absent
erect
reflex
red
filamentous
linear
free
pentaseriate
absent
erect
reflex
white
filamentous
linear
free
biseriate
white
white
white
purple
white (P. guazumaefolia )
present
red
red
white
purple
purple
purple
white
white
present
present
white/purple
purple
white (P. jardinensis )
present
present
present
absent
absent
tubular
absent
globose
elongate
present
present
campalunate
present
globose
globose/
elongate (P. quadrangularis )
present
present
tubular
present
globose
globose
present
present
campanulate
present
globose
globose
present
present
tubular
present
globose
elongate/
globose (P. pinnatistipula )
present
present
tubular
present
globose
elongate
absent
absent
flat
present
globose
globose
≤ 20 mm
≥ 30 mm
< 60 mm
< 190 mm
≥ 30 mm
< 60 mm
< 190 mm
≤ 20 mm
≥ 30 mm
< 60 mm
< 190 mm
≥ 30 mm
< 60 mm
< 190 mm
≤ 20 mm
> 30 mm
≤ 30 mm (six species)ª
< 23 mm
> 30 mm
≤ 30 mm
> 30 mm
≤ 30 mm
> 25 mm
< 23 mm
> 30 mm
≤ 30 mm (P. luzmarina and P. mathewsii )
> 25 mm
< 190 mm
> 30 mm
< 23 mm
b
vine
terete/angular/winged
absent
axillary
setaceous/linear/foliaceous-aristate
absent/present (P. maliformis )
one/three/more
cuneate-rounded/cordate
present
> 215 mm
< 190 mm
< 190 mm (P. sphaerocarpa )
≤ 30 mm
≤ 30 mm
Sepal length
vine
angular
absent
axillary
setaceous
absent
three
cordate
> 25 mm
b
b
b
a
< 190 mm
a
> 25 mm
a
< 23 mm
P. capsularis *, P. coriacea*, P. rubra *, P. suberosa *, P. sexflora , P. guatemelensis , P. bahiensis , P. edulis f. edulis , P. guazumaefolia , P. smithii , P. subpeltata , P. serrulataa.
128
a
< 190 mm
a
Chapter IV. Morphological diversity
______________________________________________________________________
Certain species show unusual trait combinations in their subgenus. This is particularly
true in subgenus Decaloba. Thus, P. adenopoda Moc. & Sessé ex DC. shows foliaceous
bracts, serrate leaf margins, with conspicuous nectary glands, orbicular petiolar glands,
and an uniseriate corona. P. guatemalensis S. Watson also shows foliaceous bracts and
glandless leaves, plus peltate leaves and a yellow uniseriate corona. P. sexflora shows
multiple peduncles. P. coriacea, P. suberosa, P. capsularis L. and P. rubra L. lack
bracts; in addition, the first two present petiolar nectaries, while the last two lack such
glands in all their parts and produce an elongate fruit.
IV.1.4.4. Cluster analysis on the reduced descriptor list
Figure 5 presents the dendrogram obtained from these observations on the first set of
descriptors. The four best-represented subgenera, Passiflora, Tacsonia, Astrophea, and
Decaloba, are supported by the analysis. Their placement on the dendrogram shows a
polarization of the latter according to several traits. On one side, we find subgenera
Passiflora, Distephana, Tacsonia, and Manicata, i.e. species producing large flowers
and fruits and very generally bearing petiolar nectaries, with a base chromosome
number of n = 9. They also share foliaceous bracts, with the relative exception of
P. vitifolia, whose long bracts appear linear. They are further divided between the
carpenter bee-pollinated species (subgenus Passiflora) and the hummingbird-pollinated
species of subgenus Tacsonia, P. vitifolia, and P. manicata. The consistency of this
subclassification compensates for the low associated bootstrap values. The placement of
P. manicata in the Tacsonia cluster supports the gathering of the Andean subgenera
Tacsonia and Manicata in a same infrageneric taxon, as in the classification of Feuillet
& MacDougal (2003). On the other hand, P. vitifolia is a good representative of the
uniform subgenus Distephana, so its position does not support its downgrading to a
supersection of subgenus Passiflora, also proposed by these authors. On the other side
of the tree, we find subgenera whose species produce small to medium flowers and
fruits, with relatively simple coronas of generally two rows of filaments (rarely one or
three), where petiolar nectaries are rare, with a base chromosome number of 12 for tree
species and 6 for the others. As could be expected from the number of their rare traits,
the subgenus Astrophea species of our sample appear very uniform, and well separated
in a very distant cluster. P. trinervia (subgenus Psilanthus) is placed on another long
branch, inserted at the same position. A third, much larger, cluster is constituted by all
the species of subgenera Tryphostemmatoides and Decaloba, but P. adenopoda. This
129
Chapter IV. Morphological diversity
______________________________________________________________________
species is placed on a well-separated branch, inserted in an intermediate position
between the Passiflora-Distephana-Tacsonia-Manicata clusters and the AstropheaDecaloba-Tryphostemmatoides
clusters.
Although
the
branch
bearing
the
representatives of subgenus Tryphostemmatoides, P. gracillima and P. arbelaezii, is
relatively long, it is clearly inserted within subgenus Decaloba, suggesting that the
qualitative morphological differentiation of subgenus Tryphostemmatoides is fragile,
which is consistent with the very low number of traits supporting it (Table 5), but
contrasting with the PCA results on quantitative traits. The position of P. adenopoda
may look surprising, as it is not consistent with either classification (Decaloba section
Pseudosysosmia for Killip, supersection Bryonioides for Feuillet & MacDougal),
however, this species showed several unusual features as compared to Decaloba as a
group. P. foetida (subgenus Dysosmia), takes a very similar position, that is clearly
more consistent with its classification in subgenus Dysosmia by Killip (1938) than with
its inclusion in subgenus Passiflora by Feuillet & MacDougal (2003). Interestingly,
these two problematic species materialize the separation between the two cytogenetic
groups in our tree. Indeed, chromosome counts for P. adenopoda give 2n = 12
(MacDougal, 1994), as in most species of subgenus Decaloba, while those for P. foetida
vary between 2n = 18, 20, and 22 (Yockteng & Nadot, 2004; De Melo et al., 2001).
According to de Melo et al. (2001) and De Melo & Guerra (2003), P. foetida appears
cytologically quite isolated, but closer to the n = 9 group, its smaller chromosomes and
areticulate interphase nuclei being similar to species with n = 6, while its chromosome
number, higher karyotype symmetry, CMA staining properties, and the number of 45S
rDNA sites make it similar to species of subgenus Passiflora.
The first set of qualitative data also allows distinguishing some structure within clusters
corresponding to subgenera. Thus, within the Passiflora cluster, one main branch
corresponds to P. edulis f. edulis and medium-flowered species as P. bahiensis Kl. and
P. guazumaefolia Juss., one to large-flowered species of series Incarnatae, i.e. P. edulis
f. flavicarpa, P. incarnata, and P. cincinnata Mast., one to series Quadrangulares and
the most typical representatives of series Tiliifoliae, one to typical representatives of
series Laurifoliae, one to accessions of P. maliformis, and one to species of series
Kermesinae. Series Lobatae is shared between the Incarnatae subcluster (including
P. caerulea L. and P. gibertii Brown) and the Kermesinae subcluster (P. subpeltata
Ortega), which shows the fragility of these subclusters, as P. gibertii and P. subpeltata
130
Chapter IV. Morphological diversity
______________________________________________________________________
are morphologically very similar. Within the Tacsonia-Manicata cluster, there are three
main branches. A first one includes the common and widely dispersed P. mixta,
P. tripartita, P. tarminiana, P. cumbalensis and endemic species related to one of them,
as P. mathewsii (Mast.) Killip and P. luzmarina. A second one includes species of
Section Colombiana, plus two accessions of P. tripartita var. mollissima, and the last
one includes the relatively short-tubed species P. pinnatistipula and P. manicata. Within
the Decaloba-Tryphostemmatoides cluster, one branch corresponds to section Cieca
(P. coriacea and P. suberosa) and one to section Xerogona (P. capsularis and P. rubra).
P. guatemalensis, the only representative of section Hahniopathantus, is placed apart.
Section Decaloba is split between three distinct branches, one for the closely related
P. alnifolia Kunth and P. bogotensis Benth., one for P. auriculata Kunth, and one for all
its other representatives.
131
Chapter IV. Morphological diversity
______________________________________________________________________
n = 12
n = 10
n=6
n=9
n = ??
Figure 5. Dendrogram obtained with first set of qualitative data. Distances of Sokal & Michener.
132
Chapter IV. Morphological diversity
______________________________________________________________________
IV.1.4.5. Cluster analysis on the global descriptor dataset
In a second analysis, all qualitative descriptors (74) were incorporated in the cluster
analysis, except those related to pubescence and anthocyanins because of a likely higher
sensitivity to environmental conditions, and the difficulty to control redundancy of
information (e.g. pubescence of different organs is often, but not always, correlated).
Removing these descriptors did not affect the structure of the dendrogram, neither at the
subgeneric nor at the interspecific level, however it increased distances and improved
bootstrap values. The general structure of the tree based on most descriptors (Figure 6)
is highly similar to the one obtained with the reduced descriptor set. Branches support is
improved, as bootstrap values increase to more than 50% between subgenera, indicating
that descriptors showing high polymorphism within subgenera, and sometimes even at
the intraspecific level, still contribute to differentiation among subgenera. Distances and
bootstrap values between species, and even accessions of a same species, are much
higher, reflecting high polymorphism at the lowest levels. However, this does not affect
the tree general consistency, as subdivisions among species can be even more easily
interpreted. Below the species level, individuals are most often grouped by accessions,
but these accessions do not cluster according to their geographic origin.
IV.1.4.6. The “Passiflora cluster”
Within subgenus Passiflora, all species are consistently separated from each other.
Several subclusters are constituted by species of obvious morphological affinity, but
these fail to reflect Killip’s series. Every time a series is represented by three species,
one of them diverges. A first example is the subcluster associating the two forms of
P. edulis and P. incarnata (series Incarnatae), but not P. cincinnata. A second one is
the P. popenovii and P. nitida subcluster (series Laurifoliae), which does not include the
less typical representative P. guazumaefolia, with small to medium flowers and a white
hirsute corona, placed closer to another relatively small-flowered species, P. bahiensis.
Third is the subcluster associating P. tiliifolia and P. ligularis (series Tiliifoliae), which
does not include P. maliformis. The latter is highly variable and forms a cluster of its
own, only including the very similar P. serrulata, which essentially differs by its
serrated and bi-/trilobed leaves and smaller flowers and fruits. Fourth, the series
Lobatae is split between the loose subcluster of P. subpeltata and P. gibertii and a
diverging P. caerulea branch. In two cases, there is no discordance, but only two
species are represented, i.e. P. quadrangularis and P. alata (series Quadrangulares),
133
Chapter IV. Morphological diversity
______________________________________________________________________
and P. lehmannii and P. smithii (series Kermesinae). The subdivisions observed within
the Passiflora cluster reinforce the general impression that the creation of series have
essentially resulted from the too obvious morphological similarity within small groups
of species of subgenus Passiflora, which have provided the core of the taxonomist’s
series or sections. This is the case for Tiliifoliae, where P. tiliifolia, P. ligularis,
P. palenquensis, P. platyloba are so closely related that their descriptions are often
accompanied by a detailed comparison. Another case is that of series Quadrangulares,
with P. alata, P. quadrangularis and P. trialata, and series Kermesinae, with
P. lehmanni, P. smithii and P. trisulca, all species that are differentiated by very few
traits, such as nectary shape. The most difficult case is that of Laurifoliae, where many
species can only be distinguished with considerable difficulties, as very few traits of
unknown genetic determinism (exact position of nectaries on petiole, number and
relative length of corona whorls), are used to differentiate them. When, in a second step,
other, better-differentiated species, are aggregated to these uniform groups, problems
and inconsistencies arise, and the solutions depend on which traits are prioritized among
the dazzling quantity of polymorphic traits. Another option is to widen the infrageneric
groups. Thus, in their classification, Feuillet & MacDougal (2003) have gathered series
Laurifoliae, Quadrangulares, and Tiliifoliae in a supersection Laurifolia. This view is
not supported by our results either. Indeed, in our tree, the link between P. popenovii,
P. nitida Kunth, P. quadrangularis, P. ligularis, and P. tiliifolia, is not supported, and
the accessions of P. maliformis (also classified in supersection-Laurifolia /seriesLaurifoliae by Feuillet & MacDougal) are placed in a distinct subcluster. In conclusion,
our analysis of the Passiflora cluster can be conciliated neither with the infrasubgeneric
classification of Killip, nor with that of Feuillet & MacDougal. Its essentially radial
structure points to the risk of overclassification instead, which is probably the reason
why Killip did not use the section level in this subgenus.
IV.1.4.7. The “Tacsonia cluster”
The structure of the branch corresponding to subgenera Manicata and Tacsonia shows
the close relationship between the most common species of the latter, P. mixta,
P. tripartita, P. tarminiana, and P. cumbalensis, forming a cluster onto which are
grafted the less common P. mathewsii (very similar to P. mixta) and P. luzmarina (very
similar to P. cumbalensis). The second subcluster, constituted by Colombian endemics,
indicates greater genetic distances. It gathers species from sections Parritana
134
Chapter IV. Morphological diversity
______________________________________________________________________
(P. parritae (Mast.) L.H. Bailey and P. jardinensis L.K. Escobar) and Colombiana of
Escobar. The latter is represented by series Colombianae (P. lanata and P. adulterina L.
f.), Quindiensae (P. linearistipula) and Leptomischae (P. antioquiensis H. Karst..,
P. flexipes Triana & Planch. and P. tenerifensis L.K. Escobar). Escobar’s classification
is not supported by our analysis. Among the common species, there appears no division
between sections Tacsonia and Bracteogama. Previous morphological and isozyme
studies (Villacís et al., 1998, Segura et al., 2002, 2005), as well as hybridization
experiments (Schöniger, 1986), have clearly shown closer similarity of P. tripartita
(section Bracteogama) with P. mixta (section Tacsonia) than with P. cumbalensis
(section Bracteogama). In the second subcluster, there appears to be a slight
differentiation of species with “normal” peduncle from those with extremely long
peduncles (series Leptomischae of section Colombiana). In the former group, all
remaining sections and series are mixed. Only the two species of series Colombianae,
P. lanata and P. adulterina, show a very close morphological similarity, with distances
remaining well under the order of intraspecific variation. These distances would have
been only slightly higher if pubescence had better been taken into account. In this
respect, it must be noted that the two specimens of P. lanata were typical for all
characters, and they exhibited a lanate ovary. Killip (1938) and Uribe (1955a) mention a
glabrous ovary for this species, a trait not mentioned in the original description by
Jussieu (1805). Escobar describes the plant as pubescent, except for leaf upper face,
with lanate flowers and fruits, which implies necessarily that the ovary is lanate too. In
herbarium material, we have found both glabrous and lanate ovaries for this species.
The confuse situation between P. lanata, P. adulterina, P. cuatrecasasii Killip, and the
recently proposed P. formosa (whose author mentions a lanate ovary as an important
distinctive feature - in fact the only qualitative one - in the comparison with P. lanata),
justifies a multivariate analysis to separate intra- and interspecific variation in what
could be a species complex or simply a lower number of variable species.
IV.1.4.8. The “Decaloba cluster”
Differentiation appears higher in subgenus Decaloba than in subgenera Passiflora and
Tacsonia, with the formation of six subclusters. The largest one is formed by accessions
of section Decaloba, including P. alnifolia, P. bogotensis, P. biflora Lam.,
P. cuspidifolia Harms, P. erythrophylla Mast., P. magdalenae Triana & Planch. (series
Punctatae), P. filipes Benth (series Lutae, although very similar to P. magdalenae),
135
Chapter IV. Morphological diversity
______________________________________________________________________
P. misera Kunth and P. trifasciata Lemaire (series Miserae). The substructure of this
large cluster does not support Killip’s subdivisions. Instead, it agrees with the grouping
of all these species in a uniform section Decaloba of supersection Decaloba, as
proposed by Feuillet & MacDougal. P. auriculata is placed at the base of this large
cluster, at a respectable distance, which may justify its classification in a distinct series
(Killip’s Auriculatae) or supersection (Feuillet & MacDougal supersection Auriculata).
P. sexflora (section Decaloba series Sexflorae) is even more isolated from the other
species of section Decaloba. Its placement in our tree questions its inclusion within
Killip’s section Decaloba, as well as in the section Decaloba sensu Feuillet &
MacDougal. P. guatemalensis, is also individualized, which is consistent with its
classification in a distinct section Hahniopathanthus in both classifications. The sixth
branch of the Decaloba cluster is subdivided in two consistent branches, one gathering
P. suberosa and P. coriacea, and the other uniting P. rubra and P. capsularis (section
Xerogona in both classifications). P. suberosa and P. coriacea are classified in a same
section Cieca in both classifications, however this section is placed in supersection
Decaloba in the proposal of Feuillet & MacDougal, which is not supported by our
results. As could be expected in such common and widely distributed species, the
accessions of P. suberosa, P. capsularis and P. rubra present a high polymorphism.
However, the two last species could not be separated on the basis of their morphology.
Their accessions cluster two by two according to the population of origin, sometimes
with relatively wide intrapopulational variation, but not following interspecific
boundaries. Indeed, all authors insist on the difficulty to distinguish them. Killip (1938)
and Holm-Nielsen et al. (1988) state that this is practically impossible on sterile
specimens. Their descriptions give particular importance to fruit shape and ovary
pubescence, although they do not agree on the distribution of ovary pubescence between
the two species. Ulmer & MacDougal (2004) insist on the color of the corona base. Our
identification was based on these three criteria, prioritizing fruit shape and corona color
(see Figure 7), but then we had to admit three different levels in ovary pubescence
(glabrous, puberulent, hirsute) in P. capsularis. Changing the priority of criteria would
just give a different but also inconsistent identification. Thus it appears that these traits
can combine in different ways, so they are not reliably discriminant, and the analysis for
all other traits, including or not pubescence, only confirms that the status of these two
species should be revised, downgrading them to botanical varieties, unless elements of
reproductive biology contradict this view. But even this is improbable, as self136
Chapter IV. Morphological diversity
______________________________________________________________________
incompatibility and self-compatibility coexist within P. capsularis (Ulmer and
MacDougal, 2004).
IV.1.4.9. The “Astrophea cluster”
Subgenus Astrophea appears very uniform, which could be expected as only one section,
Euastrophea, is represented in our small sample of this subgenus. The accessions share
most of their qualitative traits, and the divergence of one population of P. emarginata
can only be attributed to differences in leaf shape descriptors. This accession also differs
slightly in the pubescence of the ovary, but this trait did not interfere as it was not used
in the cluster analysis. The sample was too limited to draw any conclusion on intra-and
interspecific boundaries between the species represented.
n = 10
n = 12
n=6
I
n = ??
n=9
II
III
Figure 6. Dendrogram obtained on complete set of qualitative data. Distances of Sokal & Michener.
137
Chapter IV. Morphological diversity
______________________________________________________________________
I
Figure 6a. First part of the dendrogram obtained on the complete set of qualitative data.
138
Chapter IV. Morphological diversity
______________________________________________________________________
II
Figure 6b. Second part of the dendrogram obtained on the complete set of qualitative data.
139
Chapter IV. Morphological diversity
______________________________________________________________________
III
Figure 6c. Third part of the dendrogram obtained on the complete set of qualitative data.
140
Chapter IV. Morphological diversity
______________________________________________________________________
a
b
Figure 7. Morphological affinity between typical representatives of P. rubra (a) and P. capsularis (b).
Accessions from Colombia (a, Calarcá, Quindío – b, Cartago, Valle del Cauca).
IV.1.4.10. Morphological and molecular diversity
To appreciate the reliability of the morphological approach to Passiflora diversity, we
have compared some of the interspecific associations or divergences with similar results
obtained in phenetic studies with biochemical and molecular markers on samples
including some of our species. Indeed, a first series of genetic studies were carried out
on smaller species samples, mostly from Colombia too. In the trees obtained with
RAPD and cpDNA RFLP markers by Fajardo et al. (1998) and Sánchez et al. (1999),
the species of subgenus Tacsonia only constitute one subcluster within a large cluster
gathering them with species of subgenus Passiflora. Subgenus Decaloba is represented
by P. coriacea and P. adenopoda, both species strongly diverging from this Passiflora-
141
Chapter IV. Morphological diversity
______________________________________________________________________
Tacsonia cluster, but also between themselves, which appears consistent with our
results. In the RAPD study, subgenera Distephana and Astrophea are represented
respectively by P. vitifolia and P. spinosa (Poepp. & Endl.) Mast., and both species are
placed at considerable distance from the Passiflora-Tacsonia cluster, their divergence
being intermediate between that of P. adenopoda and that of P. coriacea. Within the
Tacsonia subcluster, the distances between the species, P. tripartita var. mollissima,
P. cumbalensis, P. pinnatistipula, and P. antioquiensis, follow the same order as in the
corresponding morphological cluster. This is still true when the comparison is extended
to subgenus Manicata, considering the results obtained by Segura et al. (2002, 2003)
with AFLP markers (P. tenerifensis and P. parritae also included), and with isozymes,
although P. antioquiensis is placed closer to the most common tacsos than
P. pinnatistipula in the isozyme study. Another convergence between morphological
and AFLP markers is the clear separation of P. maliformis from the typical species of
series Tiliifoliae of subgenus Passiflora (Ocampo et al., 2004).
Genetic relationships between subgenera and between particular species can also be
deduced from subsequent phylogenetic studies carried out on wider species samples by
Muschner et al. (2003) with ITS, trnL-trnF and rps4 sequences, by Yockteng (2003)
with chloroplastic matK sequences, Yockteng & Nadot (2004) with sequences of the
nuclear chloroplast-expressed glutamine synthetase gene (ncpGS), and Hansen et al.
(2006) with trnL/trnT sequences. All these studies support the existence of three major
clades, one corresponding to subgenus Astrophea, one formed around subgenus
Decaloba, and one formed around subgenus Passiflora. When included in these studies
(i.e. all studies except that of Muschner et al., 2003), species of subgenus Tacsonia form
a subclade within the Passiflora clade. P. vitifolia, and other representatives of
subgenus Distephana in the Yockteng’s studies, are also included in this large
Passiflora clade, although they are not grouped consistently in the matK tree (Yockteng,
2003). The relative position of the three major clades differs among studies. Subgenus
Astrophea takes an intermediate position in the ITS tree, it is closer to the Passiflora
clade in the other trees of Muschner et al. (trnL-trnF and rps4), but closer to the
Decaloba clade in the trnL-trnT tree of Hansen et al., while the Decaloba clade appears
closer to the Passiflora clade in the sample of Yockteng (ncpGS and matK trees).
P. foetida is clearly placed within the Passiflora clade according to rps4 and trnL-trnF
sequences, but more distant, although basal to this clade, in the trees obtained with ITS,
142
Chapter IV. Morphological diversity
______________________________________________________________________
ncpGS, and matK sequences. P. adenopoda is basal to the Decaloba clade in the ncpGS
tree, as well as in the ITS study of Krosnick & Freudenstein (2005). P. morifolia Mast.,
another species of section Pseudodysosmia, is also basal to the Decaloba clade, on a
branch between Decaloba and Astrophea, in the ITS and trnL-trnF trees. In the ncpGS
tree of Yockteng & Nadot (2004) and the ITS study of Krosnick & Freudenstein (2005),
subgenus Tryphostemmatoides, represented by P. tryphostemmatoides Harms and by
P. arbelaezii L. Uribe respectively, is placed close to subgenus Astrophea. Strangely, in
the study of Yockteng & Nadot (ncpGS tree), P. sanguinolenta Mast., a representative
of subgenus Psilanthus, is placed in a Passiflora subclade dominated by representatives
of series Tiliifoliae.
Within the Decaloba clade, we can recognize the higher level of divergence between
subclusters, the association we have observed between representatives of series
Punctatae and Miserae in the widest subclade (trnL-trnF tree of Muschner et al., 2003),
the relative separation of P. sexflora from this group, and the even more distant position
of P. coriacea and of the couple formed by P. capsularis and P. rubra (ITS, trnL-trnF,
and ncpGS sequences). However, the two last species are closely associated with
P. sexflora in the ncpGS tree.
As in our morphological study, the Passiflora clade identified in phylogenetic studies
generally shows loose relations between species. The interpretation of the poorly
supported subclades is very uneasy, with the partial exception of the ncpGS tree
presented by Yockteng & Nadot (2004), where branches are better defined, although not
easier to interpret, given, for example, the dispersion of species belonging to the series
Incarnatae and Laurifoliae. Among the close associations documented by our
morphological study, we can only recognize those of P. quadrangularis with P. alata
and P. incarnata with P. edulis (ITS and ncpGS trees).
As a first major point of conclusion on this comparison between morphological and
molecular diversity, we can underline that the major morphological divisions observed
in our study find support in the genetic studies. The cytological groups are always
validated,
with
the
clear
separation
of
subgenera
Astrophea
(n = 12),
Tryphostemmatoides and Decaloba (n = 6) between themselves and from subgenera
Passiflora,
Tacsonia,
and
Distephana
143
(n = 9).
Concerning
subgenus
Chapter IV. Morphological diversity
______________________________________________________________________
Tryphostemmatoides, the consistency between morphological and genetic studies is
clear only when considering our quantitative analysis, where it is associated with
subgenus Astrophea mostly on peduncle traits (third principal component). This trait is
represented also in the qualitative descriptors, however its effect is blurred by the high
number of traits shared with subgenus Decaloba. While the comparison is difficult for
subgenus Tryphostemmatoides, it is impossible for subgenus Psilanthus, because of
insufficient data and the unlikely placement of P. sanguinolenta in the Passiflora clade
in the ncpGS study. The two species, P. adenopoda and P. foetida, that take an
intermediate position in the general “morpho-cytological” pattern, or their close
relatives, are consistently placed in intermediate positions, in most phylogenetic studies,
P. adenopoda or P. morifolia (section Pseudodysosmia of subgenus Decaloba)
appearing basal to a general Decaloba clade and P. foetida (subgenus Dysosmia) basal
to the general Passiflora clade.
The comparison becomes more difficult at lower, infrasubgeneric, levels. Subgenus
Decaloba appears better structured than the other numerous subgenera, and shows
similarities in morphological and molecular diversity patterns, with the grouping of
Killip’s sections Punctatae and Miserae, and the differentiation of species of sections
Xerogona, Cieca, and series Auriculata and, less clearly, Sexflorae. The placement of
P. adenopoda in the different trees questions the inclusion of section Pseudodysosmia,
while the structure observed among representatives of several sections provides support
to some simplification, but not for as many fusions as those operated in the new
morphological classification of Feuillet & MacDougal. In any case, more species should
be gathered in a same phenetic study before revising objectively the morphological
classification.
Within the n = 9 group, molecular data and morphological diverge partially, as studies
of DNA sequences allow the distinction of a Tacsonia-Manicata group and fail to
separate clearly subgenus Distephana, placing both of them within a Passiflora clade,
while morphological analysis supports these three subgenera at the same level of
differentiation. The fact that species of subgenera Distephana, Tacsonia and Manicata
have developed ornithophyly is obviously related to their strong morphological
differentiation, which does not minor the importance of their separation from subgenus
Passiflora. Whether their probable evolution from a “Passiflora-like” common ancestor
144
Chapter IV. Morphological diversity
______________________________________________________________________
justifies their inclusion in the bee-pollinated Passiflora subgenus, as proposed in the
new classification, is just the same classical question about considering birds as
dinosaurs. In the end, it seems a problem of putting more emphasis on the adaptative
forces commanding evolution or more emphasis on the genetic structure that subtend
them. Concerning subgenus Passiflora sensu Killip, no clear structure appears at the
interspecific level that could result in clear subdivisions into series. The study of
sequence variation for the ncpGS gene provides the only tree with reasonably well
supported structure at this level, however several obvious abnormalities question the
robustness of the information. Our morphological observations only confirm closer
associations between the most typical representatives of some series, however the
number of contradictions with the classification and the lack of a clear hierarchy in the
branch structure point to the difficulty of the work and the risk of under- or
overclassification, leading to chose between a limited number of poorly supported series
or a great number of poorly represented series. Similarly, the structure of the TacsoniaManicata branch does not support clearly sections and series in subgenus Tacsonia,
however it allows differentiation between two groups of tacsos, one corresponding to
common species that probably have their center of diversity in Ecuador, as is obvious
for P. cumbalensis, P. luzmarina and P. matthewsii, and very likely for P. mixta,
P. tripartita and P. tarminiana (Segura et al., 2005), and another cluster only including
species endemic to Colombia, with a slight but clear differentiation related to extreme
variation for peduncle length.
IV.1.5. Conclusions
In the absence of a clear set of morphological criteria for discriminating at the different
hierarchic levels of the infrageneric classification of Passiflora, we have used a quite
exhaustive list of 43 quantitative and 84 qualitative descriptors. A shorter list of 32
qualitative traits, selected after analyzing variation among Killip’s subgenera, allowed
to classify our 60-species sample consistently, using a strictly phenetic approach. Most
discriminant characters include size of stems and leaves, presence of tendrils, number
and distribution of extrafloral nectaries, dimensions and general shape of bracts, width
and length of flowers, corona complexity, and, although they could not be
systematically analyzed, fruit size and color. Eight of the nine Killip’s subgenera
145
Chapter IV. Morphological diversity
______________________________________________________________________
represented in our sample are supported by the morphological analysis, although
subgenus Tryphostemmatoides is only supported in the quantitative analysis. By
contrast, the simplification proposed by Feuillet & MacDougal is not clearly supported
in our analyses, except for the possible inclusion of P. manicata in subgenus Tacsonia,
as this species is intermediate with subgenus Passiflora for quantitative traits but very
similar to tacsos for most qualitative traits. Furthermore, the placement of P. adenopoda
and its close relatives, in our analyses as well as in molecular ones, questions their
treatment as a division of subgenus Decaloba. More generally, as compared to
molecular results, our phenetic approach provides a better resolved vision of relations
among passifloras. Beyond the differentiation between cytological groups (n = 6, 12 or
n = 9; de Melo et al., 2001) underlined in most analyses, appear other features of
considerable importance for their evolution. The division between the two cytological
groups is particularly paralleled by a division on the presence and position of extrafloral
nectaries and the complexity of the corona, showing the importance of coevolution. In
the same line, Yockteng (2003) underlined the differentiation appearing in the spectrum
of cyanogenic components developed against herbivores in the two cytological groups.
Even clearer appears the coevolution with pollinators, causing the main line of floral
divergence between subgenera Passiflora, pollinated by large bees, Tacsonia, pollinated
only by the sword-billed hummingbird, and Distephana, pollinated by other
hummingbirds. In the n = 6/12 group, where small to medium size insects dominate as
pollinators, the morphological divergence of P. trinervia (subgenus Psilanthus),
showing exactly the same adaptation to Andean highlands and pollination by the same
bird species as subgenus Tacsonia, is not less significant. This convergence is logically
expressed in the evolution of the corona. However, the minute corona of P. trinervia,
and other typical representatives of subgenus Psilanthus, probably results from the
transformation of the two-ranked corona common in subgenera Astrophea and
Decaloba, while the reduced corona of most tacsos has evolved from the complex
corona observed in all species of subgenus Passiflora at the same time as their
specialized hypanthium. Remnants of these complex coronas can still be observed in
species with a less developed hypanthium, as P. manicata and P. antioquiensis, with
one or two external whorls, plus a few other series of thinner filaments under the floral
tube throat, and even in typical long-tubed tacso flowers. Thus, a more or less
developed second whorl is not rare in P. mixta, while loose whorls of very thin white
filaments can be observed far in the hypanthium of P. tripartita var. mollissima. The
146
Chapter IV. Morphological diversity
______________________________________________________________________
evolution has probably been different for subgenus Distephana, and its differentiation
within the n = 9 group earlier in the history of Passiflora, as its species show no such
signs of an ancestral complex corona. Instead, they are constituted by two or three rows,
which can be compared with numbers observed in subgenera Decaloba and Astrophea.
The fusion of the elements into a prolongation of the floral tube is paralleled in certain
representatives of subgenus Decaloba, such as P. tulae Urban (Murucuja), forming
another striking case of convergence between species from very different evolutive
backgrounds.
IV.1.6. Acknowledgements
This research has been funded by Colciencias and the Colombian Ministry for
Environment, with support of the Research Center of the Colombian Coffee Grower
Federation (Cenicafé) through the projects: ‘Conservación y utilización de los recursos
genéticos de pasifloras’ and ‘Estudio de la diversidad de las Passifloraceae y Caricaceae
de la zona cafetera de Colombia’. The authors are indebted to María Restrepo, Felipe
Barrera, Cristián Olaya and Lina Farfán for assistance in collecting field data, Daniel
Franco and Mario Ruiz for help in installation works of living collections (Paraguacito
Experimental Station – Cenicafé), and to German Arroyave and Juan G. Contreras
(PASSICOL S.A) for providing living collection facilities for maracuja. The first author
gratefully acknowledges financial support from the Gines-Mera Fellowship Foundation
(CIAT - CBN). Finally, they are also thankful to Dr. Philippe Feldmann (CIRAD) for
comments and suggestions.
147
CHAPTER V
____________________________________
Chloroplast and mitochondrial DNA variation in
the genus Passiflora L. (Passifloraceae) as revealed
by PCR-RFLP
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
V.1. Chloroplast and mitochondrial DNA variation in the genus Passiflora L.
(Passifloraceae) as revealed by PCR-RFLP
John Ocampo Pérez1,2*, Geo Coppens d’Eeckenbrugge2, Ange-Marie Risterucci3.
1
Bioversity International (formerly IPGRI), Regional Office for the Americas, A.A. 6713, Cali, Colombia;
CIRAD, UPR ‘Gestion des ressources génétiques et dynamiques sociales’, Campus CNRS/Cefe, 1919
route de Mende, 34293 Montpellier, France; 3CIRAD-AMIS, UMR 1096, Avenue Agropolis, TA 40/03,
34398 Montpellier, cedex 1, France.
2
V.1.1. Abstract
The chloroplast and mitochondrial DNA diversity of 213 accessions belonging to 151 Passiflora species of
15 subgenera recognized by Killip (1938) was studied by PCR-RFLP analysis of two non-coding cpDNA
regions (psbC- trnS and trnS - trnfM) and two non-coding mtDNA region (nad4-1/2 and nad1-B/C). This
sample set was supplemented with six accessions from three African Passifloraceae genera Adenia,
Barteria, and Smeathmannia, as outgroup species. The PCR-amplified cpDNA regions were digested with
six endonucleases. A total of 614 fragments were scored, of which 93% were found to be polymorphic in
the sample. Two-hundred-eighty haplotypes were found for the chloroplast and 372 for the mitochondria. A
higher level of interspecific variation was detected in the mtDNA regions than in the cpDNA regions. The
first two axes of the principal co-ordinates analysis accounted for 59% of the total variation on cpDNA
data. They allowed visualizing a strong structure, as the genera Adenia, Barteria, and the Passiflora
subgenera Astrophea, Calopathanthus, Dysosmia, Distephana, Manicata, Passiflora, Tacsonia,
Tacsonioides and Tryphostemmatoides, and P. deidamioides occupy the left half of the principal plane,
while the species of subgenera Apodogyne, Decaloba, Murucuja, Pseudomurucuja and Psilanthus form a
very well separated group, placed in a quite extreme position on the right, only the accessions of
Smeathmannia and P. lancetillensis (subgenus Deidamioides) taking intermediate positions. The
phenogram obtained by the neighbor-joining method on cpDNA data is more coherent with the major
divisions of the taxonomy proposed recently by Feuillet & MacDougal than the corresponding mtDNA
tree. The cpDNA tree shows three major, well supported clusters within Passiflora. The first one, named
the “Passiflora group”, includes subgenera Calopathanthus, Deidamioides, Distephana, Dysosmia,
Dysosmioides, Manicata, Passiflora, Tacsonia, and Tacsonioides, with a very loose substructure and
considerable intraspecific variation. The second one includes subgenus Astrophea, and the third one, named
the “Decaloba group”, comprises most species of subgenera Apodogyne, Decaloba, Murucuja,
Pseudomurucuja and Psilanthus. P. gracillima (subg.enus Tryphostemmatoides) appears in basal position,
forming a fourth cluster of its own. The position of P. lancetillensis (subgenus Deidamioides) is undefined,
as it is placed on a long branch between the outgroup and the “Decaloba group”, the outgroup itself taking
an undefined position among the three major Passiflora clusters. The phenogram obtained with mtDNA
data separates four main clusters. As for cpDNA, a first large, well supported, cluster corresponds to the
“Decaloba group”, where accessions are grouped by species, with the only exception of P. adenopoda. The
numerous accessions of section Decaloba series Punctatae and Miserae tend to form a subcluster, while
Killip’s subgenera Apodogyne, Murucuja, Pseudomurucuja and Psilanthus cannot be recognized. The other
major clusters are different from those evidenced by cpDNA data, as subgenera Astrophea and
Tryphostemmatoides appear integrated within the “Passiflora group”, while subgenus Tacsonia forms a
uniform distinct cluster, close to another one comprising species of series Kermesinae, Simplicifoliae,
Lobatae, and Menispermifoliae. The analyses of chloroplastic and mitochondrial fragments gave very
different pictures on the genetic structure of genus Passiflora. Differences appear at all levels, in the
position of the outgroup, the relative position of four subgenera, and the relationships between species. The
divergence in the information obtained from chloroplast and mitochondrial genomes are attributed to
differences in their rate of evolution and mode of transmission and to reticulate evolution in the genus.
Key words: Passiflora, PCR-RFLP, chloroplast and mitochondrial DNA, variation, evolution.
149
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
V.1.2. Introduction
The family Passifloraceae is divided into two tribes, Pariopsieae and Passifloreae, and
includes more than 650 species (Escobar, 1988; Ulmer & MacDougal, 2004) distributed
throughout the tropics. The base number of chromosome varies among x = 6 and x = 9
(De Melo & Guerra, 2003). With approximately 525 species, including several cultivated
ones, Passiflora is numerically and economically the most important genus of the family.
Passionflowers are generally vines, although some representatives are shrubs or trees.
P. edulis Sims (yellow maracuja) is by far the best-known and economically most
important species of the family.
In the last extensive revision of the genus, Killip (1938) classified 355 Passiflora species
into 22 subgenera (Annex 1). For long, while the list of species was considerably
extended, his views were only amended or supplemented. In Colombia, Escobar
(1988a,b, 1989, 1990 inedited, 1994) reviewed subgenera Astrophea, Distephana,
Manicata, Rathea and Tacsonia, merging subgenera Tacsoniopsis and Tacsonia,
subdividing them into sections and series, and proposing one additional subgenus,
Porphyropathanthus. Recently, Feuillet & MacDougal (2003; Annex 2) have proposed a
new infrageneric classification of Passiflora recognizing only four subgenera, further
divided into 16 supersections. Three of their subgenera are strictly American: Astrophea,
Deidamioides, and Passiflora, with 57, 17 and 234 species respectively, with an
essentially tropical distribution. Decaloba, with its 204 species, is mainly distributed in
America, but it is also represented in Southeast Asia and Australia (Ulmer & MacDougal,
2004). In this new proposal, the genus Tetrastylis is placed as a section of subgenus
Deidamioides, and the subgenus Tetrapathea from New Zealand is excluded from the
genus Passiflora (Hutchinson, 1967; Green, 1972; Yockteng & Nadot, 2004).
Both classifications of Killip (1938) and Feuillet & MacDougal (2003) are based on the
extreme morphological richness and complexity of Passiflora, whose species present
numerous particular traits, including a wide variation in leaf shape, even within species
and within individuals (heterophylly and heteroblasty), the presence of extrafloral
nectaries in different parts (on leaf petiole, lamina or margins, on bracts, on sepals), floral
150
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
traits showing a high level of coevolution with particular pollinators, particularly in
corolla color, the variable development of the hypanthium and that of the corona.
However, no clear hierarchy emerges in the relative contributions of these traits to
taxonomy of the genus. Ocampo & Coppens d’Eeckenbrugge (in preparation; cfr.
Chapter IV), have used a list of 127 morphological descriptors in a phenetic approach of
morphological differentiation among 60 Passiflora species from nine of the Killip’s
subgenera. The quantitative descriptors clearly separated subgenera Astrophea,
Decaloba, Passiflora, Psilanthus, Tacsonia, and Tryphostemmatoides. Subgenus
Dysosmia appeared intermediate between Passiflora and Decaloba, while the
representatives of subgenera Manicata and Distephana showed affinity with subgenera
Tacsonia and Passiflora respectively. Qualitative trait analysis showed a major
distinction between species with 2n = 12 or 24 chromosomes, and species with 2n = 18
chromosomes, and discriminated more clearly among Killip’s subgenera Astrophea,
Psilanthus, Decaloba, in the former cytological group, and Distephana, Dysosmia,
Passiflora and Tacsonia, in the second cytological group. Subgenera Manicata and
Tryphostemmatoides could not be distinguished from subgenera Tacsonia and Decaloba
respectively. While these main divisions could be easily interpreted following Killip’s
subgenera, this was not true at lower levels. No clear subdivisions were observed within
subgenus Passiflora and only a slight geographical structure was detected in subgenus
Tacsonia. Differentiation was higher in subgenus Decaloba, where subclusters partly
supported the simplification proposed by Feuillet & MacDougal. The most obvious
contradiction with both classifications was the placement of P. adenopoda (subgenus
Decaloba section Pseudodysosmia) in an intermediate position, between subgenera
Passiflora and Decaloba.
In many plant groups, molecular genetics have provided powerful new tools to
understand the structure and evolution of species diversity. The first studies of genetic
variation in Passiflora were based on RAPD, RFLP, AFLP and isozyme markers, and
initiated in a breeding perspective, focusing on cultivated species and their wild relatives
in subgenera Passiflora and Tacsonia (Fajardo et al., 1998; Sánchez et al., 1999, Segura
et al., 2002; Segura et al., 2003a; Ocampo et al., 2004; Segura et al., 2005). As discussed
151
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
in the preceding chapter, their results are consistent with our morphological analyses, but
their narrow samples limited the reach of comparisons. The application of molecular
tools to phylogenetic analysis of wider samples is more recent.
The first genus-wide molecular analysis of Passiflora, using the nuclear ribosomal
internal transcribed spacers, (ITS-1 and ITS-2), plastid trnL-F intergenic spacer and the
rps4 plastid gene, was reported by Muschner et al. (2003; Annex 4a). Eleven subgenera
were represented by 61 species. The ITS alignment appeared highly variable, with shorter
sequences in subgenera with 2n = 18 or 2n = 20 chromosomes than in subgenera with
2n = 12 or 24 chromosomes. The three phylograms consistently showed three major
clades, corresponding with three of the subgenera proposed by Feuillet & MacDougal
(2003), i.e. the ‘Passiflora clade’, composed of 2n = 18/20 species, the ‘Decaloba clade’
composed of 2n = 12 species, and one composed of 2n = 24 species (i.e. subgenus
Astrophea). The representatives of subgenus Deidamioides contributed to the Decaloba
clade in the ITS phylogram, while they appeared independent from the three major clades
in the trnL-trnF and rps4 phylograms. The substructure of the two most numerous clades,
‘Passiflora’ and ‘Decaloba’, were not consistent with either taxonomical classification.
Within the ‘Passiflora clade’, bootstrap values are low and there is no clear structure that
could correspond to sections or series, although a few associations are recognized
between closely related species, such as P. alata and P. quadrangularis, P. edulis and
P. incarnata, in the ITS tree. These associations are not expressed in the trnL-trnF tree,
where even the two species of the uniform subgenus Distephana do not appear related.
The ‘Decaloba clade’ shows a stronger substructure, with an association between species
of sections Decaloba, Miserae and Punctatae of subgenus Decaloba, an smaller
subclusters
representing sections Cieca and Xerogona, while P. morifolia (section
Pseudodysosmia) takes a relatively basal position in both phenograms. However, in the
trnL-trnF tree, P. rubra strongly diverges from the Decaloba clade, which is not the case
for P. capsularis, a species that is so similar that our morphological analysis led us to
consider the two as synonyms (see Chapter IV). An even stronger inconsistency concerns
the placement of the outgroups in both phenograms. Adenia keramanthus is positionned
as sister to the ‘Passiflora clade’, with a support value of 97%, while Mitostemma
152
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
brevifilis is placed as sister to the ‘Decaloba’ and Astrophea clades in the ITS tree, both
species diverging from Passiflora species in the trnL-trnF tree. In addition, Mitostemma
brevifilis also appears in an unexpected position at the base of the ‘Passiflora’ and
Astrophea clades in the rps4 tree. Muschner et al. (2003) concluded that the monophyly
of the genus Passiflora was not supported by any statistical or phylogenetic method, “and
in several trees there was even support of a non-monophyletic Passiflora”.
One year after Muschner et al. (2003), Yockteng & Nadot (2004; Annex 4c) published a
phylogenetic study of a 91 species-sample representing 17 of the 23 subgenera
recognized by Killip (1938) and Escobar (1988a), based on the sequences of the
chloroplast-expressed glutamine synthtase gene (ncpGS). Their phenogram showed the
same three major clades, with significant differences in length of introns. These results
provided support to three of the four subgenera defined by Feuillet & MacDougal (2003),
but indicated that three additional subgenera, Polyanthea (DC.) Killip, Dysosmia (DC.)
Killip, and Tetrapathea (DC.) Rchb., should also be recognized. The Astrophea clade is
sister to the other species, except P. cirrhiflora. P. tryphostemmatoides (subgenus
Tryphostemmatoides) appears closely related to it. Within the Passiflora clade, appears a
Tacsonia subclade that also includes P. manicata (subgenus Manicata) and P. racemosa
(subgenus Calopathantus). Two other subclades are more difficult to interpret. One of
them includes a small third-order cluster constituted by species of subgenus Distephana,
joined surprisingly by P. cincinnata (subgenus Passiflora series Incarnatae). Several
third-order subclusters appear dominated by particular morphological groups, but too
many exceptions hamper a consistent interpretation. The dispersion of species of the
uniform series Laurifoliae provides the best example of this situation. The inclusion of
P. sanguinolenta (subgenus Psilanthus) is the strangest abnormality in the Passiflora
clade. Like in the study of Muschner et al. (2003), the Decaloba clade shows a stronger
substructure, with similar associations, and relatively basal position of the species of
section Pseudodysosmia. Very surprisingly, this Pseudosysosmia cluster included
P. alnifolia, thus isolated from its closest relatives. An even more surprising placement is
that of P. candida (subgenus Astrophea) close to P. auriculata. The same large sample
was studied for sequences of the chloroplastic gene matK (Yockteng, 2003; Annex 4b).
153
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
The results were generally consistent in defining the same three clades, despite some
divergence in their relative positions. Tryphostemmatoides appeared fully included in the
Astrophea clade. Resolution was poor at the section and series level, however second
order subdivisions were also consistent with ITS and ncpGS results in the ‘Decaloba
clade’, and a Tacsonia-Manicata-Calopathanthus subclade was individualized within the
‘Passiflora clade’. P. candida was placed more logically in the Astrophea clade, which
discards the hypothesis of a misidentification invoked in the interpretation of the ncpGS
study.
Krosnick & Freudenstein (2005) showed that the 22 species of Old World Passiflora
form a monophyletic group and supported their placement by Feuillet & MacDougal in
supersection Disemma, within subgenus Decaloba, using plastid (trnL-F) and nuclear
(ITS) DNA sequences. More recently, Hansen et al. (2006; Annex 4d) analyzing
chloroplast sequences rpoC1 and trnL/trnT from 136 species of genus Passiflora, also
obtained the three major clades, with problems of resolution and consistency at section
and series level, and a few striking cases of inconsistent positioning. The most
spectacular example came from P. microstipula, a species with 2n = 18 chromosomes,
two accessions being placed in the Passiflora clade, close to the very different P. nitida,
and a third one at the base of the Decaloba clade. As stated by the authors, the most
likely explanation is an extreme case of chloroplast capture. Hansen et al. (2006) sustain
that their data unequivocally support the subgeneric classification of Feuillet &
MacDougal, however the position and composition of subgenus Deidamioides is not
clearly defined, and the phylogram presented places Dilkea and Tetrapathea as sister
group to a Decaloba-Deidamioides clade, which throws additional doubts on Passiflora
monophyly.
This revision of the phylogenetic studies consistently shows the existence of three major
clades in the genus Passiflora, with a stronger structure in the Decaloba clade than in the
Passiflora clade. When tacsos (species of subgenus Tacsonia) are included in the sample,
they are placed in a particular subclade of the latter. Beyond this general pattern, several
points of conflict are obvious too. First are the inconsistencies in the relative placement
154
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
of the three major clades. Such variation affects more markedly the placement of several
morphologically atypical species. Also, there are cases of highly improbable placement
concerning a few species. The variable placement of the outgroup species is even more
problematic, generating doubts on the monophyly of the genus. A comparison of all
phenograms also shows that chloroplastic sequence data tend to produce trees with less
consistent and less supported structure than the nuclear ncpGS and ITS trees.
In the last decade, the genetic information present in the plant chloroplastic genome has
been particularly used for phylogenetic purposes (Clegg et al., 1994; Vekemans et al.,
1998), because of its non-Mendelian mode of inheritance and low rate of evolution as
compared to that of the nuclear genome, which makes it very useful in studying the
variation at levels higher than the species, considering that the probability of detecting
intraspecific variation is low. This trend has been furthered accentuated by the
development of efficient techniques to study this variation, among them PCR-RFLP or
CAPS (cleaved amplified polymorphic sequence), which employs universal primers for
amplification of specific noncoding DNA, followed by their restriction. This technique
soon appeared to be more efficient than traditional RFLP to reveal polymorphism. Its
successful application to a number of crop and forest species for which extensive
nucleotide information is now available (Mes et al., 1997; Lakshmi et al., 2000; Parani et
al., 2000; Duval et al., 2003; Van Droogenbroeck et al., 2004; Kyndt et al., 2005), has
changed the common perception of the variability of chloroplast genome. Reviewing
cases of cpDNA variation at the intraspecific level, and even within individuals, Harris &
Ingram (1991) have underlined that hybridization, introgression and lineage sorting all
may influence a taxon’s position in cpDNA phylogenies. They have also underlined that
the occurrence of widespread biparental plastid transmission in the Angiosperms makes
the implicit assumption that hybridization can be ignored not always tenable. The
mitochondrial genome has been less used because of its high degree of intramolecular
recombination and low rate of base substitution. In Passiflora, biparental plastid
transmission was first suspected by Corriveau & Coleman (1988), following
epifluorescence microscopy observations on P. edulis. Do et al. (1992) showed that
RFLP markers of cpDNA were mostly inherited maternally in crosses between yellow
155
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
and purple maracuja when the former (P. edulis f. flavicarpa) was used as female parent,
and biparental in the reciprocal cross, suggesting asymmetric post-fertilization exclusion
processes. Transmission was paternal in the hybrid ‘P. coccinea x P. edulis f. flavicarpa’.
More recently, Mráček (2005) observed biparental transmission between P.
menispermifolia and P. oerstedii and heteroplasmy in resulting hybrids. Muschner et al.
(2006) established paternal transmission of cpDNA in four interspecific hybrids of
subgenera Passiflora and Dysosmia and maternal transmission in an interspecific hybrid
of subgenus Decaloba. All mtDNA were maternally transmitted in these five hybrids.
Hansen et al. (2007) studied 17 crosses and found paternal or biparental inheritance of
cpDNA, except in their intraspecific crosses, where it was predominantly maternal.
Hansen et al. (2007) evoked the possible effect of genetic divergence between the
parents, however it should be noted that the progeny studied was from P. costaricensis,
the only species of subgenus Decaloba in their study. Hence, an alternative and simpler
explanation, consistent with the results of Muschner et al. (2006), could be differences in
cpDNA transmission between subgenera.
The study presented here was planned before the publication of phylogenetic studies in
Passiflora. The implementation of projects including extensive collecting of Passiflora
germplasm in Colombia, the country with the widest Passiflora diversity (Chapter II),
provided a unique opportunity to study a very wide sample of the genus, taking into
account possible variation at the infraspecific level. The species sample was significantly
widened thanks to the contribution of the French National collection. The PCR-RFLP
technique was chosen because of its relative simplicity, rapidity, and cost efficiency. At
that time, sequence analyses would have been more expensive, not allowing the study of
the genomes of both organelles on so many individuals. Two cpDNA spacer regions,
(psbC-trnS and trnS-trnfM), and two mtDNA introns (nad4-1/2 and nad1-B/C) were
analyzed for variation at both intra-and intergeneric levels in the 151 Passiflora species
of our sample and four species of the Old World genera Adenia, Barteria and
Smeathmannia. The objective was to study the general structure of genetic diversity in
the genus, with particular attention to the cultivated species and their close relatives.
156
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
V.1.3. Materials and methods
V.1.3.1. Taxon sampling
The germplasm sample consisted of 213 individuals from 151 species, representing 15 of
the subgenera recognized by Killip (1938), Escobar (1988) and MacDougal (1994), and
eight outgroup accessions of Adenia, and Barteria, and Smeathmannia, from Africa
(Table 1). This sample included several representatives of the principal cultivated
species. Most of the materials were collected in Colombia, from the wild, home gardens,
and farms (see Chapter II). Voucher specimens were deposited at COL and VALLE
herbaria. Other samples were obtained from the living National Collection in Blois
(France), mainly collected in French Guiana and Brazil, and maintained by Christian
Houel. Specimen identification was based on the morphological descriptions of Killip
(1938), Holm-Nielsen et al. (1988), Escobar (1988a,b, 1994) and MacDougal (1992,
1994), and the comparison with herbarium material (mostly COL, HUA, MEDEL,
COAH, K, MA, MO, NY, P, PSO). Infrageneric taxonomy follows the same references,
unless specified otherwise.
Table 1. List of species used in this study according to classification by Killip (1938), Escobar (1988a,b)
and MacDougal (1994).
Infrageneric clasification
Collection data
Status
Passifloraceae Juss. ex Kunth.
Tribe Paropsieae D.C., 1828
Genus Barteria Hooker, J.D., 1860
B. fistulosa Mast.
Cameroon, Kandara (Central Province)
Wild
B. fistulosa Mast.
Cameroon, Ebodjé (Southwest Province)
Wild
B. nigritiana Hook. f.
Cameroon, Ebodjé (Southwest Province)
Wild
B. nigritiana Hook. f.
Gabon
Wild
B. solida F. J. Breteler
Cameroon, Dikome Balue (Southwest Province)
Wild
Genus Smeathmannia Solander ex R. Brown, 1821
S. pubescens Sol. ex R. Br.
Cameroon, Ebodje (Southwest Province)
Wild
S. pubescens Sol. ex R. Br.
Cameroon, Mamalles (Southwest Province)
Wild
National Collection - Blois - France
Wild
Tribe Passifloreae D.C., 1828
Genus Adenia Forsskal, 1775
Adenia glauca Schinz
Genus Passiflora L., 1753
Subgenus Astrophea (DC.) Mast., 1871
Section Dolichostemma
157
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
P. mariquitensis Mutis ex L. Uribe
Colombia, Tolima, Mariquita
Wild
Section Euastrophea
P. arborea Spreng.
Colombia, Caldas, Manizales
Wild
P. emarginata Humb. & Bonpl.
Colombia, Caldas, Manizales
Wild
P. emarginata Humb. & Bonpl.
Colombia, Valle del Cauca, Yotoco
Wild
P. sphaerocarpa Triana & Planch.
Colombia, Valle del Cauca, Cali
Wild
P. sphaerocarpa Triana & Planch.
Colombia, Tolima, Ibagué
Wild
P. macrophylla Spruce ex Mast.
Ecuador, National Collection - Blois - France
Wild
Section Pseudoastrophea
P. citrifolia (Juss.) Mast.
Guiana French, Belgian National Botanic Garden - Meise
Wild
P. citrifolia (Juss.) Mast.
National Collection - Blois - France
Wild
P. haematostigma Mart. ex Mast.
Guiana French, National Collection - Blois - France
Wild
P. kawensis Feuillet
National Collection - Blois - France
Wild
Subgenus Calopathanthus (Harms) Killip, 1938
P. racemosa Brot.
National Collection - Blois - France
Wild
Subgenus Decaloba (DC.) Rchb.
Section Cieca
P. apoda Harms
Colombia, Caldas, Manizales
Wild
P. coriacea Juss.
Colombia, Caldas, Palestina
Wild
P. coriacea Juss.
Colombia, Tolima, Ibagué
Wild
P. coriacea Juss.
Colombia, Valle del Cauca, Palmira
Wild
P. exoperculata Mast.
Ecuador, Tunguragua, Ceballos
Wild
P. gracilis J. Jacq. ex Link
USA, National Collection - Blois - France
Wild
P. holosericea L.
National Collection - Blois - France
Wild
P. monadelpha P. Jorg. & Holm-Niels.
Colombia, Valle del Cauca, El Cerrito
Wild
P. mutiflora L.
National Collection - Blois - France
Wild
P. suberosa L.
Colombia, Antioquia, Jerico
Wild
P. suberosa L.
Colombia, Caldas, Manizales
Wild
Series Apetalae
P. apetala Killip
National Collection - Blois - France
Wild
Series Auriculatae
P. auriculata Kunth
Colombia, Caldas, Victoria
Wild
P. auriculata Kunth
Guiana French, National Collection - Blois - France
Wild
P. jatunsachensis Schwerdtfeger
National Collection - Blois - France
Wild
Series Luteae
P. filipes Benth.
Colombia, Risaralda, Pereira
Wild
Series Miserae
P. amalocarpa Barb. Rodr.
National Collection - Blois - France
Wild
P. misera Kunth
Colombia, Valle del Cauca, Jamundi
Wild
P. misera Kunth
National Collection - Blois - France
Wild
P. trifasciata Lem
Colombia, Quindio, Buenavista - (introduced)
Wild
Section Decaloba
P. tricuspis Mast.
Bolivia, Santa Cruz
Wild
P. tricuspis Mast.
Brasil, National Collection - Blois - France
Wild
Series Punctatae
P. alnifolia Kunth
Colombia, Valle del Cauca, El Cerrito
Wild
P. alnifolia Kunth
Colombia, Caldas, Manizales
Wild
P. alnifolia Kunth
Colombia, Nariño, Pasto
Wild
P. biflora Lam.
Colombia, Caldas, Victoria
Wild
P. biflora Lam.
Colombia, Tolima, Mariquita
Wild
P. boenderi MacDougal
National Collection - Blois - France
Wild
P. bogotensis Benth.
Colombia, Cundinamarca, Bogotá
Wild
P. colinvauxii Wiggins
Ecuador, National Collection - Blois - France
Wild
158
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
P. cuneata Wild.
Colombia, Boyacá, Duitama
Wild
P. cuneata Wild.
National Collection - Blois - France
Wild
Wild
P. cupraea L.
National Collection - Blois - France
P. erytrophylla Mast.
Colombia, Boyacá, Duitama
Wild
P. gilbertiana MacDougal
National Collection - Blois - France
Wild
P. occidentalis
Colombia, Valle del Cauca, Buenaventura
Wild
P. pohii Mast.
National Collection - Blois - France
Wild
P. magdalenae Triana & Planch.
Colombia, Caldas, Victoria
Wild
P. magdalenae Triana & Planch.
Colombia, Cundinamarca, San Juan de Rio Seco
Wild
P. micropetala Mast.
National Collection - Blois - France
Wild
P. vespertilio L.
Guiana French, National Collection - Blois - France
Wild
P. yucatanensis Killip
Mexico, National Collection - Blois - France
Wild
Series Sexflorae
P. allantophylla Mast.
National Collection - Blois - France
Wild
Section Eudecaloba
P. aurantia G. Forst.
Australia, National Collection - Blois - France
Wild
P. herbertiana Ker-Gawl.
National Collection - Blois - France
Wild
Section Hahniopathanthus
P. guatemalensis S. Watson
Colombia, Caldas, Filadelfia
Wild
Section Organenses
P. ornithoura Mast.
National Collection - Blois - France
Wild
Section Pseudodysosmia
P. adenopoda Moc. & Sessé ex DC.
Colombia, Caldas, Manizales
Wild
P. adenopoda Moc. & Sessé ex DC.
National Collection - Blois - France
Wild
P. karwinskii Mast.
National Collection - Blois - France
Wild
P. lobata (Killip) Hutchinson ex MacDougal
National Collection - Blois - France
Wild
Section Pseudogranadilla
P. bicornis Mill.
National Collection - Blois - France
Wild
P. bicornis Mill.
National Collection - Blois - France
Wild
P. indecora Kunth
National Collection - Blois - France
Wild
P. telesiphe Knapp & Mallet
National Collection - Blois - France
Wild
Section Xerogona
P. rubra L.
Colombia, Quindio, Calarca
Wild
Subgenus Deidamioides (Harms) Killip, 1938
P. deidamioides Harms
National Collection - Blois - France
Wild
P. lancetillensis MacDougal & Meerman
National Collection - Blois - France
Wild
Subgenus Distephana (Juss,) Killip, 1938
P. aimae Annonay & Feuillet
Guiana French, National Collection - Blois - France
Wild
P. coccinea Aubl.
Guiana French, Belgian National Botanic Garden - Meise
Wild
P. quadriglandulosa Rodschied
National Collection - Blois - France
Wild
P. speciosa Gardn.
National Collection - Blois - France
Wild
P. variolata Poep & Endl.
National Collection - Blois - France
Wild
P. vitifolia (Harv.) Harms
Colombia, Chocó, Quibdo
Home garden
P. vitifolia (Harv.) Harms
Colombia, Tolima, Ibagué
Wild
Subgenus Disosmioides Killip, 1938
P. campanulata Mast.
Brasil, National Collection - Blois - France
Wild
P. setulosa Killip
National Collection - Blois - France
Wild
Subgenus Dysosmia DC., 1938
P. foetida var. moritziana (Planch.) Killip ex Pull
Guiana French, National Collection - Blois - France
Wild
P. foetida var. gossypiifolia (Desv.) Mast.
P. arida (Mast. & Rose) Killip
Colombia, Chocó, Quibdo
National Collection - Blois - France
Wild
Subgenus Manicata (Harms) Escobar, 1988
P. macropoda Killip
National Collection - Blois - France
Wild
159
Wild
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
P. manicata (Juss.) Pers.
Colombia, Quindio, Salento
Wild
P. manicata (Juss.) Pers.
Ecuador, Tunguragua, Ambato
Wild
P. trisecta Mast.
National Collection - Blois - France
Wild
Subgenus Murucuja (Medic.) Mast., 1871
P. murucuja L.
National Collection - Blois - France
Wild
P. tulae Urban
National Collection - Blois - France
Wild
Subgenus Pseudomurucuja (Harms.) Killip
P. perfoliata L.
National Collection - Blois - France
Wild
Subgenus Psilanthus (DC.) Killip, 1938
P. trinervia (Juss.) Poir.
Colombia, Quindio, Salento
Wild
National Collection - Blois - France
Wild
Subgenus Passiflora
Series Digitatae Killip
P. serrato-digitata L.
Series Incarnatae
P. cincinnata Mast.
Brasil
Wild
P. edulis f. edulis Sims
Colombia, Cauca, Timbio
Home garden
P. edulis f. edulis Sims
Colombia, Antioquia, Fredonia
Wild
P. edulis f. edulis Sims
Colombia, Cundinamarca, San Juan de Río Seco
Home garden
P. edulis f. flavicarpa Degener
Brasil, Araguari
Cultivated
P. edulis f. flavicarpa Degener
Colombia, Amazonas, Leticia
Home garden
P. edulis f. flavicarpa Degener
Ecuador, Guayas, Guayaquil
Cultivated
P. edulis f. flavicarpa Degener
Perú, Santa Vilma
Cultivated
P. edulis f. flavicarpa Degener
Colombia, Chocó, Quibdo
Cultivated
P. edulis f. flavicarpa Degener
Colombia, Valle del Cauca, La Unión
Cultivated
P. edulis f. flavicarpa Degener
Brasil, National Collection - Blois - France
Wild
P. incarnata L.
USA, Florida, Miami
Home garden
P. incarnata L.
USA, National Collection - Blois - France
Wild
Series Kermesinae
P. edmundoi Sacco
Brasil, National Collection - Blois - France
Wild
P. kermesina Link & Otto
National Collection - Blois - France
Wild
P. lehmanni Mast
Colombia, Quindio, Calarca
Wild
P. lehmanni Mast
Colombia, Caldas, Manizales
Wild
Wild
P. loefgrenii Vitta
Brasil, National Collection - Blois - France
P. miersii Mast.
National Collection - Blois - France
Wild
P. smithii Killip
Colombia, Tolima, Fresno
Wild
P. smithii Killip
Colombia, Tolima, Ibagué
Wild
P. trisulca Mast.
National Collection - Blois - France
Wild
Series Laurifoliae
P. acuminata DC.
National Collection - Blois - France
Wild
P. ambigua Hermsl.
Colombia, Belgian National Botanic Garden - Meise
Wild
P. carinata
National Collection - Blois - France
Wild
P. crenata Feuillet & Cremers
National Collection - Blois - France
Wild
P. fernandezii L. K. Escobar
Bolivia, National Collection - Blois - France
Wild
Wild
P. gabrielliana Vanderplank
Guiana French, National Collection - Blois - France
P. laurifolia L.
Guiana French, National Collection - Blois - France
Wild
P. nigradenia Rusby
National Collection - Blois - France
Wild
P. nitida Kunth
National Collection - Blois - France
Wild
P. nitida Kunth
Colombia, Chocó, Quibdo
Home garden
P. odontophylla Harms ex Glaz.
Brasil, National Collection - Blois - France
Wild
P. popenovii Killip
Colombia, Cauca, El Tambo
Home garden
P. popenovii Killip
Colombia, Nariño, Chachagui
Home garden
P. riparia Mart. ex Mast.
National Collection - Blois - France
Wild
P. rufostipulata Feuillet
Guiana French, Belgian National Botanic Garden - Meise
Wild
Series Lobatae
160
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
P. amethystina Mikan
National Collection - Blois - France
Wild
P. caerulea L.
France, Herault, Montpellier - (introduced)
Home garden
P. caerulea L.
Brasil, National Collection - Blois - France
Wild
P. eichleriana Mast.
National Collection - Blois - France
Wild
P. elegans Mast.
National Collection - Blois - France
Wild
P. garckei Mast.
Guiana French, Belgian National Botanic Garden - Meise
Wild
Wild
P. garckei Mast.
Guiana French, National Collection - Blois - France
P. gibertii N. E. Brown
National Collection - Blois - France
Wild
P. gritensis Karts.
National Collection - Blois - France
Wild
P. mooreana Hook.
National Collection - Blois - France
Wild
P. pallens Poepp. ex Mast.
National Collection - Blois - France
Wild
P. sprucei Mast.
National Collection - Blois - France
Wild
P. stipulata Aubl.
National Collection - Blois - France
Wild
P. subpeltata Ortega
National Collection - Blois - France
Wild
P. tucumanensis Hook.
National Collection - Blois - France
Wild
P. urubicensis Cervi
National Collection - Blois - France
Wild
Series Marginatae
P. marginata Mast.
National Collection - Blois - France
Wild
Series Menispermifolia
P. crassifolia Killip
National Collection - Blois - France
Wild
P. menispermifolia Kunth
Colombia, Tolima, Ibagué
Wild
P. menispermifolia Kunth
National Collection - Blois - France
Wild
P. nephrodes Mast.
Bolivia, National Collection - Blois - France
Wild
P. reitzii Sacco
National Collection - Blois - France
Wild
Series Quadrangulares
P. alata Curtis
Brasil
Cultivated
P. alata Curtis
Brasil, National Collection - Blois - France
Wild
P. quadrangularis L.
Colombia, Huila, Paicol
Cultivated
Series Serratifoliae
P. bahiensis Klotzsch
Brasil, Bahia, Salvador
Wild
P. serratifoliaL.
National Collection - Blois - France
Wild
Series Simplicifoliae
P. dispar Killip
National Collection - Blois - France
Wild
P. galbana Mast.
Guiana French, National Collection - Blois - France
Wild
P. mapiriensis Harms
Guiana French, National Collection - Blois - France
Wild
P. oerstedii Mast.
Colombia, Tolima, Ibagué
Wild
P. oerstedii Mast.
Colombia, Cauca, Popayán
Wild
P. oerstedii var. choconiana (S. Watson) Killip
National Collection - Blois - France
Wild
P. subrotunda Mast.
National Collection - Blois - France
Wild
Series Tiliifoliae
P. ligularis Juss.
Colombia, Nariño, Ipiales
Cultivated
P. ligularis Juss.
Colombia, Risaralda, Marsella
Cultivated
P. ligularis Juss.
Colombia, Caldas, Manizales
Wild
P. ligularis Juss.
Colombia, Quindio, Calarca
Home garden
P. maliformis L.
Colombia, Santander, Barichara
Home garden
P. maliformis L.
Colombia, Antioquia, Fredonia
Home garden
P. maliformis L.
Colombia, Huila, La Plata
Cultivated
P. multiformis Jacq.
Colombia, Norte de Santander, Ocaña
Home garden
P. palenquensis Holm-Niels. & Lawesson
Colombia, Chocó, Quibdo
Cultivated
P. seemannii Griseb
National Collection - Blois - France
Wild
P. serrulata Jacq.
Colombia, Magdalena, Plato
Wild
P. tiliifolia L.
Colombia, Quindio, Filandia
Wild
P. tiliifolia L.
Colombia, Valle del Cauca, El Cerrito
Wild
Subgenus Tacsonia (Juss.) Tr. & Planch, 1873
161
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
Section Colombiana
Series Colombiana
P. adulterina L. f.
Colombia, Boyacá, Duitama
P. adulterina L. f.
Colombia, National Collection - Blois - France
Wild
P. lanata (Juss.)
Colombia, Boyacá, Duitama
Wild
Series Leptomischae
P. ampullacea (Mast.) Harms
National Collection - Blois - France
Wild
P. antioquiensis Karst.
Colombia, Caldas, Villamaria
Wild
P. antioquiensis Karst.
Colombia, Antioquia, Santa Rosa de Osos
Wild
P. antioquiensis Karst.
Colombia, Caldas, Manizales
Wild
P. flexipes Triana & Planch
Colombia, Quindio, Salento
Wild
P. tenerifensis L. K. Escobar
Colombia, Valle del Cauca, El Cerrito
Wild
Series Quindensae
P. linearistipula L. K. Escobar
Colombia, Caldas, Manizales
Wild
Section Bracteogamma
P. cumbalensis var. cumbalensis (Karst.) Harms
Ecuador
P. cumbalensis var. goudotiana (Tr. & Planch.) L.K.
Escobar
Colombia, Boyacá, Duitama
Wild
Wild
Wild
P. luzmarina Jorgensen
Ecuador, Loja, Loja
Wild
P. tarminiana Coppens & Barney
Colombia, Caldas, Villamaria
Home garden
P. tarminiana Coppens & Barney
Perú, Huancavelica
Home garden
P. tarminiana Coppens & Barney
Venezuela, Táchira
Home garden
P. tarminiana Coppens & Barney
Venezuela, Tachira
Home garden
P. tarminiana Coppens & Barney
Colombia, Cauca, Slivia
Wild
P. tripartita var. mollissima (Kunth) Holm-Niel & Jorg
Colombia, Boyacá, Nuevo Colón
Cultivated
P. tripartita var. mollissima (Kunth) Holm-Niel & Jorg
Colombia, Cundinamarca, Tequendama
Wild
P. tripartita var. mollissima (Kunth) Holm-Niel & Jorg
Venezuela, Táchira, Villa Paez, Betania
Home garden
P. parritae (Mast.) L. H. Bailey
Colombia, Tolima, Herveo
Wild
Section Poggendorffia
P. pinnatistipula Cav.
Ecuador, Tunguragua, Ambato
Wild
Section Parritana Escobar
Section Tacsonia
P. matthewssii (Mast.) Killip
Ecuador
Wild
P. mixta L. f.
Colombia, Valle del Cauca, El Cerrito
Wild
P. mixta L. f.
Colombia, Cauca, Totoró
Wild
Subgenus Tacsonioides (DC.) Killip, 1938
P. mendoncaei Harms
National Collection - Blois - France
Wild
P. reflexiflora Cav.
National Collection - Blois - France
Wild
P. umbilicata (Griseb.) Harms
National Collection - Blois - France
Wild
Colombia, Salento, Quindío
Wild
Subgenus Tryphostemmatoides (Harms) Killip, 1938
P. gracillima Killip
V.1.3.2. DNA Extraction and PCR-RFLP analyses
Total genomic DNA was extracted from frozen leaves following the protocol of Doyle &
Doyle (1987) with minor modifications and purified on anion exchange micro-columns
(Qiagen). All samples were qualified on a 0.8% agarose gel, in 1x TBE buffer and
electrophoresed at 110 V during 1 hour and visualized under UV light.
162
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
Two cpDNA regions, intergeneric spacer (psbC- trnS and trnS - trnfM), and two mtDNA
introns (nad4-1/2 and nad1-B/C), were amplified using PCR and the universal primer
pairs indicated by Demesure et al. (1995) and evaluated in the subgenus Passiflora by
Vargas (2000). The PCR products were subsequently digested with six endonucleases
selected by Varón (2000) (Hae III, Hha I, Hinf I, Hpa II, Taq I, and Rsa I - BioLabs©;
Table 2). The PCR reaction mix contained 25 ng of template DNA, 10 mM Tris-HCl, 50
mM KCL, 1.5 mM MgCl2, 0.15 µM of each primer, 2.5 µM of each dNTP and 10 units
of Taq polymerase, in a total volume of 50 µl. PCR-amplification was performed at 95°C
for 4 min for initial denaturation, followed by 35 cycles at 94°C for 30 s, 57°C for 1 min,
72°C for 3 min, and was terminated by 10 min at 72°C. To confirm successful
amplification and to determine the size of amplified fragments, 8 µl of PCR products
were separated by electrophoresis in a 1% agarose gel, in 1 x TBE buffer. The DNA
fragments were visualized by UV fluorescence after staining with ethidium bromide. The
approximate product size was calculated by comparison of the migration distance of the
PCR product with a DNA marker (1-kb ladder).
Ten microliters of each PCR product was restricted in a volume of 20 µl containing 5
units of restriction enzyme for 3 hours at 37°C, according to the manufacturer’s
procedures. Restriction fragments were separated on 1.2% agarose gels containing
ethidium bromide in 1x Trisborate EDTA (TBE), run at 110 V for 3 hours, and visualized
under UV light. Figure 1 shows the general schema of the PCR-RFLP markers.
Table 2. DNA sequence and type of primer pairs used in the present study by Demesure et al. (1995).
Primers
psbC [psII 44 kd protein]
trnS [tRNA-Ser (UGA)]
trnS [tRNA – Ser (UGA)]
trnfM [tRNA- fMet (CAU)]
nad4 exon 1
nad4 exon 2
nad1 exon B
nad1 exon C
Code
PC1 – PC2
Sequence (5’- 3’)
GGTCGTGACCAAGAAACCAC
GGTTCGAATCCCTCTCTCTC
GAGAGAGAGGCATTCGAACC
CATAACCTTGAGGTCACGGG
CAGTGGGTTGGTCTGGTATG
TCATATGGGCTACTGAGGAG
GCATTACGATCTGCAGCTCA
GGAGCTCGATTAGTTTCTGC
TS1 – TS2
N41 – N42
N1B – N1C
163
Organelle - DNA type
Chloroplast
Chloroplast
Mitochondrial
Mitochondrial
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
V.1.3.3. Data analysis
Most restriction patterns were impossible to interpret in terms of mutations, because of
the high number of bands. Instead, these were scored for presence (1) or absence (0) and
used to identify chloroplast and mitochondria haplotypes, generating a binary data matrix
for the subsequent phenetic analyses. The Sokal & Michener (1958) coefficient of genetic
similarity was calculated for each pair of haplotypes. The similarity matrix was employed
in a principal co-ordinate analysis (PCO) and the construction of a phenogram by the
neighbor-joining method (Saitou & Nei, 1987), calculating bootstrap values from 1000
replicates. These analyses were performed using the DARwin 5.0 software (Perrier et al.,
2003).
A/A
R R
B/B
A/B
R R R
R R R
DNA
PCR + Digestion with R + gel electrophoresis
Figure 1. General schema of the PCR-RFLP markers (CAPS).
164
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
V.1.4. Results
V.1.4.1. PCR amplification
All four primer pairs used in the present study successfully amplified the corresponding
cpDNA and mtDNA regions in 151 species of genus Passiflora and four species of
Adenia, Barteria and Smeathmannia, producing fragment whose length varied from 820
to 2,040 bp (Table 3). Polymorphism for chloroplast fragment (PC1-PC2) distinguishes
P. tripartita var. mollissima, P. antioquiensis, P. pinnatistipula and P. murucuja from all
other species. Four different fragment lengths were observed for TS1-TS2, two for
subgenus Decaloba, with a difference between Australian and American species, one for
three species of subgenus Passiflora series Laurifoliae (P. popenovii, P.nitida and
P. laurifolia), and one for all the other species. Eight fragment lengths are observed for
the mitochondrial N1B-N1C, discriminating P. coriacae, P. multiflora, P. perfoliata and
all the species of subgenus Tacsonia. For the N41-N42 region, only one fragment was
detected with a weak degree of amplification.
Table 3. Numbers of haplotypes and fragments for each combination primer/ enzyme.
Primers
Size of amplification
products (bp)
Degree
amplification
Hinf I
Rsa I
Hpa II
Hae III
Hha I
Taq I
Total number
haplotypes / fragments
PC1 – PC2
TS1 – TS2
N41 – N42
N1B – N1C
1875-1896-1930-2000
820-1200-1250-1345-14001470-1500-1550-1650-16931735-1780-1810
2036
1500-1580-1605-1690-18851910-2000-2040
strong
13 / 17
2 / 4*
19 / 18
29 / 34
27 / 34
15 / 16
strong
60 / 28
34 / 26
32 / 46
10 / 11
13 / 16
30 /22
weak
27 /22
34 / 32
48 / 59
36 / 31
50 / 47
22 / 29*
good
26 / 29
17 / 21
27 / 22
20 / 22
16 / 16
12 / 31
* not interpretable.
V.1.4.2. Restriction analysis
All twenty-four fragment/enzyme combinations revealed polymorphism, but restriction
profiles could not be read in two cases [(N41-N42/TaqI) and (PC1PC2/RsaI)]. A total of
614 fragments were scored, of which 93% were found to be polymorphic in the sample
(Table 3). All the polymorphism was due to insertion-deletion (indel) mutations. Because
of the high number of fragments in relation to the sample size (213 individuals), banding
165
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
patterns could not be interpreted in terms of particular mutations. Figure 2 presents
examples of the interspecific variation for cp/mtDNA.
The PCR-RFLP analysis of the chloroplast regions showed polymorphism with all
enzymes, resulting in 268 interpretable polymorphic fragments for the 11
fragment/enzyme combinations. The 11 mitochondrial fragment/enzyme combinations
exhibited a high polymorphism too, with 307 interpretable fragments, but the resolution
of the fragments was less clear. The combination N41-N42/HpaII was by far the most
Ladder 1kb
P. oerstedii
P. maliformis
P. smithii
P. lehmannii
P. alata
P. ambigua
P. cincinnata
P. garckei
P. popenovii
P. trisulca
P. caerulea
P. quadrangularis
P. citrifolia
P. coriacea
P. filipes
P. erytrophylla
P. suberosa
P. cuneata
P. magdalenae
P. linearistipula
P. antioquiensis
P. parritae
P. tricuspis
P. alnifolia
P. luzmarina
P. flexipes
P. mixta
P. tarminana
Ladder 1kb
polymorphic one, with 59 fragments differentiating among genera and subgenera.
12.216 bp.
3.054 bp.
2.036 bp.
1.636 bp.
1.018 bp.
506 bp.
396 bp.
344 bp.
298 bp.
220 bp.
a
TS1II))
TS1-TS2 (Hpa
(HpaII
12.216 bp.
3.054 bp.
2.036 bp.
1.636 bp.
1.018 bp.
506 bp.
396 bp.
344 bp.
298 bp.
220 bp.
b
N1B1II))
N1B1-N1C (Hpa
(HpaII
Figure 2. Interspecific variation for cpDNA (a) and mtDNA (b) among different Passifora species.
166
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
V.1.4.3. PCR-RFLP haplotypes
A total of 280 chloroplast haplotypes and 372 mitochondrial haplotypes were found
(Table 3). Their global patterns of distribution among genera and subgenera are
summarized in Table 4.
For the chloroplast, the TS1-TS2 region displayed more variation (177 haplotypes) than
the PC1-PC2 region (103 haplotypes). Chlorotypes clearly distinguished subgenus
Decaloba from the rest of the sample, as its species exhibited two or three specific
haplotype patterns in all fragment/enzyme combinations. Subgenus Astrophea showed
specific chlorotypes after the restriction of the TS1-TS2 region with Rsa I and Taq I.
Subgenus Distephana was differentiated by the TS1-TS2/Taq I combination. Subgenera
Dysosmia, Manicata and Tacsonia shared haplotypes for many combinations, so they
cannot be clearly distinguished from each other. Subgenus Tacsonia appeared highly
uniform for most combinations. Within subgenus Passiflora, most species of series
Laurifoliae presented specific restriction patterns of the TS1-TS2 fragment (Hha I, Rsa I,
and Taq I. Several combinations (Hinf I and Taq I) allowed distinguishing another
morphological group, the typical representatives of series Tiliifoliae (P. ligularis,
P. palenquensis, P. tiliifolia, P. seemannii). The outgroup (Adenia, Barteria and
Smeathmannia) showed specific chlorotypes in most combinations. When this was not
the case, they usually showed similarities with species of subgenus Passiflora.
A higher number of mtDNA haplotypes were observed, 245 for the N41-N42 region and
127 for the N1B-N1C region. Subgenus Decaloba again displayed many specific
haplotypes, however it shared several mitotypes with Barteria and Smeathmannia. All
accessions of subgenus Tacsonia but four were clearly differentiated by highly specific
and uniform mitotypes for the N1B-N1C regions. The four exceptions were Ecuadorian
accessions of P. pinnatistipula, P. matthewsii, P. luzmarina and P. cumbalensis, sharing
most restriction patterns with subgenus Passiflora. Subgenus Murucuja was
differentiated by the N1B-B1C/Hha I combination.
167
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
In most species represented by several individuals, intraspecific variation was observed
for both cpDNA and mtDNA. Although represented by two accessions only, P. foetida
and P. vitifolia displayed the highest diversity of haplotypes. Other remarkable cases are
those of P. ligularis, P. maliformis, P. alata, P. caerulea, and P. sphaerocarpa.
168
Table 4. Global distribution of the haplotypes among the genera and subgenera studied.
Primers
Hinf I
Rsa I
PC1 – PC2
Decaloba two groups
- Adenia - Barteria Smeathmannia - other
subgenera
Impossible
to interpret
TS1 – TS2
Decaloba two groups
- Passiflora four
groups - (Barteria Smeathmannia) other subgenera
N41 – N42
N1B – N1C
Hpa II
Hae III
Hha I
Taq I
Decaloba three groups Decaloba - Adenia - other subgenera
Smeathmannia - other
subgenera
Decaloba three groups Decaloba - other
subgenera
- Passiflora two
groups - Tacsonia
two groups - other
subgenera
Decaloba three groups
- Passiflora four
groups - Astrophea other subgenera
Decaloba two groups
- Passiflora three
groups - Adenia (Barteria Smeathmannia) other subgenera
Decaloba three groups
- Passiflora two
groups - other
subgenera
Decaloba two groups
- Passiflora two
groups - other
subgenera
Decaloba three groups
- Passiflora two
groups - Distephana Astrophea - Adenia (Barteria Smeathmannia) other subgenera
Decaloba two groups
- Passiflora Astrophea - (Barteria
- Smeathmannia) other subgenera
Decaloba three groups
- Passiflora two
groups - Adenia other subgenera
Decaloba two groups
Passiflora four groups
- Tacsonia two groups
- Barteria two groups
- other subgenera
Decaloba two Passiflora five groups
- Adenia - other
subgenera
Decaloba two groups
- Passiflora three
groups - Barteria two
groups - other
subgenera
Impossible
to interpret
Decaloba three groups
- Passiflora two
groups - Tacsonia two
groups - (Barteria Smeathmannia) Adenia - other
subgenera
Decaloba two groups
- Passiflora two
groups - Tacsonia (Barteria Smeathmannia) Adenia - other
subgenera
Decaloba two groups
- Passiflora two
groups - Tacsonia two
groups - (Barteria Smeathmannia) other subgenera
Decaloba two groups
- Passiflora two
groups - Tacsonia
two groups - other
subgenera
Decaloba two groups Decaloba two groups
- Passiflora two
- Tacsonia - Barteria
groups - Tacsonia two two groups - Adenia groups - Murucuja Adenia - other
subgenera
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
V.1.4.4. Principal co-ordinates analysis
The first two axes of the PCO on cpDNA data accounted for 59% of the total variation
(Figure 3). They allowed visualizing a strong structure in the diversity under study, as the
genera Adenia, Barteria, and Passiflora subgenera Astrophea, Calopathanthus,
Dysosmia,
Distephana,
Manicata,
Passiflora,
Tacsonia,
Tacsonioides
and
Tryphostemmatoides, and P. deidamioides occupy the left half of the principal plan,
while the species of Passiflora subgenera Apodogyne, Decaloba, Murucuja,
Pseudomurucuja and Psilanthus form a very well separated group, placed in a quite
extreme position on the right, only the accessions of Smeathmannia and P. lancetillensis
(subgenus Deidamioides) taking intermediate positions. Within genus Passiflora, only
subgenus Astrophea is individualized. The subgenera on the right side form a wide group
that we shall call the “Decaloba group”, by analogy with the “Decaloba clade” obtained
in previous molecular work (Muschner et al., 2003; Yockteng, 2003; Yockteng & Nadot,
2004, Hansen et al., 2006) and with the subgenus Decaloba of the new classification
(Feuillet & MacDougal, 2003). In the same way, the major group on the left will be
called the “Passiflora group”. In the latter, only a small group dominated by species of
subgenus Passiflora series Laurifoliae are separated from a bulk comprising species of
subgenera
Tacsonia,
Manicata,
Calopathanthus,
Dysosmia,
Distephana
and
Tacsonioides. P. gracillima (subgenus Tryphostemmatoides) is positioned between the
‘Passiflora group’ and subgenus Astrophea.
V.1.4.5. Cluster analysis
The Neighbor-Joining tree obtained with cpDNA data shows three major, well-supported
clusters, within genus Passiflora (Figure 4). The first one corresponds to the ‘Passiflora
group’,
i.e.
subgenera
Calopathanthus,
Deidamioides,
Distephana,
Dysosmia,
Dysosmioides, Manicata, Passiflora, Tacsonia, and Tacsonioides, the second one to
subgenus Astrophea, the third one to the ‘Decaloba group’, i.e. most species of subgenera
Apodogyne, Decaloba, Murucuja, Pseudomurucuja and Psilanthus. P. gracillima
(subgenus Tryphostemmatoides) appears in basal position, forming a fourth cluster of its
own. The position of P. lancetillensis (subgenus Deidamioides) is undefined, as it is
170
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
placed on a long branch between the outgroup and the ‘Decaloba group’ clusters, the
outgroup itself taking an undefined position among the three Passiflora major clusters.
Figure 3. Principal co-ordinates on cpDNA data (PC1-PC2 and TS1-TS2 regions) estimated with 268
CAPS marker.
The Astrophea cluster shows no particular substructure, however accessions are logically
grouped
by
species.
Within
the
‘Decaloba
group’,
P.
karwinskii
(section
Pseudodysosmia) and the Australian species (section Eudecaloba) take a basal position.
After their separation, the ‘Decaloba group’ is split in two loose subclusters. The first one
contains small branches corresponding to particular sections as Pseudodysosmia and
Cieca or to series Auriculatae of section Decaloba (P. auriculata and P. jutasanchensis).
The second subcluster gathers species of subgenera Murucuja, Pseudomurucuja,
Psilanthus, section Pseudogranadilla of subgenus Decaloba and series Apetalae,
Miserae, Punctatae, Sexflorae, and Luteae of section Decaloba. This weak structure
171
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
related to such low-level infrageneric taxa contrasts with the complete lack of
differentiation among higher-level divisions of Killip, i.e. the subgenera constituting the
‘Decaloba group’. Thus, P. trinervia (subgenus Psilanthus), P. murucuja (subgenus
Murucuja), P. tulae (subgenus Murucuja), and P. perfoliata (subgenus Pseudomurucuja)
are well integrated among species of subgenus Decaloba. In addition, different
accessions from a same species do not cluster close together, as in P. bicornis,
P. magdalenae, P. biflora and P. tricuspis. Within the ‘Passiflora group’, resolution is
very poor, and the substructure very weak. Branches are short and not supported. The
only clear subcluster roughly corresponds to the small division of the ‘Passiflora group’
in the PCO principal plane. Species from a same subgenus do not tend to form even loose
subclusters, with the partial exception of tacsos that show a higher uniformity. Even
intraspecific variation can be compared in many cases to interspecific variation.
172
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
Figure 4. Phenogram derived from on cpDNA (PC1-PC2 and TS1-TS2 regions) data illustrating the
distribution of the different Passiflora subgenera studied. Boostrap values above 20% are indicated under
branches.
173
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
Figure 4a. Cluster analysis on cpDNA data, ‘Decaloba group’, subgenera Astrophea and Tryphostemmatoides, and outgroup genera.
174
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
Figure 4b. Cluster analysis on cpDNA data, ‘Passiflora group’. Subgenus Tacsonia in clear-brown.
175
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
The dendrogram obtained with mtDNA data separates four poorly to moderately
supported clusters (Figure 5). As for cpDNA, a first large cluster corresponds to the
‘Decaloba group’. Its accessions are grouped by species, with the only exception of
P. adenopoda. Again the Australian P. aurantia/P. herbertiana are paired, as is the case
for P. suberosa and P. coriacea. The numerous accessions of section Decaloba series
Punctatae and Miserae tend to form a subcluster, while species of subgenera Apodogyne,
Murucuja, Pseudomurucuja and Psilanthus are diluted in subgenus Decaloba. The other
major clusters are different from those evidenced by cpDNA data.
The ‘Passiflora group’ is split among three clusters. The largest one includes a subcluster
composed of all species of subgenus Astrophea, but two. P. gracillima (subgenus
Tryphostemmatoides) is basal to this Astrophea subgroup. Within this wide PassifloraAstrophea cluster, accessions tend to cluster by species, although strong divergences may
appear within particular species, as is the case for P. alata and P. vitifolia. Some loose
subclusters can be interpreted, as one dominated by part of the series Laurifoliae, a
branch grouping the most typical species of series Tiliifoliae (P. ligularis,
P. palenquensis, P. seemannii and P. tiliifolia), a branch grouping four species of
subgenus Distephana, and a branch bearing three Ecuadorian and one southern
Colombian accessions of subgenus Tacsonia. On the other hand, the most typical
representatives of series Incarnatae, i.e. the two forms of P. edulis and P. incarnata, are
clearly separated on the tree. Another surprising splitting is that of the morphologically
uniform series Quandrangulares, with the divergence of P. alata and P. quadrangularis.
A third important cluster is consistently composed by the majority of species from series
Menispermifoliae, Kermesinae, and Lobatae. However, at lower levels, this
morphological consistency is lost, as indicated by the separation of the very similar
P. lehmannii, P. trisulca and P. smithii. Even more surprising is the separation of the
morphologically very close P. gibertii and P. subpeltata, the latter on a smaller, distinct
cluster, together with two other representatives of series Lobatae.
176
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
The fourth subcluster is very uniform, as it includes all species of subgenus Tacsonia,
except the four Ecuadorian accessions mentioned above and an accession of
P. pinnatistipula, also from Ecuador, which appears associated with the morphologically
very similar P. macropoda in the Passiflora-Astrophea cluster.
A few species constitute a cluster of their own. Among them, we find again
P. deidamioides, (subgenus Deidamioides), but this time at the base of the Decaloba
cluster, P. umbilicata (subgenus Tacsonioides), P. serratifolia and P. multiformis
(subgenus Passiflora), and P. haematostigma (subgenus Astrophea), adding to an already
difficult interpretation of the mitochondrial tree. Two other surprising informations are
the positioning of P. lancetillensis within the Decaloba cluster and the splitting of the
two varieties of P. foetida, var. hispida appearing in the Passiflora-Astrophea cluster and
var. gossypifolia appearing in the Lobatae-Menispermifoliae-Kermesinae cluster. Last but
not least, the African Passifloraceae of the sample again take an abnormal position for an
outgroup.
177
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
Figure 5. Phenogram derived from on mtDNA (N41-N42 and N1B-N1C regions) data illustrating the
distribution of the different Passiflora subgenera studied. Boostrap values above 20% are indicated under
branches.
178
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
Fugure 5a. Cluster analysis on mtDNA data, ‘Passiflora group’ and subgenera Astrophea and Tryphostemmatoides.
179
Figure 5b. Cluster analysis on mtDNA data, ‘Decaloba group’.
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
V.1.5. Discussion
V.1.5.1. Chloroplast DNA diversity
Concerning cpDNA diversity, our results are globally consistent with those previously
obtained by Muschner et al. (2003) on trnL-F sequences, by Yockteng (2003) with matK
sequences (for the best resolved branches of the tree), and Hansen et al. (2006) with trnLtrnT sequences, as we observed three major groups, clearly corresponding to the three
major clades of these phylogenetic studies and to the three major subgenera proposed by
Feuillet & MacDougal (2003). The clusters/clades corresponding to subgenus Astrophea
appear well differentiated and uniform in all studies. Within the ‘Decaloba group’, there
appears no structure corresponding to Killip’s subgenera Apodogyne, Decaloba,
Murucuja, Pseudomurucuja and Psilanthus, so their fusion into a unique subgenus is
again justified by our cpDNA data. The marginal position of the Australian species can
be interpreted by an ancient isolation. The two loose subclusters observed after their
separation, are partly paralleled by subclades in the trnL-F and trnL-T trees. In the first
one, we can recognize the association between P. coriacea and P. suberosa and, more
distantly, with P. capsularis and P. morifolia; while subclades including species of series
Organenses, Miserae and Punctatae of subgenus Decaloba and/or subgenera Murucuja
and Pseudomuruja are observed in both trees. This subdivision within the Decaloba
group is not very reliable, mostly because of the intraspecific variation that sometimes
results in the placement of accessions of the same species in different subclusters or
branches (e.g. P. adenopoda and P. bicornis). However it appears much more consistent
with the classification of subgenus Decaloba proposed by Feuillet & MacDougal (2003)
than with Killip’s, as all the species grouped in the second subcluster are classified in
supersection Decaloba - section Decaloba by the former authors.
In the three cpDNA studies, the ‘Passiflora group/clade’ is the one showing the loosest
structure and the highest number of inconsistencies between morphological and genetic
similarities. The subgenera that compose it are not differentiated as their species tend to
be widely dispersed among poorly supported subclusters/subclades, with the exception of
P. foetida (subgenus Dysosmia), whose accessions appear as a well supported sister
181
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
group of the ‘Passiflora clade’ in the matK tree of Yockteng (2003), while the
representatives of this very particular subgenus are not individualized, neither in the trnLF tree of Muschner et al. (2003) nor in our study. In our tree, the three accessions of
P. foetida and P. arida even appear very dispersed within the ‘Passiflora group’. This
observation can be generalized to most species represented by several accessions, which
underlines that intraspecific variation is of the same order as interspecific and even
intersubgeneric variation. Only two small clusters are supported by bootstrap values
exceeding 50% and minimal taxonomic interpretation. One includes all species of series
Laurifoliae of subgenus Passiflora, except P. fernandezii and P. odontophylla, plus three
species of other series, P. crassifolia (series Menispermifoliae) P. kermesina (series
Kermesinae), P. oerstedii var. choconiana (series Simplicifoliae). The other one includes
three accessions of P. tripartita var. mollissima and one of P. antioquiensis.
Killip’s subgenus Tryphostemmatoides is supported by the placement of P. gracillima in
our tree, in an independent branch, but slightly closer to subgenus Astrophea. Its close
relative, P. tryphostemmatoides, was placed among the arborescent and semi-arborescent
species of subgenus Astrophea in the matK study of Yockteng (2003) and as a sister
group of subgenus Astrophea in the nuclear ncpGS tree of Yockteng & Nadot (2004).
Thus, the three datasets converge in providing reasons to maintain these small herbaceous
vines in a distinct subgenus, surprisingly associated with the arborescent or semiarborescent species of subgenus Astrophea. Feuillet & MacDougal classified them as a
section of their fourth subgenus, Deidamioides. However, the type species of this
subgenus, i.e. P. deidamioides, is placed in the Passiflora group in our tree, which
contradicts this view. P. lancetillensis and P. microstipula, two species of subgenus
Deidamioides sensu Killip, were transferred to a supersection Pterosperma of subgenus
Decaloba in the new classification. In our study, P. lancetillensis is placed at the base of
the Decaloba group, but in a distant position, as in the phylogenetic trees of Muschner et
al. (2003) and Hansen et al. (2006). In the latter, P. microstipula was very strangely
dispersed, with one accession close to P. lancetillensis, and two accessions in the
‘Passiflora clade’, a fact that was later attributed to a case of heteroplasmy (Hansen et al.
2007). In any case, neither the contours of Killip’s subgenus Deidamioides, nor its
182
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
splitting between subgenus Deidamioides and supersection Pterosperma by Feuillet &
MacDougal (2003) are clearly supported by cpDNA data.
V.1.5.2. Mitonchondrial DNA diversity
The results on mtDNA diversity are consistent with those from cpDNA on two main
points at the intersubgeneric level. These are the clear separation of the ‘Decaloba group’
and the unexpected placement of the outgroup species, at a comparable distance. Within
the ‘Decaloba group’ cluster, we recognize several structural traits observed on cpDNA,
such as a better grouping of accessions by species, as compared to the situation in the
‘Passiflora group’, the integration of subgenera Apodogyne, Decaloba, Murucuja,
Pseudomurucuja and Psilanthus, the formation of a subcluster supporting the section
Decaloba of subgenus Decaloba in the new classification, and associations between
particular species as P. suberosa and P. coriacea or the two Australian species. A third
important convergence at the subgeneric level is the grouping of P. gracillima (subgenus
Tryphostemmatoides) with most species of subgenus Astrophea.
The splitting of the ‘Passiflora group’ and the insertion of the AstropheaTryphostemmatoides cluster in its largest division are the two most striking divergences
with all chloroplastic and nuclear DNA, as well as morphological data (except for the
separation of tacsos), and cannot be explained on the basis of any of the two subgeneric
classifications. The largest cluster resulting from the division of the ‘Passiflora group’
shows some loose clustering for species from subgenus Distephana (four of its six
representatives), series Tiliifoliae, or simply for accessions from a same species, whereas
other clear morphological groups are split, as series Quadrangulares and Incarnatae. The
most uniform cluster dissident from the ‘Passiflora group’ is constituted by the bulk of
subgenus Tacsonia species. The five other accessions of this subgenus, placed in the
largest ‘Passiflora group’ subcluster, share a common recent Ecuadorian or southernColombian origin, constituting a particular case of geographic structure in the data.
Interestingly, P. umbillicata, of the subgenus Tacsonioides, also characterized by a
relatively long hypanthium, is placed at the base of the Tacsonia subcluster, while
P. manicata is placed in the larger Passiflora-Astrophea cluster. The third ‘Passiflora
183
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
group’ cluster seems less easy to interpret, however it integrates remarkably the series
Kermesinae, Simplicifoliae, Lobatae, and Menispermifoliae. All these series are
differentiated in Killip’s key by their foliaceous, semi-ovate to semi-oblong stipules,
“attached on one side above base, hence often appearing reniform”. As for the Tacsonia
cluster, some representatives of these series are placed in the larger Passiflora-Astrophea
cluster: P. galbana, P. mapiriensis and P. subrotunda (Simplicifoliae), P. elegans,
P. tucumanensis (Lobatae), P. reitzii (Menispermifoliae). In addition, three species of
series Lobatae, P. mooreana, P. pallens and P. subpeltata form a small independent
cluster.
In
conclusion,
the
series
Kermesinae,
Simplicifoliae,
Lobatae,
and
Menispermifoliae can be compared to those of subgenus Tacsonia for their relative
divergence from the bulk of the ‘Passiflora group’, however they show a much higher
differentiation, both between divergent clusters and within their two specific clusters. The
presence of P. racemosa, the unique species of Killip’s subgenus Calopathanthus, among
the species of these four series, is not really surprising, considering the shape of its
stipules. In contrast, the presence of one of the accessions of P. foetida in this same
cluster is very surprising.
As in the cpDNA tree, the relative distances between genus Passiflora, or its main
divisions, and the outgroup, as well as the placement of the latter, do not provide any
support to the monophyly of the genus.
V.1.5.3. Divergences in the evolutions of chloroplast and mitochondrial genomes
The analyses of chloroplastic and mitochondrial fragments gave very different pictures
on the genetic structure of genus Passiflora. Differences appear at all levels, in the
position of the outgroup, the relative position of four subgenera, and the relationships
between species.
According to cpDNA data, the three genera of the outgroup are placed near the Decaloba
cluster, giving more emphasis to its separation from the two other subgenera. In the
mtDNA tree, they are placed between divisions of the ‘Passiflora group’. In addition,
184
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
while Adenia appears more closely related to Smeathmannia for the chloroplast
sequences, it appears more closely related to Barteria for the mitochondrial sequences.
The divergences between both datasets have opposite consequences for Astrophea and
Tacsonia, the former being clearly differentiated by cpDNA but not so for mtDNA, and
the latter being clearly differentiated by mtDNA and diluted in the ‘Passiflora group’ in
the cpDNA tree. Subgenus Tryphostemmatoides appears associated with subgenus
Astrophea in both datasets. Its independence in the cpDNA tree is blurred in the mtDNA
tree as both get integrated in the ‘Passiflora group’. The fourth case concerns the small
subgenus Deidamioides. It is more complex as its two representatives are separated in
both analyses, P. lancetillensis appearing in basal position close to the ‘Decaloba group’,
while its type species, P. deidamioides, is placed in the ‘Passiflora group’ in the cpDNA
tree, but at the base of the ‘Decaloba group’ in the mtDNA tree.
At the infrasubgeneric level, clustering of chlorotypes and haplotypes appear globally
consistent for the ‘Decaloba group’, while we found replicas of the intersubgeneric
discrepancies within the ‘Passiflora group’. Thus series Laurifoliae and Incarnatae tend
to differentiate in the cpDNA tree, but not in the mtDNA tree, while the reverse is true,
and clearer, for series Tiliifoliae and the Kermesinae-Lobatae-MenispermifoliaeSimplicifoliae series.
The divergence in the information obtained from chloroplast and mitochondrial genomes
could be due to differences in their rate of evolution and mode of transmission. While
most phylogenetic studies are based on the chloroplast genome because of its maternal
transmission and slow evolution, as compared to the nuclear and mitochondrial genomes,
which makes them less sensitive to hybridization events, this assumption has been
contradicted in a significant number of cases. In their review of the question, Harris and
Ingram had already shown in 1991 that mutation and hybridization/introgression may
generate considerable intraspecific variation, constituting a potential problem for
cpDNA-based phylogenetic reconstructions. According to their revision, 27% of the
families and 21% of the genera show potential for biparental inheritance, and 23% of the
185
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
species show only biparental inheritance, suggesting the existence of a continuum rather
than an alternative mode of transmission. A few studies have also shown the possibility
of intraindividual variation (heteroplasmy), resulting from biparental transmission, as in
Pelargonium (Metzlaff et al., 1981), Oryza sativa (Moon et al., 1987), Medicago (Lee et
al., 1988), Musa (Fauré et al., 1994), Cucumis (Harvey et al., 1998) and Actinidia (Chat
et al., 2004). In Actinidia, where transmission of chloroplasts is paternal while that of
mitochondria is maternal, chloroplastic phylogenetic information provides clear
evidences of conflicts with morphological classification and striking incongruences have
been observed between maternal and paternal phylogenies, evidencing reticulate
evolution related to frequent hybridization events, sometimes involving distant species
(Chat et al., 2004). Passiflora probably constitutes a very similar case, where reticulate
evolution could account for the fast radiation in the genus, as well as the lack of clear
morphological discontinuities at subgeneric levels, while variable patterns in transmission
and evolution of organellar genomes explain striking incongruences between different
datasets, reflecting multiple origins of species.
Indeed, since the observations of Corriveau & Coleman (1988) led them to suspect
biparental plastid transmission in Passiflora, several molecular studies have pointed to a
high frequency of paternal or biparental inheritance of the chloroplast genome in the
genus. Studying a case of genome-plastome incompatibility in Passiflora hybrids,
Mráček (2005) showed that biparental transmission of chloroplasts resulted in
heteroplasmy of the whole plant, down to the single leaf level, and perhaps even the
single cell level. The case of heteroplasmy reported by Hansen et al. (2007) involves two
“extremely divergent chloroplast types” in a single individual. All species where
biparental and/or paternal transmission was detected belong to the ‘Passiflora group’
(including subgenus Dysosmia), while the only cases studied in subgenus Decaloba
evidenced maternal transmission (Do et al., 1992; Mráček, 2005; Muschner et al., 2006)
or predominantly maternal transmission (Hansen et al., 2007), which would then explain
lower intraspecific variation in the latter. More studies are needed to ascertain whether
this was coincidence or one more fundamental difference between the ‘Decaloba group’
186
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
and the ‘Passiflora group’. Concerning mitochondrial inheritance, mtDNA was
maternally transmitted in the five hybrids studied by Muschner et al. (2006).
For Passifloraceae phylogenetic studies, the consequences of paternal/biparental
transmission of chloroplasts would include all the most severe problems anticipated by
Harris & Ingram (1991), i.e. high intraspecific and intraindividual variation, random
survivorship of lineages from a polymorphic progenitor (lineage sorting), reticulate
evolution of the chloroplast genome through hybridization and introgression, all factors
that, interfering with interspecific genetic differentiation, may explain the confuse
situation, particularly in the ‘Passiflora group’. The differentiation among the major
group/clades (‘Passiflora’, ‘Decaloba’, Astrophea) would limit genetic exchanges
between them, explaining the distances observed in most studies. If the maternal
transmission of mtDNA observed by Muschner et al. (2006) were further confirmed, this
genome would become essential in the elucidation of the relations between Passiflora
species. However, the numerous cases of wide intraspecific variation for mtDNA in our
data suggest that there is no such clearcut difference in the transmission of the two
organellar genomes.
V.1.5.4. Diversity of organellar genomes and Passiflora systematics
Obviously, the central expectation in the recent molecular studies of Passiflora was that
they would give key elements for the understanding of its evolution and the definition of
its subgeneric divisions, compensating for the pitfalls in the morphological diversity
analysis. What they have produced instead is a series of divergent pictures of Passiflora
diversity, one for each genome, plus one from morphological diversity, while providing
strong evidence for reticulate evolution in the genus and, very likely, the whole family.
The divergences between corresponding datasets are such that they disqualify any intent
of deriving a consensual picture by simply piling up these datasets in a global analysis. In
other words, there is no basis for a consensus tree. All we can do is trying to draw the
clearest lessons from the different analyses.
187
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
While all inferior divisions of the two competing taxonomical treatments only appear
occasionally supported by molecular and morphological data, revealing our general
tendency to overclassification, the three main divisions of the new classification proposed
by Feuillet & MacDougal appear clearly in all molecular studies. The situation for
smaller clades is less clear. The fourth subgenus, Deidamioides, is not supported, while
our data confirm the statement of Yockteng & Nadot (2004) supporting the persistence of
the small subgenus Tryphostemmatoides.
All studies on cpDNA diversity have indicated a higher differentiation within the
‘Decaloba group/clade’, and a better relation between genetic and morphological
diversities. In addition, our data show a better relation between chlorotype and mitotype
diversities in this group. This may be attributed to one or several of three causes. A first
hypothesis is that this taxon is more ancient than the other major groups/clades, which
would be consistent with the fact that it is present also in Asia and Australia, while the
others are of strictly American origin. A second one would be that these mostly
herbaceous vines present a shorter generation time, contributing to a faster evolution
(Yockteng, 2003). A third likely explanation is that maternal transmission of cpDNA is
the most common case in the ‘Decaloba group’. In addition, if both genomes are mostly
transmitted in similar ways, it also explains why mtDNA and cpDNA trees are relatively
congruent only in this group.
The ‘Passiflora group/clade’, as defined by cpDNA and nuclear DNA data, appears as
the group where reticulation events have the wider impact on radiation and evolution. Its
division in three clusters in the mtDNA tree is the most difficult point to interpret. In this
division, the only logics we can see follow ecological adaptation and morphological
differentiation. The best-differentiated and most uniform cluster corresponds to subgenus
Tacsonia. This taxon is composed of species specifically adapted to the cool conditions
of Andean highlands, between 2,500 and 4,000 m.a.s.l, and pollination by the long-billed
hummingbird Ensifera ensifera. They have developed a particularly long hypanthium, so
the presence of the long-tubed P. umbillicata in a neighbor branch of the tree is not
surprising. This young group clearly constitutes the most efficient adaptative answer of
188
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
Passiflora to the rise of the Andes, as they are responsible for a species diversity peak
along the elevational gradient (see Chapter III). Their recent evolution explains their
mtDNA and morphological uniformity (see Chapter IV). The formation of a geographic
subcluster by four accessions of Tacsonia from Ecuador and southern Colombia within
the largest cluster of Passiflora, close to lowland species, indicates the possibility of
capture of mitochondrial genetic material through relatively wide hybridization, although
this appears much less frequent than exchanges of chloroplast material.
The
relative
separation
of
series
Kermesinae,
Simplicifoliae,
Lobatae,
and
Menispermifoliae shows a similar correspondence with morphological traits (stipule
shape) and ecoclimatic adaptation, although the climatic trend is less marked. Most
species represented in this cluster are vines adapted to the mild climates of Andean
hillsides, between 1,200 and 2,000 m, or of subtropical regions of southern South
America (northern Argentina to Brazilian Minas Gerais).
By contrast, the Passiflora-Astrophea-Tryphostemmatoides cluster is a composite group,
including the trees and treelets of subgenus Astrophea and all vigorous lianas of
subgenera Passiflora and Distephana, mostly originating from tropical lowlands or
Andean foothills, plus their close relatives from other climates. Thus, in the series
Incarnatae, P. incarnata is adapted to the temperate or subtropical conditions of the
southern USA, while P. edulis f. edulis is adapted to the mild conditions of southern
South America. In the series Tiliifoliae we find two high hillside species, P. tiliifolia and
P. ligularis (1,500-2,500m). Interestingly, we also find species of the reniform stipule
cluster and species that are intermediate between subgenera Passiflora and Tacsonia, as
P. manicata (subgenus Manicata), P. pinnatistipula (a tacso with a shorter tube, less
reduced corona and round fruit) and P. macropoda (very similar to the latter, with an
even shorter tube and a more complex corona). P. pinnastipula is known to cross
spontaneously with typical tacsos, although most of the resulting P. x rosea hybrids are
sterile.
189
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
Subgenera Astrophea and Tryphostemmatoides are much less divergent from the
‘Passiflora group’ than the ‘Decaloba group’. They are only slightly differentiated by
mtDNA, while their restriction profiles only diverge for the cpDNA fragment (trnStrnfM regions), which suggests that their separation was more recent than that of the
‘Decaloba group’. As their chromosome number is 2n = 24, this is consistent with the
scheme of genome evolution proposed by De Melo et al. (2001). According to these
authors, the 2n = 20 (subgenus Dysosmia) and 2n = 18 (‘Passiflora group’) evolved from
the 2n = 12 of the ‘Decaloba group’ via a duplication followed by descendent disploidy
(Figure 6). The tree proposed by De Melo et al. (2001) would be more parsimonious if
we accept that the divergence of the ‘Decaloba group’ is more ancient than that of
Adenia (2n = 12), or, in other words, if we accept that genus Passiflora is not
monophyletic. This latter possibility is consistent with the cpDNA studies of Muschner et
al. (2003), Hansen et al. (2006) and our cpDNA data, as well as the geographic
distributions of these taxa.
The question of Passiflora monophyly unavoidably imposes to remember that Decaloba
and Astrophea were once generic names and the possibility of reconsidering the relative
status of the three major clades. Morphology, biogeography, cytogenetics and possibly
differences in inheritance of plastid genomes converge with molecular data in
differentiating the ‘Decaloba group/clade’ and, to a lesser degree, clades corresponding
to the current subgenera Astrophea and Tryphostemmatoides. The future of lower-level
divisions, as sections and series, will obviously depend on the answer to the main
question. In a hypothetical simplified genus Passiflora, many problems would be solved
spontaneously. For example, Tacsonia or Distephana would very logically recover a
subgeneric status, reducing the need for lower taxonomic levels. On the other hand,
whatever the levels of the divisions in the classification, some species will remain
problematic, such as P. manicata and P. pinnatistipula, which show morphological
affinities with both Killip’s subgenera Tacsonia and Passiflora, nuclear DNA affinities
with the former (Segura et al., 2002; Yockteng & Nadot, 2004) and mtDNA affinity with
the latter.
190
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
Figure 6. Probable relationships among main haploid numbers known in Passiflora subgenera and other
Passifloraceae genera, as proposed by De Melo et al. (2001).
V.1.6. Conclusions
The clear conflicts observed between classifications based on morphological, chloroplast
DNA and mitochondrial DNA data can be explained by a combination of reticulate
evolution and biparental/paternal transmission of organellar genomes in Passiflora and
probably the whole family. This situation appears further complicated by differences in
the transmission between the chloroplastic and the mitochondrial genomes, the latter
being probably more frequently maternal. There is even a possibility that organellar
transmission differs among subgenera. In particular, the situation in subgenus Decaloba
sensu Feuillet & MacDougal should be further assessed.
This situation imposes particular caution in using molecular data for the resolution of the
many problems in Passiflora systematics and evolution. Only major divisions have been
confirmed by the molecular studies. On one hand, these tend to confirm the necessary
191
Chapter V. Chloroplast and mitochondrial DNA variation
________________________________________________________________________
simplification of Passiflora taxonomy and the strong reduction in the number of
subgenera, as undertaken by Feuillet & MacDougal (2003). On the other hand, the
monophyly of the genus is questioned by the data, a problem which must be solved
before any significant taxonomic modification. The possibility of restoring genera
Decaloba, as established by Roemer in 1846, and Astrophea, by Reichenbach in 1828
(Killip, 1938), should be considered.
Interspecific relations appear particularly confusing in the large-fruited species of the
‘Passiflora clade’, which includes all the passion fruits presenting economic importance
or breeding potential. In the absence of a strong structure, this entire group can be
considered as genetic resources for developing new passion fruit species and cultivars.
Breeders should not suspect interspecific barriers a priori. On the contrary, the ‘Decaloba
clade’ and subgenus Astrophea should not be attributed the same potential for fruit crop
development.
V.1.7. Acknowledgements
The authors wish to acknowledge Christian Houel (Passiflora National Collection, Blois,
France) and Doyle McKey (CNRS, France) for assistance in obtaining plant material. A
great part of this research has been funded by the Ministerio del Medio Ambiente from
Colombia and Cenicafé through the project CEN-303-2003 “Estudio de la diversidad de
las Passifloraceae y Caricaceae en la zona cafetera de Colombia”. The first author is
grateful to the Gines-Mera Foundation (CIAT) for financial support.
192
General discussion
General discussion
_______________________________________________________________________
1. Discussion
1.1. Biogeography and conservation
Colombia has been subject to many studies focused on inventories of plant species
groups (Gentry, 1993; Silverstone-Sopkin & Ramos, 1995; Galeano et al., 1998; Rangel,
1995, 2002). Passifloraceae have been inventoried in taxonomical works by Escobar
(1998a,b, 1989, 1990 inedited) and Hernández & Bernal (2000). As compared to the
latter, we have added new information on geographical distribution of each taxon and
extended the list to a total of 167 Passifloraceae species, from three genera and the five
biogeographic regions, with reports of 26 species new to Colombia. Our list ranks
Colombia as the country with the highest richness of Passifloraceae, followed by Brazil
with 127 species. Colombian species richness and diversity is more than twice that of
Peru and Venezuela, two countries of similar surface and latitude. Given its much smaller
area, Ecuador also presents an impressive diversity. Thus, the northern Andes of
Colombia and Ecuador clearly constitute the center of diversity for the genus Passiflora.
Escobar (1988a) had already underlined that 40% of the New World Passifloraceae are
found in the Andes. Indeed, radiation has been very active in the northern Andes, with
particular contribution of recent and fast evolving groups, such as subgenus Tacsonia,
particularly adapted to highlands in comparison to other infrageneric groups. Not
surprisingly, among the more than 41 tacso species in Colombia and Ecuador, 21 (14%)
are endemics. Colombian highlands are also rich in representatives of subgenus
Decaloba. On the whole, habitats between 1000 and 3,000 m account for only 27% of the
Colombian land area, but 81% of the species of Passifloraceae are found there. With 123
species, the Andean region concentrates the highest richness, mainly between 1000 and
2,000 m. The increase of species richness and endemism with the elevation is generally
interpreted as a result of the increasing isolation and decreasing habitat surface in high
mountain regions, leading to small, fragmented populations which are prone to speciation
(Simpson, 1975; Jørgensen et al., 1995). Another contribution to the particular species
richness in Colombia and Ecuador is that of the Pacific Coast region, continuous with the
similar highly diverse ecosystems of Central America (Chocó-Darién/Western Ecuador
194
General discussion
_______________________________________________________________________
hotspot of Myers et al., 2000), and receiving one of the highest rainfalls in the world, in
strong contrast with the conditions prevailing in the westerns Andes and coast of Peru
that are arid or semi-arid, or the drier and more contrasted climate of Venezuela.
The distribution of Passifloraceae has been drastically affected by deforestation,
principally in the Andean region, as their favorite habitats correspond to areas with a long
history of livestock and agriculture that now supports extensive plantations of coffee,
sugar cane, rice, bananas, and potatoes. Forests in the northern Andes are currently one of
the major conservation priorities on a global scale due to their fragility, biological
richness, high rates of endemism and multiple anthropogenic threats (Olson & Dinerstein,
1998). As Passifloraceae display very high species richness, endemism and extinction
risk in this area, and given their multiple ecological interactions with many organisms, as
well as their economic potential, this family should constitute both an important target of
conservation efforts and a good indicator of their success. This appears particularly
important as we have shown that the conservation of Passifloraceae and their habitat is
not taken into account by the current system of protected areas in the Colombian Andes,
due to a general lack of correspondence between the modeled distribution of
Passifloraceae diversity and these areas, as presented in Figure 8 (Chapter III). Instead,
the striking superposition of areas of high Passifloraceae diversity on certain Colombian
coffee growing zone ecotopes (Figure 9; Chapter III) points to the need of developing a
completely different approach, integrating agriculture, tourism, watershed management,
and nature conservancy at the landscape level, in a region whose conservation is of
utmost importance for the country.
1.2. Morphological and molecular diversity
As a first major point, we can underline that the major morphological divisions observed
in our study find support in the genetic studies. The cytological groups are always
validated,
with
the
clear
separation
of
subgenera
Astrophea
(n = 12),
Tryphostemmatoides and Decaloba (n = 6) between themselves and from subgenera
Passiflora,
Tacsonia,
and
Distephana
(n = 9).
Concerning
subgenus
Tryphostemmatoides, the consistency between morphological and genetic studies is clear
195
General discussion
_______________________________________________________________________
only when considering our quantitative analysis, where it is associated with subgenus
Astrophea mostly on peduncle traits (third principal component). This trait is represented
also in the qualitative descriptors, however its effect is blurred by the high number of
traits shared with subgenus Decaloba. While the comparison is difficult for subgenus
Tryphostemmatoides, it is impossible for subgenus Psilanthus, because of insufficient
data and the unlikely placement of P. sanguinolenta in the ‘Passiflora clade’ in the
ncpGS study. The two species, P. adenopoda and P. foetida, that take an intermediate
position in the general “morpho-cytological” pattern, or their close relatives, are
consistently placed in intermediate positions, in most phylogenetic studies, P. adenopoda
or P. morifolia (section Pseudodysosmia of subgenus Decaloba) appearing basal to a
general ‘Decaloba clade’ and P. foetida (subgenus Dysosmia) basal to the general
‘Passiflora clade’.
The comparison becomes more difficult at lower, infra subgeneric, levels. Subgenus
Decaloba appears better structured than the other subgenera, and shows similarities in
morphological and molecular diversity patterns, with the grouping of Killip’s sections
Punctatae and Miserae, and the differentiation of species of sections Xerogona, Cieca,
and series Auriculata and, less clearly, Sexflorae. The placement of P. adenopoda in the
different trees questions the inclusion of section Pseudodysosmia, while the structure
observed among representatives of several sections provides support to some
simplification, but not for as many fusions as those operated in the new morphological
classification of Feuillet & MacDougal (2003). In any case, more species should be
gathered in a same phenetic study before revising objectively the morphological
classification.
Within the n = 9 group, molecular and morphological data diverge partially, as studies of
DNA sequences allow the distinction of a Tacsonia-Manicata group and fail to separate
clearly subgenus Distephana, placing both of them within a Passiflora clade, while
morphological analysis supports these three subgenera at the same level of
differentiation. The fact that species of subgenera Distephana, Tacsonia and Manicata
have developed ornithophyly is obviously related to their strong morphological
196
General discussion
_______________________________________________________________________
differentiation, which does not minor the importance of their morphological separation
from subgenus Passiflora. Whether their probable evolution from a “Passiflora-like”
common ancestor justifies their inclusion in the bee-pollinated subgenus Passiflora, as
proposed in the new classification, is just the same classical question about considering
birds as dinosaurs. In the end, it seems a problem of putting more emphasis on the
adaptative forces commanding evolution or more emphasis on the genetic structure that
subtend them. Concerning subgenus Passiflora sensu Killip, no structure appears at the
interspecific level that could result in clear subdivisions into series. The study of
sequence variation for the ncpGS gene provides the only tree with reasonably wellsupported structure at this level, however several obvious abnormalities question the
robustness of the information. Our morphological observations only confirm closer
associations between the most typical representatives of some series, however the number
of contradictions with the classification and the lack of a clear hierarchy in the branch
structure point to the difficulty of the work and the risk of under- or overclassification,
leading to chose between a limited number of poorly supported series or a great number
of poorly represented series. Similarly, the structure of the Tacsonia-Manicata branch
does not support clearly sections and series in subgenus Tacsonia, however it allows
differentiation between two groups of tacsos, one corresponding to common species that
probably have their center of diversity in Ecuador, as is obvious for P. cumbalensis,
P. luzmarina and P. matthewsii, and very likely for P. mixta, P. tripartita and
P. tarminiana (Segura et al., 2005), and another cluster only including species endemic to
Colombia, with a slight but clear differentiation related to extreme variation for peduncle
length.
1.3. Importance of reticulate evolution in Passiflora
Many Passifloraceae species are cross-compatible (Vanderplank, 2000; Ulmer &
MacDougal, 2004). Based on morphological observations, Escobar (1980) stated that also
in nature these interspecific hybridizations occasionally occur in areas where species
distributions overlap. In the molecular analyses presented here, several discrepancies
were observed between chloroplast and mitochondrial DNA phenograms, suggesting both
recent and ancient hybridization and introgression events and corroborating the
197
General discussion
_______________________________________________________________________
observations of Escobar (1980). These data clearly suggest that evolution in Passiflora is
reticulate rather than exclusively dichotomous and branching. Especially the origin of
P. foetida provides an excellent example of reticulate evolution and once more illustrates
the importance of hybridization in speciation within Passiflora. Such phenomena are
well-documented and considered fairly common in the evolution of plants (Van
Droogenbroeck et al., 2004; Chat et al., 2004; Vriesendorp & Bakker, 2005).
The combination of significant reticulate evolution, with a biparental/paternal inheritance
of the choloroplast genome and, although to a lesser extent, the mitochondrial genome,
particularly in the ‘Passiflora group’, is the most likely explanation to the observations of
wide intraspecific, and even intra-individual, organelle DNA variation, or conversely
morphologically distant species sharing very similar organellar genomes, indicating
captures of chloroplasts and mitochondria, and the resulting striking inconsistencies
between the ITS, matK, ncpGS and TrnL/trnT phylogenies published so far (Muschner et
al., 2003; Yockteng, 2003; Yockteng & Nadot, 2004; Hansen et al., 2006; see Annex 4ad). Taking into account these elements, the molecular phenograms presented in the
present thesis can serve as a base for further exploration of the genetic diversity and
interspecific relationships, phenotypic evolution, historical ecology and phylogeography
of the wide Passiflora diversity.
At another level, the results presented in this work are also valuable for the development
of better in situ and/or ex situ conservation strategies of Passiflora diversity. The extent
of natural hybridization occuring in the different areas of sympatry should be considered
when dealing with Passiflora genetic resources. Although hybridization has long been
recognized as an important factor in the evolution of plant species, the harmful effects of
hybridization have also led to the extinction of many populations and species (Allendorf
et al., 2001; Barton, 2001). Such effects of hybridization can be most problematic in the
situation when a rare Passiflora species comes into contact with a more abundant one.
Such cases could get more common as the abundance of several Passiflora species has
been reported to decline steadily. Thus, in Colombia, we have determined that three
Passifloraceae species can be considered extinct and 70% of the species are threatened
198
General discussion
_______________________________________________________________________
(Chapter II). Conservation programs, not only in Colombia, should be aware of such
effects in preserving the genetic integrity of Passifloraceae species in the long term.
1.4. Phylogeography
As discussed in the Chapter III, South America and especially the biodiversity hotspot
described as ‘the Tropical Andes’ have higher plant species diversity than any other
habitat on the planet. How this diversity arose is unexplained. One theory suggested that
species richness in the tropics is the result of the gradual accumulation of species over a
long geological period, with low rates of extinction in stable equatorial climates
(Stebbins, 1974). However, more recent discoveries suggest that Neotropical climates
were unstable over the past 2 million years during the Pleistocene (Whitmore & Prance,
1987). Cyclical glacial events led to periods of cooler and/or drier climate in which forest
species may have withdrawn to small refugial pockets. According to this view, the
present species diversity could be more recent, resulting from speciation through
allopatric differentiation of populations in separate refugia (Haffer, 1982). Other recent
geological phenomena that have been suggested as driving neotropical speciation are the
uplift of the northwestern Andes from about 5 million years ago and the bridging of the
Isthmus of Panama some 3.5 million years ago (Simpson & Todzia, 1990). Gentry (1989)
speculated that nearly half of the Neotropical flora might be accounted for by explosive
speciation. This rapid diversification is characteristic of plant evolution on the SouthAmerican continent since its isolation from Africa, Meso-America and AntarcticaAustralia during the late Cretaceous and early Tertiary Period, about 60 to 70 million
years ago (Burnham & Graham, 1999; Dino et al., 1999). These geographic connections
allowed high exchange of flora, according to fossil registers of many families (Taylor &
Taylor, 1993; Taylor, 1995). In the genus Passiflora this exchange is evident with the
presence
of
the
subgenus
Tetrapathea
and
the
section
and
series
Hollrungiella/Eudocaloba (of subgenus Decaloba) in Oceania and South Asia.
Taking into account all these elements, it is most likely that isolation and adaptive
radiation into ecologically extreme habitats at the time of the Andean uplift also have led
to rapid diversification and differentiation in the genus Passiflora. The relatively wide
199
General discussion
_______________________________________________________________________
size and restricted distribution of the genus Passiflora and paleobotanical data from 1720 millions of years (Miocene) provide support to this hypothesis (Dorofeev, 1963). As
illustrated by Richardson et al. (2001), molecular evidence can be useful in assessing the
validity of phylogeographical hypotheses. The sequence analyses reported by Muschner
et al. (2003), Yockteng (2003), Yockteng & Nadot (2004), Hansen et al. (2006) and our
results with cp/mtDNA, revealed only low levels of sequence divergence among
Passiflora taxa, suggesting that these have diversified recently. In particular all molecular
data place the species of the highland subgenus Tacsonia (with a cylindrical elongate
floral tube) within the ‘Passiflora group’. However, in the mtDNA phenogram, this
subgenus forms a well-supported independent group. This stronger differentiation of
mtDNA appears consistent with a faster evolution of the mitochondrial genome, as
compared to the chloroplast genome, the particular altitudinal distribution of subgenus
Tacsonia and the hypothesis of a rapid diversification and differentiation in the genus
Passiflora in relation to the uplift of the Andes.
200
Conclusions
&
future prospects
Conclusions and prospects
_______________________________________________________________________
1. Conclusions
With 167 reported species, Colombia is the country with the highest Passifloraceae
richness. This richness is concentrated in the Andean region, particularly in the
departments of Antioquia, Valle del Cauca and Cundinamarca. Comparisons with other
countries indicate that the northern Andes of Colombia and Ecuador constitute the center
of diversity for the most important genus Passiflora.
Collections of Passifloraceae have not been uniform as a consequence of difficulty of
access and/or chronic social conflict in many areas. They have been much denser in the
central coffee growing zone, Antioquia, Valle del Cauca and Cundinamarca. The
southern and northeastern Andes, and the Caribbean have been little explored. For the
lowland forests of the Pacific, the Orinoquian and the Amazonian, data are so poor that
they are misleading. Despite the resulting sampling bias, collecting parameters clearly
point to the concentration of observed Passifloraceae diversity in the Andes, and more
particularly the central coffee growing zone.
The analysis of species distribution areas shows a trend for a wider dispersion of species
occurring at low and intermediate elevations. On the contrary, narrow endemics are more
frequent among highland species.
The modeled species richness map allowed identifying nine richness spots of variable
size, three of which, located in the southern and southeastern Andes of Colombia,
correspond to collection gaps, as they were not detected in the analysis of observed
diversity. Another probable collection gap, not detected by diversity modeling,
corresponds to the Sierra Nevada de Santa Marta, an isolated mountain range with both
high diversity and endemism. The proportion of endemics living in high richness spots is
lower than the proportion of all species used for modeling, confirming the lack of relation
between diversity concentration and endemism reported in other studies. If this is further
substantiated in different groups of organisms, it could limit the application of the
202
Conclusions and prospects
_______________________________________________________________________
biodiversity hotspot concept, as the best-protected areas for diversity would not
necessarily provide protection to a high proportion of narrow endemics.
Passifloraceae diversity is not conserved by the current network of Colombian protected
areas. On the contrary, it is particularly concentrated on certain ecotopes of the coffee
growing zone, i.e. highly disturbed habitats, so any conservation effort must be integrated
in local management strategies at the landscape level. Passifloraceae may provide an
interesting indicator to evaluate the outcome of such efforts.
In the absence of a clear set of morphological criteria for discriminating at the different
hierarchic levels of the infrageneric classification of Passiflora, we have used a quite
exhaustive list of 43 quantitative and 83 qualitative descriptors. A shorter list of 32
qualitative traits, selected after analyzing variation among Killip’s subgenera, allowed
classifying our 60-species sample consistently, using a strictly phenetic approach. Eight
of the nine Killip’s subgenera represented in our sample are supported by the
morphological analysis, although subgenus Tryphostemmatoides is only supported in the
quantitative analysis. By contrast, the simplification proposed by Feuillet & MacDougal
is not clearly supported in our analyses, except for the possible inclusion of P. manicata
in subgenus Tacsonia, as this species is intermediate with subgenus Passiflora for
quantitative traits but very similar to tacsos for most qualitative traits.
Chloroplast and mitochondrial molecular trees provide clear evidence of conflicts with
morphological classifications. This suggests that the infrageneric classifications that have
been established in the past by Killip (1938), Escobar (1988a,b, 1989) and Feuillet &
MacDougal (2003) do not reflect the molecular evolutionary history of the genus
Passiflora, at least with respect to the supersections and series. Moreover, the frequent
occurrence of reticulation events in the evolution of Passiflora could explain the lack of
morphological discontinuities at subgeneric levels, while variable patterns in transmission
and evolution of organellar genomes explain striking incongruences between different
datasets, reflecting multiple origins of species.
203
Conclusions and prospects
_______________________________________________________________________
Breeding programs aimed at producing interspecific hybrids involving the cultivated
species of Passiflora should therefore focus on the species belonging to the same clade as
subgenus Passiflora. According to our results, the ‘Decaloba clade’ and subgenus
Astrophea do not constitute interesting genetic resources for passion fruit breeding.
These results constitute potentially crucial inputs for the development of a coherent
strategy for the conservation and use of these genetic resources. Studies of Passiflora
diversity in the Andean countries, and the maps presented here, will be used in future
prospecting and identifying sites for in situ conservation, and more generally guiding
government conservation strategies.
2. Future prospects
The low level of exploration in parts of the Andes, the Amazonian and the Orinoquian
raises expectations that Colombia may still harbor many unknown species. Future studies
should encompass new regions, including protected areas and current conflict zones.
Indeed, a better knowledge of this diversity, and its distribution, is urgent for the in situ
conservation of this threatened richness, targeting the conservation of these resources as
well as their habitat. Both aspects may even be combined if the genus Passiflora can be
used as an indicator of biodiversity in the Andean region, as was the objective of a
project in the coffee growing zone. Another important aspect is its direct valorization as a
germplasm resource for crop diversification programs, implying the need for a better
understanding of its morphological and genetic diversity.
Another direction that should be considered is the sequencing of regions different from
those analyzed in this study. By doing so, the value of the phenograms obtained in this
work can be assessed and further complemented. A promising source is the nuclear
ribosomal internal transcribe spacers (ITS). This region has been widely used in plant
phylogenetics because of their high rate of nucleotide substitutions and their power in
elucidating infrageneric relations (Taberlet et al., 1991; Alvarez & Wendel, 2003) Also,
204
Conclusions and prospects
_______________________________________________________________________
the ITS have already been used in Passiflora species by Muschner et al. (2003) with
success, although the size and composition of the sample limited the interpretation at the
level of subgenera. It would also be necessary to increase the number of species and in
particular to include Asian members of Passiflora as well as species of subgenera not
represented in our analysis.
Given the importance of hybridization and introgression in the evolution of the genus
Passiflora, more studies are also needed to investigate and document hybridization in
Passiflora, and its potential implications in the taxonomic problems concerning the genus
and the whole family. Detailed studies should focus on the morphologically variable taxa,
along with their close relatives in areas where they co-occur. A large number of different
methods can provide valuable information regarding hybridization, such as:
•
•
Artificial hybridizations in many species.
•
artificially produced hybrids.
Investigation and characterization of pollen viability and seed germination in
Molecular techniques to confirm the hybrid origin of some taxa, combined with
cytogenetic techniques, including chromosome counts, flow cytometry,
fluorescence in situ hybridization (FISH) and genomic in situ hybridization
(GISH).
205
Bibliography
Bibliography
________________________________________________________________________
Allendorf F.W., Leary R.F., Spruell P., and Wenburg J.K. (2001) The problems with
hybrids: setting conservation guidelines. Trends in Ecology and Evolution
16:613-622.
Alvarez I. and Wendel J.F. (2003) Ribosomal ITS sequences and plant phylogenetic
inference. Molecular Phylogenetics and Evolution 29(3): 417-434.
Apple J. and Feener D. (2001) Ant visitation of extrafloral nectaries of Passiflora: the
effects of nectary attributes and ant behavior on patterns in facultative antplant mutualisms. Oecologia 127(3): 409-416.
Aublet F. (1775) Histoire des plantes de la Guiane française. Paris : P.F. Didot, 2: 828324.
Barbosa L.V. and Vieira A.M. (1997) Meiotic behavior of passion fruit somatic hybrids,
Passiflora edulis f. flavicarpa Degener + P. amethystina Mikan. Euphytica
98: 121-127.
Barrios L., Caetano C.M., Cardoso C.I., Coppens d´Eeckenbrugge G., and Arroyave J.A.,
Olaya C. (2005) Caracterización del polen de especies de los géneros
Passiflora y Dilkea. Acta Agronómica 54(3): 1-8.
Barton N.H. (2001) The role of hybridization in evolution. Molecular Ecology 10: 551 568.
Beal P.R. (1972) Two new interspecific hybrids in the genus Passiflora. Sabrao
Newsletter 4(2): 113 115.
Beal P.R. (1975) Hybridization of Passiflora edulis Sims and P. edulis Sims f. flavicarpa
Degener. Queensland Journal of Institute of Agriculture and Animal Sciences
32 (1): 101-111.
Benson W.W., Brown K.S., and Gilbert L.E. (1976) Coevolution of plants and
herbivores: Passion vine butterflies. Evolution 29: 659-680.
Bergner P. (1995) Passionflower. Medical Herbalism 7: 13–14.
BRC (Banco de la República de Colombia) (2007) Museo de Oro de Bogotá D.C.,
Cultura Malagana. http://www.banrep.gov.co/museo/esp/s_calima.htm. Cited
25 January 2007.
Brooks T.M., Mittermier R. A., Mittermier C.G., da Fonseca G.A.B., Rylands A.B.,
Konstant W.R., Flick P., Pilgrim J., Oldfield S., Magin G., and Hilton-Taylor
C. (2002) Habitat loss and extinction in hotspot of biodiversity. Conservation
Biology 16: 909-923.
207
Bibliography
________________________________________________________________________
Bruckner C.H. and Otoni W.C. (1999) Hibridação em maracujá. In: Borém, A. (Ed.),
Hibridação artificial de plantas. UFV, Viçosa, pp. 379–399.
Büchert A. and Mogens J. (2001) The fragility of extreme specialization: Passiflora
mixta and its pollinating hummingbird Ensifera ensifera. Journal of Tropical
Ecology 17: 323-329.
Burnham R.J. and Graham A. (1999) The history of neotropical vegetation: New
developments and status. Annals of Missouri Botanical Garden 86:546-589.
Busby J.R. (1991) BIOCLIM - a bioclimate prediction system. In: Margules CR, Austin
MP (eds.) Nature conservation: cost effective biological surveys and data
analysis. CSIRO, Melbourne, Australia, pp. 4-68.
Campos T. (2001) La Curuba: Su cultivo. Bogotá, Colombia, IICA. 30pp.
Calderón E. (2005) Familia Passifloraceae. Programa de biología de la conservación,
proyecto flora amenazada. Instituto Alexander von Humboldt.
http://www.humboldt.org/ListasRojas/PASSIFLORACAE. Cited 3 abril de
2005.
Cancino O. and Hodson E. (1994) Cultivo de tejidos y micropropagación en "maracuyá"
Passiflora edulis var. flavicarpa Degener. Tablero, Revista del Convenio
Andrés Bello, 18: 81-83.
Cavanilles A.J. (1790) Monadelphia classis dissertation decen. Diss. 10: Decima
dissertatio botanica de Passiflora. Madrid: Typographia regia. 439-463.
Cenicafé (2005) Biodiversidad en la zona cafetera. http://www.cenicafe.org. Cited 25
July 2005.
Chat J., Urregui B., Petir R.J., and Nadot S. (2004) Reticulate evolution in kiwifruit
(Actinidia, actinidiaceae) identified by comparing their maternal and paternal
phylogenies. American Journal of Botany 9(15): 736-747.
Chopra R.N., Badhwar R.L., and Ghosh S. (1944) Poisonous Plants of India. Public
Service Commission, Govt of West Bengal, Calcutta, pp. 469–472.
Chopra R.N., Nayar S.L., and Chopra I.C. (1956) Glossary of Indian Medicinal Plants.
CSIR, New Delhi, India, pp. 186–187.
Coppens d’Eeckenbrugge G. (2003) Promesas de las pasifloras. Memorias del X
Seminario Nacional y IV Internacional sobre Especies Promisorias, Medellín
Octubre 29-31 de 2003. CD.
208
Bibliography
________________________________________________________________________
Coppens d`Eeckenbrugge G., Barney V., Møller-Jørgensen P., and MacDougal J.M.
(2001) Passiflora tarminiana, a New Cultivated Species of Passiflora
subgenus Tacsonia (Passifloraceae). Novon 11: 8-15.
Coppens d’Eeckenbrugge G., Segura S.D., Hodson de Jaramillo E. and Gongora G.A.
(2001) Passion Fruits. In: Charrier et al. A. (eds.), Tropical Plant Breeding.
Montpellier, France, Cirad; Enfield, U.S.A., Science Publishers, pp. 381-401.
Coppens d’Eeckenbrugge, G., Restrepo, M.T., Ocampo, J.A., Garcin, F., Olaya, C.A.,
Farfán, L.C., Vega, J., Rioux, V., Barrera, F., Petersen, J., Guzmán, L.,
Primot, S., Quispe, P., Caetano, C.M. (2002) Diversidad morfológica en
Passiflora subgéneros Tacsonia y Manicata. VIII Congreso Latinoamericano
de Botánica & II Congreso Colombiano de Botánica, Cartagena (Colombia),
October 13-18, Resúmenes, 417.
Corriveau J.L, and Coleman AW. (1988) Rapid screening method to detect potential
biparental inheritance of plastid DNA and results for over 200 angiosperm
species. American Journal of Botany 75: 1443-1458.
Clegg M.T., Gaut B.S., Learn G.H., and Morton B.R. (1994) Rates and patterns of
chloroplast DNA evolution. Proceedings of the National Academy of Sciences
of the United States of America 91(15): 6795-6801.
Croat T. (1978) Flora of Barro Colorado Island. Stanford University Press, Stanford.
943pp.
Crochemore M., Molinare H., and Colauto N. (2003) Caracterizao agromorfologica do
Maracujazeiro (Passiflora spp.). Revista Brasilera de Fruticultura 25 (1): 510.
De Candolle A.P. (1822). Passifloraceae. Prodromus Systematis Naturalis Regni
Vegetabilis 3 : 321-338. Paris : Treuttel & WÜrtz.
De Candolle A.P. (1828) Mémoires de la Société de Physique et d’Histoire Naturelle de
Geneve 1 : 434-436.
De Melo F., Cervi A., and Guerra M. (2001) Karyology and cytotaxonomy of the genus
Passiflora L. (Passifloraceae). Plant Systematics and Evolution 226: 69-84.
De Melo, F. and Guerra M. (2003) Variability of 5S and 45S rDNA sites in Passiflora L.
species with distinct base chromosome numbers. Annals of Botany 92: 309316.
De Oliviera R., Falvo M., Gomes J., Bernacci L., Carneiro M., Oliveira G., and Martinez
O. (2005) Leaf shape analysis using the multiscale Minkonwski fractial
209
Bibliography
________________________________________________________________________
dimension, a new morphometric method: a study with Passiflora
(Passifloraceae). Canadian Journal of Botany 83: 287-301.
De Wilde W.J.J.O. (1972) The indigenous Old World Passifloras. Blumea 20: 227-250.
Debouck D.G. and Libreros Ferla D. (1995) Neotropical montane forests: A fragile home
of genetic resources of wild relatives of New World crops. In Biodiversity and
Conservation of Neotropical Montane Forests, Proceedings of a Symposium,
New York Botanical Garden, 21-26 June 1993 (S.P. Churchill, H. Balslev, E.
Forero and J.L. Luteyn, eds.). New York Botanical Garden, New York, pp.
561-577.
Deginani N.B. (2001) Las especies argentinas del género Passiflora (Passifloraceae).
Darwiniana 39: 43-129.
Demesure B., Sodzi N., and Petit R.J. (1995) A set of universal primers for amplification
of polymorphic non-coding regions of mitochondrial and chloroplast DNA in
plants. Molecular Ecology 4: 129-131.
Dévalos L.M. (2001) The San Lucas mountain range in Colombia: how much
conservation is owed to the violence?. Biodiversity and Conservation 10: 6978.
Dharwan K., Dharwan S., and Sharma A. (2004) Passiflora: a review update. Journal
Ethno-Pharmacology 94: 1-23.
Dino R., Pocknall D.T., and Dettman M.E. (1999) Morphology and ultrastructure of
elater-bearing-pollen from the Albian to Cenomanian of Brazil and Ecuador:
implications for botanical affinity. Review of Paleobotany and Palynology
105: 201-235.
Do Y.Y., Shii C.T., and Huang P.L. (1992) Restriction patterns and inheritance of
chloroplast DNA in Passiflora edulis Sims and its related species. P. 10-13. In
the impact of biological research on agricultural productivity (eds.). Taipei,
Taiwan.
Dornelas M.C., Tavares J.C. de O., and Vieira M.L. (1995) Plant regeneration from
protoplast fusion in Passiflora spp. Plant Cell Reports 15: 106-110.
Dorofeev P.I. (1963) Primulaceae. In A. L. Takhtajan (ed.), Oznovij Paleontologii, 517–
518. Akademia Nauk, Moscow, Russia.
Doyle J.J., and Doyle J.L. (1987) A rapid DNA isolation procedure from small quantities
of fresh leaf tissues. Phytochemical Bull 19:11-15.
210
Bibliography
________________________________________________________________________
Drew R.A. (1991) In vitro culture of adult and juvenile bud explants of Passiflora spp.
Plant Cell Tissue and Organ Culture 26: 23-28.
Duval M.F., Buso G.S.C., Ferreira F.R., Noyer J.L., Coppens d'Eeckenbrugge G., Hamon
P., and Ferreira M.E. (2003) Relationships in Ananas and other related genera
using chloroplast DNA restriction site variation. Genome 46: 990-1004.
Endress P.K. (1994) Diversity and evolutionary biology of tropical flowers. Cambridge,
England: Cambridge University Press.
Erdtman G. (1952) Pollen morphology and plant taxonomy: Angiosperms (An
introduction to Palinology). Leiden: E.J. Brill. (Reprinted in 1986). 553p.
Escobar L.K. (1980) Interrelationships of the edible species of Passiflora centering
around Passiflora mollissima (H.B.K.) Bailey subgenus Tacsonia. Ph.D
Dissertation, University of Texas.
Escobar, L.K. (1981) Experimentos preliminares en la hibridación de especies
comestibles de Passiflora. Actualidades Biológicas 10: 103-111.
Escobar L.K. (1985) Biología reproductiva de Passiflora manicata e hibridación con la
curuba Passiflora mollissima (H.B.K.) Bailey. Actualidades. Bioliologicas. 14
(54): 111-121.
Escobar L.K. (1986) New species and varieties of Passiflora (Passifloraceae) from the
Andes of South America. Systematic Botany 11(1): 88-97.
Escobar L.K. (1987) A taxonomic revision of the varieties of Passiflora cumbalensis
(Passifloraceae). Systematic Botany 12 (2): 238-250.
Escobar, L. K. (1988a) Monografía No. 10. Passifloraceae. Passiflora. Subgéneros:
Tacsonia, Rathea, Manicata and Distephana. Bogotá, D.E., Colombia,
Universidad Nacional de Colombia.
Escobar L.K. (1988b) Passifloraceae. Flora de Colombia 10. Instituto de Ciencias
Naturales, Universidad Nacional de Colombia, 138pp.
Escobar L.K. (1988c) Novedades en Passiflora (Passifloraceae) de Colombia Mutisia
71:1-8.
Escobar L. K. (1989) A new subgenus and five new species in Passiflora (Passifloraceae)
from South America. Annals of the Missouri Botanical Garden 76: 877-885.
Escobar L.K. (1990) Two new species of Passiflora (Passifloraceae) from northern
South America. Phytologia 69 (5): 364-365.
211
Bibliography
________________________________________________________________________
Escobar L.K. (1990, inedited). Una revisión taxonómica de Passiflora subg. Astrophea
(Passifloraceae). 289 pp.
Escobar L.K. (1992) La sistemática y evolución de las Pasifloras. In: Primer simposio
internacional de Pasifloras. Universidad Nacional de Colombia sede Palmira,
Colombia, pp. 51-54.
Escobar L.K. (1994) Two new species and a key to Passiflora subg. Astrophea.
Systematic Botany 19 (2): 203-210.
Fajardo D., Angel F., Grum M., Tohmé J., Lobo M., Roca W., and Sánchez I. (1998)
Genetic variation analysis of the genus Passiflora L. using RAPD markers.
Euphytica 101: 341-347.
Faure S., Noyer J., Carreel F., Horry J.P., Bakry F., and Lanaud C. (1994) Maternal
inheritance of chloroplast genome and paternal inheritance of mitochondrial
genome in banana (Musa acuminata). Current Genetics 25: 265-269.
Ferguson M.E., Jarvis F., Stalker H.T., Williams D.E., Guarino L., Valls J.M., Pittman
R.N., Simpson C.E., and Bramel P.J. (2005) Biogeography of wild Arachis
(Leguminosae): distribution and environmental characterization. Biodiversity
and Conservation 14: 1777- 1798.
Feuillet C. (2002) A new series and three new species of Passiflora subgenus Astrophea
from Guianas. Brittonia 54: 18-29.
Feuillet C. (2004) Passiflora phellos, a new species in subgenus Passiflora
(Passifloraceae). Novon 14: 285-287.
Feuillet C. and MacDougal J.M. (2002) Checklist of recognized species names of passion
flowers. Passiflora 12 (2): 41-43.
Feuillet C. and MacDougal J.M. (2003) A new infrageneric classification of Passiflora L.
(Passifloraceae). Passiflora 13 (2): 34-38.
Fouqué A. and Fouqué R. (1980) Quelques notes sur la granadille jaune. Passiflora edulis
Sims var. flavicarpa. Fruits 35: 309-312.
Franklin J. (1995) Predictive vegetation mapping: geographic modeling of biospatial
patterns in relation to environmental gradients. Progress in Physical
Geography 19: 474-499.
Galeano G., Suarez S., and Balslev H. (1998) Vascular plant species count in a wet forest
in the Chocó area on the Pacific coast of Colombia. Biodiversity and
Conservation 7: 1563-1572.
212
Bibliography
________________________________________________________________________
García
M.T.A., Galati B.G., and Anton A.M. (2002) Microsporogenesis,
microgametogenesis and pollen morphology of Passiflora spp.
(Passifloraceae). Botanical Journal of the Linnean Society 139:383-394.
García-Barriga H. (1975) Flora medicinal de Colombia. Tomo II, Instituto de Ciencias
Naturales, Universidad Nacional – Bogotá, D.E. Colombia. 253-264.
Gaston K.J. (1996) Species richness: measure and measurement. In: Gaston K.L. (ed.)
Biodiversity: A biology of numbers and difference. Blackwell Science,
Oxford, UK, pp. 77-113.
Gentry A.H. (1976) Additional Panamanian Passifloraceae. Annals of the Missouri
Botanical Garden 63: 341-345.
Gentry A.H. (1982) Neotropical floristic diversity: phytogeographical connections
between Central and South America, Pleistocene climatic fluctuations, or an
accident of the Andean orogeny?. Annals of Missouri Botanical Garden
69:557-593.
Gentry A.H. (1986) Endemism in tropical versus temperate plant communities. In Soulé
M.E. (ed), Conservation biology. The science of scarcity and diversity.
Sinauer Associates, Sunderland, Massachussets, pp. 153-181.
Gentry A.H. (1986) Species richness and floristic composition of Choco´ Region plant
communities. Caldasia 15: 71–75.
Gentry A.H. (1989) Northwest South America (Colombia, Ecuador and Peru). In
Campbell D.G., and Hammond H.D. (eds), Floristic inventory of tropical
countries. New York Botanical Garden, Bronx. Pp. 391-400.
Gentry A.H. (1992) New species of woody plants from Amazonian Peru. Novon 2(4):
333-338.
Gentry A.H. (1993) “Riqueza de especies y composición florística.” In: P. LEYVA (Ed.).
Colombia Pacífico, Vol. 1. Fondo Protección del Medio Ambiente José
Celestino Mutis, Publicaciones Financiera Eléctrica Nacional (FEN), Bogotá.
Gerlach G. and Ulmer T. (2000) Una nueva especie de Passiflora, subgénero Passiflora
(Passifloraceae) del oeste de Colombia. Caldasia 22 (2): 231-235.
Gilbert L.E. (1982) The evolution of a butterfly and a vine, Heliconius butterflies.
Scientific American 247: 110-121.
Green P.S. (1972) Passiflora in Australasia and the Pacific. Kew Bulletin 26(3): 539-558.
213
Bibliography
________________________________________________________________________
Gremillion K.J. (1989) The development of a mutualistic relationship between humans
and maypops (P. incarnata L.) in the Southeastern United States. Journal of
Ethnobiology 9: 135–158.
Haffer J. (1982) General Aspects of the Refuge Theory In: Prance GT (Ed) Biological
diversification in the Tropics. Columbian University Press, New York, USA,
pp 6-24.
Hall R. M., Drew R.A., Higgins C.M., Dietzgen R.G. (2000) Efficient organogenesis of
an Australian passion fruit hybrid (Passiflora edulis x Passiflora edulis var.
flavicarpa) suitable for gene delivery. Austin Journal Botany 48:673–680.
Handler N. (1962) Psychomimetic Medicine, The First Hahnemann Symposium. Lea and
Febiger, Philadelphia.
Hansen A.K., Lawrence G., Simpson B.B., Downie S.R., Stephen R., Cervi A.C., and
Jansen R.K. (2006) Phylogenetic relationships and chromosome number
evolution in Passiflora. Systematic Botany 31(1): 138-150.
Hansen A.K., Escobar L.K., Lawrence E., Gilbert E., and Jansen K. (2007) Paternal,
maternal and biparental inheritance of chloroplast genome in Passiflora
(Passifloraceae): American Journal of Botany 94 (1): 42-46.
Harms H. (1898) Zür morphologie der ranken und blütenstände bei den Passifloraceae.
Botanische Jahrbücher für Systematik 24: 163-178.
Harms H. (1925) Passifloraceae. In Die naturlichen Pflanzenfamilien 21, Germany: W.
Engelmann, 470-507.
Harris S.A. and Ingram R. (1991) Chloroplast DNA and biosystematics: the effects of
intraspecific diversity and plastid transmission. Taxon 40:393–412.
Havey M.J., McCreight J.D., Rhodes B., and Taurick G. (1998) Differential transmission
of the Cucumis organellar genomes. Theoretical and Applied Genetics 97:
122-128.
Hernández J., Hurtado A., Ortiz R., and Walschburger T. (1991) Unidades
Biogeográficas de Colombia En: Hernández J., Ortiz R., Walshburger T., y
Hurtado A. (Eds.) Estado de la Biodiversidad en Colombia Informe Final
Santafé de Bogotá, Instituto Colombiano para el Desarrollo de la Ciencia y la
Tecnología “Francisco José de Caldas” - Colciencias.
Hernández A. (2003) Revisión taxonómica de Passiflora, subgénero Decaloba
(Passifloraceae) en Colombia. Tesis, Facultad de Ciencias, Departamento de
Biología, Universidad Nacional de Colombia sede Bogotá. 138pp.
214
Bibliography
________________________________________________________________________
Hernández A. and Bernal R. (2000) Lista de especies de Passifloraceae de Colombia.
Biota Colombiana 1 (3): 320- 335.
Hernández A.. and García N. (2006) Las pasifloras (familia Passifloraceae). En: Libro
rojo de plantas de Colombia. Las bromelias, las labiadas y las pasifloras.
Garcia N., and Galeano G. (eds.). volumen 3: 553-663.
Hickey M. and King C. (1988) 100 Families of Flowering Plants. Cambridge University
Press, Cambridge, pp. 130–133.
Hijmans R.J., Garrett K.A., Huamán Z., Zhang D.P., Schreuder M., and Bonierbale M.
(2000) Assessing the geographic representativeness of genebank collections:
the case of Bolivian wild potatoes. Conservation Biology 14(6): 1755-1765.
Hijmans R.J., Schreuder M., De la Cruz M., and Rojas E. (2001) Computer tools for
spatial analysis of plant genetic resources data: DIVA-GIS. Plant Genetic
Resources Newsletter 27: 15-19.
Hijmans R.J., Guarino L., Bussink C., Mathur P., Cruz M., Barrentes I., and Rojas E.
(2005) DIVA-GIS. A geographic information system for the analysis of
species distribution data. Manual, version 5.2.
Hijmans R.J. and Graham C.H. (2006) The ability of climate envelope models to predict
the effect of climate change on species distributions. Global Change Biology
12(12): 2272-2281.
Hijmans R.J. and Spooner D.M. (2001) Geography of wild potato species. American
Journal of Botany 88: 2101-2112.
Hill T. and Lewicki P. (2006) STATISTICS Methods and Applications. StatSoft, Tulsa,
UK.
Hno. Daniel (J. González Patiño) (1968) Curiosidades de una flor y de una familia
botánica (La Flor de la Pasión). Boletín del Instituto de la Salle, Bogotá. 208:
261-270.
Ho W.F., and Shii C.T. (1986) Incompatibility system in passion fruit (Passiflora edulis
Sims). Acta Horticulturae 194: 31–38.
Holm-Nielsen L.B. (1974) Notes on central Andean Passifloraceae. Botaniska Notiser
127:338-351.
Holm-Nielsen L.B., Jørgensen P.M., and Lawesson J.E. (1988) Passifloraceae. In:
Harling & L. Andersson (eds.), Flora of Ecuador 31: 124pp.
215
Bibliography
________________________________________________________________________
Howell C.W. (1976) Edible fruited Passiflora adapted to South Florida growing
conditions. Proceeding Florida State Horticultural Society 89: 236-238.
Hutchinson J. (1967) Passifloraceae. The general of florewing Plants. Vol. 2. Oxford:
Clarendon Press. 364-374.
IGAC (Instituto Geográfico Agustín Codazzi) (2006) Mapas departamentales.
http://www.igac.gov.co. Cited 18 November 2005.
IUCN (World Institute for Conservation and Environment) (2003) Guidelines for
application of IUCN red list criteria at regional levels. Version 3.0. IUCN
Species Survival Commission. IUCN, Gland, Switzerland and Cambridge,
UK.
IUCN (World Institute for Conservation and Environment) (2004) Guidelines for using
the IUCN Red List categories and criteria. Standards and petitions
subcommittee of the IUCN SSC Red List programme committee. IUCN,
Gland, Switzerland and Cambridge, U.K.
Jamir T.T., Sharma H.K., and Dolui A.K. (1999) Folklore medicinal plants of Nagaland,
India. Fitoterapia 70, 395–401.
Jarvis A., Guarino L., Williams D., Williams K., & Hyman G. (2002) Spatial analysis of
wild peanut distributions and the implications for plant genetic resource
conservation. Plant Genetic Resources Newsletter 131: 29-35.
Jarvis A., Ferguson M.E., Williams D.E., Guarino L., Jones P.G., Stalker H.T., Valls
J.F.M., Pittman R.N., Simpson C.E., and Bramel P. (2003) Biogeography of
wild Arachis: assessing conservation status and setting future priorities. Crop
Science. 43(3): 1100-1108.
Jarvis A., Williams D., Guarino L., Caballero P.J. and Mottram G. (2005) Use of GIS for
optimizing a collecting mission for a rare wild pepper (Capsicum flexuosum
Sendtn.) in Paraguay. Genetic Resources and Crop Evolution 52 (6): 671-682.
Jones P.G., Beebe S.E., Tohme J., and Galwey N.W. (1997) The use of geographical
information systems in biodiversity exploration and conservation. Biodiversity
and Conservation 6: 947-958.
Jones, P., Guarino L., and Jarvis A. (2002) Computer tools for spatial analysis of plant
genetic resources data: 2. FloraMap. Plant Genetic Resources Newsletter
130:1–6.
Jørgensen P.M., Lawesson J.E.., and Holm-Nielsen L.B. (1984) A guide to collecting
passion flowers. Annals of the Missouri Botanical Garden 71(4): 1172-1174.
216
Bibliography
________________________________________________________________________
Jørgensen P.M., Ulloa C., Madsen J.E., and Valencia R. (1995) A floristic analysis of the
high Andes of Ecuador. Churchill S.P., Balslev H., Forero E., and Luteyn
J.L.(eds.). Biodiversity and Conservation of Neotropical Montane Forests,
Proceedings of the Neotropical Montane Forest Biodiversity and Conservation
Symposium, The New York Botanical Garden, Bronx, NY: 221-237.
Jørgensen P.M., and León-Yánez S. (1999) Results. In: P.M. Jørgensen and S. LeónYánez (eds.): Catalogue of the vascular plants of Ecuador. Monograph
Systematic Botonay, Missouri Botanical Garden 75: 42-108.
Judd W.S., Campbell C.S., Kellogg, E.A., Stevens P.F., and Donoghue M.J. (2002) Plant
systematics: a phylogenetic approach. Sinauer Associates, Inc., Sunderland,
Massachusetts, USA.
Junqueira N.T.V., Bragal M.F., Faleiro F.G., Peixoto J.R., e Bernacci L.C. (2005)
Potencial de espécies silvestres de maracujazeiro como fonte de resistência a
doenças. 81-108p. En: Maracujá: germoplasma melhoramento genético. (Eds.)
Fábio Gelape Faleiro, Nilton Tadeu Vilela Junqueira, Marcelo Fideles Braga.
Planaltina, DF : Embrapa Cerrados. 670 p.l.
Jussieu A.L. de. (1805) Second Mémoire sur la famille des Passiflorées, et
particulièrement sur quelques espèces nouvelles du genre Tacsonia. Annales
du Muséum d’Histoire Naturelle 6: 388- 396.
Kay E. (2001) Observations on the pollination of Passiflora penduliflora. Biotropica
33:709–713.
Kessler M. (2002) The elevational gradient of Andean plant endemism: varying
influences of taxon-specific traits and topography at different taxonomic
levels. Journal of Biogeography 29: 1159-1165.
Killip E.P., Meggers W.F., and Hewett D.F. (1924) New species of Passiflora from
Tropical America. Journal of the Washington Academy of Sciences 14 (5):
108-112.
Killip E.P. (1930) Ten new species of Passiflora, mainly from Colombia and Peru.
Journal of the Washington Academy of Sciences 20 (15): 374-381.
Killip E.P. (1938) The American Species of Passifloraceae. Field Museum of Natural
History Publication. Botanical Series 19: 613pp.
Killip E.P. (1960) Supplemental notes to the American species of Passiflora with
descriptions of new species. Contributions from the U.S. National Herbarium
35: 2 (Tomo I).
217
Bibliography
________________________________________________________________________
Kingston N. and Waldren S. (2005) A conservation appraisal of the rare and endemic
vascular plants of Pitcairn Island. Biodiversity and Conservation 14: 781-800.
Kirtikar K.R. and Basu, B.D. (1975) Indian Medicinal Plants. Periodical Experts,
Dehradun, India, p1103.
Knight R.J. and Winters H.F. (1962) Pollination and fruit set of yellow passion fruit in
Southern Florida. Proceeding of the Florida State Horticultural Society 75:
412-418.
Knight R.J. (1991) Development of tetraploid hybrid passion fruit clones with potential
for the north temperate zone. HortScience 26: 1541-1543.
Knight R.J. (1992) Characters needed for commercially successful passion fruit.
Proceeding Florida State Horticultural Society 105: 280-282.
Krosnick S.E. and Freudestein J.V. (2005) Monophyly and floral character homology of
Old Word Passiflora (Subgeneus Decaloba: supersection Disemma).
Systematic Botany 20(1): 139-152.
Kugler E.E. and King L.A. (2004) A brief history of the Passionflower. 15-26 pp. In
Ulmer T., and MacDougal J.M. (eds). Passiflora: passionflowers of the word.
Timber Press Portland, Oregon. 430 pp.
Kyndt T., Romeijn-Peeters E., Van Droogenbroeck B., Romero-Motochi J.P., Gheysen
G., and Goetghebeur P. (2005) Species relationships in the genus
Vasconcellea (Caricaceae) based on molecular and morphological evidence.
American Journal of Botany 92: 1033–1044.
Lakshmi M., Senthilkumar P., Parani M., Jithesh M.N., and Parida A. (2000) PCR-RFLP
analysis of chloroplast gene regions in Cajanus (Leguminosae) and allied
genera. Euphytica 116: 243–250.
Lee D.J., Blake T.K., and Smith S.E. (1988) Biparental inheritance of chloroplast DNA
and the existence of heteroplasmic cells in Alfalfa. Theoretical and Applied
Genetics 76:545–549.
Lehmann A., Leathwick J.R., and Overton J.M. (2002) Assessing New Zealand fern
diversity from spatial predictions of species assemblages. Biodiversity and
Conservation 11: 2217-2238.
Leimberck R., Valencia R., and Balslev H. (2004) Landscape diversity patterns and
endemism of Araceae in Ecuador. Biodiversity and Conservation 13 (9):
1755-1779.
218
Bibliography
________________________________________________________________________
Lemarck J.B. Monenet de. (1789) Encyclopédie Méthodique, Botanique 3(1). Paris:
Panckoucke.
Linneaus C. (1753) Species Plantarum. Vol. 2. Stockholm: L. Salvius.
MacDougal J.M. (1992) New species of Passiflora subg. Plectostemma (Passifloaceae).
Novon 2(4): 358-367.
MacDougal J.M. (1994) Revision of Passiflora section Decaloba, Pseudodysosmia
(Passifloraceae). Systematic Botany Monographs 14: 146pp.
MacDougal J.M. (2006) Passiflora sandrae (Passifloraceae), a new species from Panama.
Novon 16: 85-88.
MacMillan R.T. and Graves W.R. (1992) Susceptibility of Passiflora spp. to Alternaria
passiflorae. In: Memorias Primer Simposio Internacional de Passifloras,
Universidad Nacional de Colombia, Palmira, 123-124.
MacNeely J.A., Miller K.R.., Reid N.A., Mittemer R.A., and Wainer T.B.. (1990)
Conserving the world’s biological diversity World Conservation Union,
World Resources Institute, World Wildlife Fund - U.S. World Bank,
Washigton, D.C.
Manders G., Otoni W.C., d'Utra-Vaz F.B., Blackhall N.W., Power J.B., and Davey M.R.
(1994) Transformation of passionfruit (Passiflora edulis f. flavicarpa
Degener) using Agrobacterium tumefaciens. Plant Cell Reports 13: 697-702.
Martin F.W. and Nakasone H.Y. (1970) The edible species of Passiflora. Economic
Botany 24 (3): 333-343.
Martin P.S. and Szuter C.R. (1999) War zones and game sinks in Lewis and Clark’s
West. Conservation Biology 13: 36-45.
Master M.T. (1872) Passifloraceae. In Flora Brasiliensis 13 (1). Ed. K. Mrtius and A.
Eichler. 529-628.
Master M.T. (1877) A classified synonymic list of all the species of Passiflorae cultivated
in Europe gardens, with references to the works in which they are figured.
Journal of the Royal Horticultural Society 4: 125-149.
Maxted N., van Slageren M.W., and Rihan, J.R. (1995) Ecogeographic surveys. In
‘Collecting plant genetic diversity. Technical guidelines’. (eds L. Guarino,
R.V. Ramantha & R.Reid). Chapter 14. CAB International.
Maxted N., Mabuza-Dlamini P., Moss H., Padulosi S., Jarvis A., and Guarino L. (2004)
African Vigna : An Ecogeographic Study, IPGRI, Rome, Italy.
219
Bibliography
________________________________________________________________________
May P.G. and Spears E.E. (1988) Andromonoecy and variation I in phenotypic gender of
Passiflora incarnata (Passifloraceae). American Journal of Botany 75: 18301841.
May R.M. (1992) "How many species inhabit the Earth?". Scientific American 267(4):
18-24.
McNeely J.A., Miller K.R., Reid N.A., Mittemer R.A., and Wainer T.B. (1990)
Conserving the world’s biological diversity World Conservation Union,
World Resources Institute, World Wildlife Fund – U.S. World Bank,
Washigton, D.C.
Medina, C. I., Lobo, M., and Correa, R. 2000. Caracterización morfológica y química de
Pasifloras Andinas como apoyo al desarrollo de estas especies. III Seminario
de Frutas de Clima Frío Moderado. Manizales, Colombia. 13-18.
Meletti L.M.M., Santos R.R., and Minami K. (2000) Melhoramento do maracujazeiroamarelo: obtenção do Composto IAC-27. Scientia Agricola, Piracicaba, v. 56,
n. 3, p. 17-24.
Meletti L.M.M., Soares-Scott D., Bernacci L.C., e Passos I.R.da S. (2005) Melhoramento
genético do maracujá: passado e futuro. 55-78pp. En: Maracujá: germoplasma
melhoramento genético. (Eds.) Fábio Gelape Faleiro, Nilton Tadeu Vilela
Junqueira, Marcelo Fideles Braga. Planaltina, DF : Embrapa Cerrados. 670
p.l.
Mes T. H. M., Friesen N., Fritsch R. M., Klass M., and Konrad B. (1997) Criteria for
sampling in Allium based on chloroplast DNA PCR-RFLP''s. Systematic
Botany 22: 701–712.
Metzlaff M., Borner T., and Hagemann R. (1981) Variations of chloroplast DNAs in the
genus Pelargonium and their biparental inheritance. Theoretical Applied
Genetics 60: 37-41.
Midgley G.F., Hannah L., Millar D., Rutherford M.C., and Powrie L.W. (2002)
Assessing the vulnerability of species richness to anthropogenic climate
change in a biodiversity hotspot. Global Ecology Biogeography 11:6 445.
Moon E., Kao T-H., and Wu R. (1987) Rice chloroplast DNA molecules are
heterogeneous as revealed by DNA sequences of a cluster of genes. Nucleic
Acids Research 15:611–630.
Morton J.F. (1981) Atlas of Medicinal Plants of Middle America. Spring- field, IL, p.
1281.
220
Bibliography
________________________________________________________________________
Mowrey D. (1993) Herbal Tonic Therapies. Keats Publishing Incorporation, New
Canaan, CT.
Mráček J. (2005) Investigation of interspecific genome-plastome incompatibility in
Oenothera and Passiflora. Ph.D Dissertation. Munich University, Germany.
110pp.
Muschner V., Lorenz-Lemke A., Cervi A.C., Bonatto S., Souza-Chies T., Salzano F., and
Freitas L. (2003) A first molecular phylogenetic analysis of Passiflora
(Passifloraceae). American Journal of Botany 90: 1229-1238.
Muschner V., Lorenz-Lemke A., Vecchia M., Bonatto S., Salzano F., and Freitas L.
(2006) Differential organellar inheritance in Passiflora’s (Passifloraceae)
subgenera. Genetica 128 (1-3): 449-453.
Myers J.E. (2000) The biodiversity challenge: Expanded hot spots analysis. The
Environmentalist 10: 243–256.
Myers N., Mittermeier R.A., Mittermeier C.G., da Fonseca G.A.B., and Kent J. (2000)
Biodiversity hotspots for conservation priorities. Nature 403: 853-858.
Nakasone H.Y., Hirono R., and Ito P. (1967) Preliminary observations on the inheritance
of several factors in the passionfruit (Passiflora edulis and forma flavicarpa).
Tech. Prog. Rpt. 161. U. Hawaii, Agr. Exper. Sta., Honolulu.
Nascimento W.M.O., Tomé A.T., Oliviera M.do S.P.de., Muller C H., e Caravalho J.E.U.
(2003) Seleção de progênies de maracujazeiro-amarelo (Passiflora edulis f.
flavicarpa) quanto à qualidade de frutos. Revista Brasileira Fruticultura
25(1): 186-188.
Negreiros J.R. da Silva., Bruckner C.H., Cruz C.D. De Siqueira D.L., and Pimentel L.D.
(2004) Selection of yellow passion fruit vines for vigor and resistance to scab
(Cladosporium cladosporioides). Revista Brasilera de Fruticultura 26(2): 272275.
Nishida T. (1958) Pollination of the passion fruit in Hawaii. Journal of Economic
Entobotanic 51: 146–149.
Nwosu M.O. (1999) Herbs for mental disorders. Fitoterapia 70: 58–63.
Ocampo J., Coppens d’Eeckenbrugge G., Olano C., and Schnell R. (2004) AFLP analysis
for the study of genetic relationships among cultivated Passiflora species of
the subgenera Passiflora and Tacsonia. Proceedings of the Interamerican
Society for Tropical Horticulture 47: 72-76.
221
Bibliography
________________________________________________________________________
Ocampo J., Coppens d’Eeckenbrugge G., Restrepo M., Salazar M., Jarvis A., and
Caetano C. (2007) Diversity of Colombian Passifloraceae: biogeography an
updated list for conservation. Biota Colombiana (in press).
Oga S., de Freitas P.C.D., Gomes da Silva A.C., and Hanada S. (1984) Pharmacological
trials of crude extracts of Passiflora alata. Planta Medica 51: 303–306.
Oliveira J.C. de., Ferreira F.R.., Ruggiero C., Nakamura K. (1987) Caracterização e
avaliação de germoplasma de Passiflora edulis. In: Congresso Brasileiro
Fruticultura, 9. Campinas. Anais. Campinas: SBF, v.2, p. 591-596.
Oliveira J.C. and Ferreira F.R. (1991) Melhoramento genético do maracujazeiro. In: "A
cultura do maracuja no Brasil", São José, A.R., Ferreira, F.R. and Vaz, R.L.
(eds.), Jaboticabal, FUNEP, pp. 187-200.
Olson D.M. and Dinerstein E. (1998) “The Global 200: A representation approach to
conserving the earth’s most biologically valuable ecoregions.” Conservation
Biology 12: 502–515.
Otoni V.C., Blackhall N.W., D'Utra Vaz F.B., Casali V.W., Power J.B., and Davey M.R.
(1995) Somatic hybridization of the Passiflora species, P. edulis var.
flavicarpa Degener and P. incarnata L. Journal of Experimental Botany 46:
777-785.
Parani M., Lakshmi M., Ziegenhagen B., Fladung M., Senthilkumar P., and Parida A.
(2000) Molecular Phylogeny of Mangroves VII. PCR-RFLP of trnS-psbC and
rbcL gene regions in 24 mangrove and mangrove associate species.
Theoretical Applied and Genetics 100: 454-460.
Parques Naturales de Colombia (2006) http://www.parquesnacionalesdecolombia.gov.co.
Cited 10 Sept 2006.
Passionfruit (2006) Supply and demand. http://www.passionfruitjuice.com. Cited 3
October 2006.
Payán F.R. and Martín F.W. (1975) Barriers to the hybridization of Passiflora species.
Euphytica 24: 709-716.
Pearson D.L. (1994) Selecting indicator taxa for the quantitative assessment of
biodiversity. Philosophical Transactions of the Royal Society of London 345:
75–79.
Pérez E. (1956) Plantas útiles de Colombia. Tercera redacción muy corregida y
aumentada, con XLV laminas en negro, otras en color y 752 figuras en el
texto. Librería Colombiana - Camacho Roldán (Cia, Ltda.) - Bogotá. 611614.
222
Bibliography
________________________________________________________________________
Perrier X., Flori A., and Bonnot, F. (2003) Data analysis methods. In: Hamon, P., Seguin,
M., Perrier, X., Glaszmann, J. C. Ed., Genetic diversity of cultivated tropical
plants. Enfield, Science Publishers. Montpellier. pp 43 - 76.
Peterson A.T. (2001) Predicting species geographic distribution based on ecological
niche modeling. Condor 10: 599-605.
Peterson A.T. (2004) Predictability of the geography of species invasions via ecological
niche modeling. Quarterly Review of Biology 78: 419-433.
Peterson A.T. and Robins C.R. (2003) When endangered meets invasive: ecological
niche modeling predicts double trouble for spotted owls, Strix occidentalis.
Conservation Biology 17: 1161-1165.
Porter-Utley K. (2003) Revision of Passiflora subgenus Decaloba supersection Cieca
(Passifloraceae). Ph.D dissertation, University of Florida. 444p.
Presting D. (1969) Zür morphology der pollenkorner der passifloraceen. Pollen et Spores
(2): 193-247.
Primot S., Coppens d’Eeckenbrugge G., Rioux V., Ocampo J., and Garcin F. (2005)
Variación morfológica de tres especies de curubas (Passiflora tripartita var.
mollissima, P. tarminiana y P. mixta) y sus híbridos en el Valle del Cauca
(Colombia). Revista Brasilera de Fruticultura 27(3): 467-471.
Radhamani T.R., Sudarshana L., and Krishnan R. (1995) Defence and carnivory: dual
roles of bracts in Passiflora foetida L. Journal of Biosciences 20: 657-664.
Rangel J.O. (1995) La diversidad florística en el espacio andino de Colombia. En S.P.
Churchill et al., eds. Churchill, S. P., H. Balslev, E. Forero & J. L. Luteyn
(eds.). 1995. Biodiversity and Conservation of Neotropical Montane Forests,
Proceedings of the Neotropical Montane Forest Biodiversity and Conservation
Symposium, The New York Botanical Garden, Bronx, NY: 187-205.
Rangel J.O. (2002) El estado actual del conocimiento de la flora de Colombia. Pág. 570
en: Rangel J.O., J. Aguirre-C & M.G. Andrade-C. (eds), Libro de resúmenes
octavo congreso latinoamericano y segundo Colombiano de botánica Instituto
de Ciencias Naturales, Universidad Nacional de Colombia, Bogotá.
Raven P.H. (1975) The bases of angiosperm phylogeny: cytology. Annals of Missouri
Botanical Garden 62:734-764.
Rawat P.S. (1987) Select Your Dose and Potency. B. Jain Publishers (P) Ltd., New Delhi,
pp. 481–482.
223
Bibliography
________________________________________________________________________
Rebelo A.G. (1994) Iterative selection procedures: Centers of endemism and optimal
placement of reserves. Strelitzia 1:231–257.
Rêgo M.M, Bruckner C.H., Finger F.L., Siqueira D.L., and Fernandes A.A. (1998) Selfincompatibility: evidence for two-locus genetic control in yellow passion fruit.
Theoretical and Applied Genetics 98:564–568.
Rêgo M.M., Rêgo E.R., Bruckner C.H., Da Silva E.A.M., Finger F.L., and Pereira K.J.C.
(2000) Pollen tube behavior in yellow passion fruit following compatible and
incompatible crosses. Theoretical and Applied Genetics 101:685–689.
Reichenbach H.G.I. (1828) Passifloraceae. Conspectus regni vegetabilis. Leipzig,
Germany: Carl Cnobloch. 132.
Richardson J.E., Pennington R.T., Pennington T.D., and Hollingsworth P.M. (2001)
Rapid diversification of a species-rich genus of neotropical rain forest trees.
Science 293: 2242-2245.
Robbirt K.M., Roberts D.L., and Hawkins J.A. (2006) Comparing IUCN and
probabilistic assessments of threat: do IUCN red list criteria conflate rarity
and threat?. Biodiversity and Conservation 15: 1903-1912.
Rodríguez D., Marín C., Quecán H., and Ortiz R. (2005) Áreas potenciales para colectas
del género Vasconcellea Badillo en Venezuela. Bioagro 17 (1): 3-10.
Roemer M.J. (1846) Passifloracea. Familiarum naturalium regni vegetabilis synopses
monographicae 2. Weimar, Germany. 15-207.
Romero-C.R. (1956) Plantas de valor comercial del género Passiflora: Granadilla,
curuba, badea, y otras. Agricultura Tropical 12 (6): 403-407.
Romero-C.R. (1991) Frutas silvestres de Colombia. Segunda edición actualizada,
Volumen con 207 ilustraciones. Instituto Colombiano de Cultura Hispánica.
423-429 p.
Royal Horticultural Society. (2002) Royal Horticultural Society Colour Chart. London.
Ruberté-Torres R. and Martín F.W. (1974) First generation hybrids of edible passion fruit
species. Euphytica 23: 61-70.
Ruggiero C., Banzatto DA., and Lam-Sanchez A. (1976) Studies on natural and
controlled pollination in yellow passion fruit (Passiflora edulis f. flavicarpa
Deg.). Acta Horticulturae 57:121–124.
224
Bibliography
________________________________________________________________________
Saitou N. and Nei M. (1987) The neighbor-joining method: a new method for
reconstructing phylogenetic trees. Molecular Biology and Evolution 4(4):406425.
Sánchez I., Fajardo A., Grum M., Duque M., Lobo M., Tohme J., and Roca W. (1999)
Variability of chloroplast DNA in the genus Passiflora L. Euphytica 106 : 1526.
Sazima M. and Sazima I. (1978) Bat pollination of the passion flower, Passiflora
mucronata, in southeastern Brazil. Biotropica 10: 100-109.
Scheldeman X., Willemen L., Coppens d'Eeckenbrugge G., Romeijn-Peeters E., Restrepo
M.T., Romero Motoche J., Jimenez D., Lobo M., Medina C.I., Reyes C.,
Rodriguez D., Ocampo J.A., Van Damme P., and Goetgebeur P. (2006)
Distribution, diversity and crop ecology of highland papayas (Vasconcellea
spp.) and papaya (Carica papaya L.) in Latin America. Biodiversity and
Conservation. DOI 10.1007/s10531-006-9086-x.
Schöniger G. (1986) La curuba: técnicas para el mejoramiento de su cultivo. Bogotá: Ed.
Guadalupe. 256p.
Seaforth C.E., Adams C.D., and Sylvester Y. (1983) A Guide for the Medicinal Plants of
Trinidad & Tobago. Commonwealth Secretariat, Marlborough House, Pall
Mall, London.
Sechrest W., Brooks T.M., da Fonseca G.A.B., Konstant W.R., Mittermeier R.A., Purvis
A., Rylands A.B., and Gittleman J.L. (2002) Hotspots and the conservation of
evolutionary history. Proceeding National Academy Science 99(4): 20672071.
Segura S., Coppens d'Eeckenbrugge G., Ocampo C., and Ollitrault P. (2002) Isozyme
variation in Passiflora subgenera Tacsonia and Manicata. Relationships
between cultivated and wild species. Genetic Resources and Crop Evolution
50: 417-427.
Segura S., Coppens d’Eeckenbrugge G., Bohórquez A., Ollitrault P., and Tohmé J.
(2002) An AFLP study of the genus Passiflora focusing on subgenus
Tacsonia. Genetic Resources and Crop Evolution 49: 111-123.
Segura S., Coppens d’Eeckenbrugge G., López L., Grum M., and Guarino L. (2003)
Mapping the potential distribution of five species of Passiflora in Andean
countries. Genetic Resources and Crop Evolution 50: 555-566.
Segura S., Coppens d'Eeckenbrugge G., Ocampo C-H., and Ollitrault P. (2005) Isozyme
variation in Passiflora subgenus Tacsonia: geographic and interspecific
differentiation among the three most common species. Genetic Resources and
Crop Evolution 52: 455-463.
225
Bibliography
________________________________________________________________________
Silva M.B. (1998) Transformação genética de maracujá-amarelo (Passiflora edulis f.
flavicarpa Degener) mediada por Agrobacterium tumefaciens. M.Sc. thesis,
Universidade Federal de Vicosa, Vicosa, Brazil.
Silverstone-Sopkin P.A. and Ramos-Pérez J.E. (1995) Floristic exploration and
phytogeography of the cerro del Torrá, Chocó, Colombia. Biodiversity of
Conservation of Neotropical Montane Forests: 169-86. In Biodiversity and
Conservation of Neotropical Montane Forests, Proceedings of a Symposium,
New York Botanical Garden, 21-26 June 1993 (S.P. Churchill, H. Balslev, E.
Forero and J.L. Luteyn, eds.). New York Botanical Garden, New York.
Simpson B.B. (1975) Pleistocene changes in the flora of high tropical Andes.
Paleobiology 1:273-294.
Simpson B.B. and Todzia C.A. (1990) Patterns and processes in the development of the
high Andean flora. American Journal of Botany 77:1419-1432.
Skov F. (2000) Potential plant distribution mapping based on climatic similarity. Taxon
49: 503-515.
Snow A. (1982) Pollination intensity and potential seed set in Passiflora vitifolia.
Oecologia 55: 231-237.
Snow N., and MacDougal J.M. (1993) New chromosome reports in Passiflora
(Passifloraceae). Systematic Botany 18 (2): 261-273.
Sokal R.R. and Michener C.D. (1958) A statistical method for evaluating systematic
relationships. U. Kansas Sci. Bull .38:1409-1438.
Stebbins G.L. (1974) Evolution above the species level. Harvard University Press,
Cambridge, USA, 399 p.
Storey W.B. (1950) Chromosome numbers of some species of Passiflora occurring in
Hawaii. Pacific Science 4: 37-42.
Suassuna T.M.F., Bruckner C.H., Carvalho C., and Borém A. (2003) Selfincompatibility
in passion fruit: evidence of gametophytic– sporophytic control. Theoretical
and Applied Genetics, 106: 298–302.
Svenning J.C. (1998) The effect of land-use on the local distribution of palm species in
an Andean rain forest fragment in northwestern Ecuador. Biodiversity and
Conservation 7: 1529–1537.
226
Bibliography
________________________________________________________________________
Taberlet, P., Gielly L., Patou G., and Bouvet J. (1991) Universal primers for
amplification of three noncoding regions of chloroplast DNA. Plant
Molecular Biology 17: 1105.1109.
Taylor T. N., & Taylor E.L. (1993) The Biology and Evolution of Fossil Plants. Prentice
Hall, NJ, USA. 982pp.
Taylor D.W. (1995) Cretaceous to tertiary geologic and angiosperm paleobiogeographic
history of the Andes. 3-22pp. In Biodiversity and Conservation of Neotropical
Montane Forests, Proceedings of a Symposium, New York Botanical Garden,
21-26 June 1993 (S.P. Churchill, H. Balslev, E. Forero and J.L. Luteyn, eds.).
New York Botanical Garden, New York, pp. 561-577.
Tillet S. (2003) Passifloraceae. In: Flora of the Venezuelan Guayana Volume 7:
Myrtaceae-Plumbaginaceae by Julian A. Steyermark, Paul E. Berry, Kay
Yatskievych, and Bruce K. Holtz (eds.): 625-667.
Ulmer T. (1999) Passiflora formosa sp. nov., a hitherto misunderstood taxon in
Passiflora subgenus Tacsonia (Passifloraceae) from Colombia. Edinburgh
Journal of Botany 56 (2): 195-198.
Ulmer T. and MacDougal J.M. (2004) Passiflora: passionflowers of the word. Timber
Press Portland, Oregon. 430 pp.
UNEP-WCMC. (2004) World Conservation Monitoring Centre of the United Nations
Environment Programme. Species Data (unpublished, September 2004). Web
site at: http://www.unep-wcmc.org. Cambridge, England: UNEP-WCMC.
Uribe L. (1954) Dos nuevas Passifloraceae colombianas. Mutisia 21: 1-5.
Uribe L. (1955a) Pasifloráceas y Begoniáceas de la Real Expedición Botánica del Nuevo
Reino de Granada. Ediciones Cultura Hispánica. Madrid 26:1-98.
Uribe L. (1955b) Sertula Florae Colombiana 2. Caldasia 7(32):159-165.
Uribe L. (1957) Una nueva e interesante Passiflora de Colombia. Caldasia 7(35):335338.
Uribe L. (1958) Sertula Florae Colombiae 4. Caldasia 8 (37):127-130.
Uribe L. (1972) Catalogo ilustrado de las plantas de Cundinamarca: Passifloraceae,
Begoniaceae, Melastomataceae. Instituto de Ciencias Naturales, Facultad de
Ciencias, Universidad Nacional. (5): 11-41.
Uribe L. (1977) Sertula Florae Colombiae 14. Caldasia 12 (56): 13-18.
227
Bibliography
________________________________________________________________________
Van Droogenbroeck B., Kyndt T., Maertens I., Romeijn-Peeters E., Scheldeman X.,
Romero-Motochi J., Van Damme P., Goetghebeur P., and Gheysen G. (2004)
Phylogenetic analysis of the highland papayas (Vasconcellea) and allied
genera (Caricaceae) using PCR-RFLP. Theoretical Applied and Genetics 108:
1473-1486.
Vanderplank J. (2000) Passion Flowers. 3nd ed. The MIT Press. Cambridge,
Massachusetts. 224 pp.
Vanderplank J. (2006) Passiflora miniata: Passifloraceae. Curtis’s Botanical Magazine
23 (3): 223-230.
Varassin I.G., Trigo J.R., and Sazima M. (2001) The role of nectar production, flower
pigments and odour in the pollination of four species of Passiflora
(Passifloraceae) in south-eastern Brazil. Botanical Journal of the Linnean
Society 136: 139–152.
Vargas J.D. (2000) Relaciones filogenéticas de especies del genero Passiflora
(Passifloraceae) con énfasis en el subgénero Tacsonia a partir de RFLP y
PCR-RFLP. Tesis (Biólogo), Facultad de Ciencias, Pontificia Universidad
Javeriana, Santa fe de Bogota - Colombia. 168pp.
Vargas J.H., Consiglio T., Jørgensen P.M., and Croat T.B. (2004) Modeling distribution
patterns in a species-rich plant genus, Anthurium (Araceae), in Ecuador.
Diversity and Distribution 10(3): 211-216.
Varón M.H. (2000) Determinación de polimorfismo en regiones amplificadas de cpDNA,
mtDNA y rDNA de Passiflora utilizando enzimas de restricción
Tesis (Bióloga), Facultad de Ciencias, Pontificia Universidad Javeriana, Santa
fe de Bogota - Colombia. 150pp.
Vasconcellos M. (1991) Biología floral do maracujá doce Passiflora alata Dryan nas
condiçöes de Botucatu-SP. Tesis Faculdade de Ciencias Agrarias.
Universidade Estatudual Paulista. Botucatu. 98p.
Vekemans X., Hardy O., Berken B., Fofana B., and Baudoin J-P. (1998) Use of PCRRFLP on chloroplast DNA to investigate phylogenetic relationships in the
genus Phaseolus. Biotechnology, Agronomy, Society and Environment 2 (2):
128–134.
Whitmore T.C., and Prance G.T. (1987) Biogeography and Quaternary history in
Tropical America. Oxford Science Publications, Oxford, UK, 214 p.
Villacís L., Grum. M., and Coppens d'Eeckenbrugge G. (1998) Morphological
characterization of Andean Passifloras (Passiflora spp.) from Ecuador. Plant
Genetic Resources Newsletter 115: 51-55.
228
Bibliography
________________________________________________________________________
Villareal H., Álvarez M., Córdoba S., Escobar F., Fagua G., Gast F., Mendoza H.,
Ospina M., y Umaña A.M. (2006) Manual de métodos para el desarrollo de
inventarios de biodiversidad. Programa de Inventarios de Biodiversidad.
Instituto de Investigación de Recursos Biológicos Alexander von Humboldt.
Bogotá, Colombia. 2d. edition. 236 pp.
Vina A.P. (2001) Correlações e parâmetros genéticos e populações de maracujazeiroamarelo (Passiflora edulis f. flavicarpa) e diversidade molecular no gênero
Passiflora. 98p. Tese (Doutorado em Produção Vegetal) - Universidade
Estadual do Norte Fluminense, Campos dos Goytacazes.
Winks C.W., Menzel C.M., and Simpson D.R. (1988) Passionfruit in Queensland. 2.
Botany and cultivars. Queensland Agriculture Journal 114: 217-224.
World Press Review. (1993) Colombia's vanishing forests, World Press Review, June
1993, Vol. 40 (6): 43pp.
Vriesendorp B., & Bakker F.T. (2005). Reconstructing patterns of reticulate evolution in
Angiosperms: what can we do? Taxon 54(3): 593-604.
Yockteng R. (2003) Phylogénies: outils pour l’étude de l’histoire évolutive des
organismes. L’exemple des pensées (genre Viola L.) et des passiflores (genre
Passiflora L.). Université Paris XI – Orsay. Docteur en Sciences. 142pp.
Yockteng R. and Nadot S. (2004) Phylogenetic relationships among Passiflora species
based on the glutamine synthetase nuclear gene expressed in chloroplast
(ncpGS). Molecular Phylogenetics and Evolution 31: 379-396.
Zaharieva M., Prosperi J. M., and Monneveux P. (2004) Ecological distribution and
species diversity of Aegilops L. genus in Bulgaria. Biodiversity and
Conservation 13: 2319-2337.
229
Congress communications
Congress communications
________________________________________________________________________
Ocampo Pérez John, Coppens d’Eeckenbrugge Geo, Salazar Mike, Jarvis Andy. (2007).
La familia Passifloraceae en Colombia: diversidad y conservación. IV Congreso
Colombiano de Botánica. Medellín, abril 2007. Resumen aprobado.
Salazar Mike, Ocampo Pérez John, Restrepo María, Coppens d’Eeckenbrugge Geo,
Caetano Creuci, Jarvis Andrew, Villegas Mauricio. (2006). First results from
diversity mapping of Passiflora (Passifloraceae) and Vasconcellea (Caricaceae)
in the Colombian coffee growing zone. First DIVERSITAS Open Science
Conference: Integrating biodiversity science for human well-being. Oaxaca,
Mexico. November 9-12, 2006. Abstracts (CD).
Ocampo Pérez John, Coppens d’Eeckenbrugge Geo, Risterucci Ange-Marie. (2006).
Relaciones genéticas en el género Passiflora (Passifloraceae) usando
marcadores CAPS sobre el ADN citoplasmático. IX Congreso Latinoamericano
de Botánica. Republica Dominicana, Santo Domingo, junio de 2006, p 325.
Ocampo Pérez John, Salazar Mike, Coppens d’Eeckenbrugge Geo, Restrepo María,
Jarvis Andrew. (2006). Distribución y riqueza de especies de la familia
Passifloraceae en Colombia.
IX Congreso Latinoamericano de Botánica.
Republica Dominicana, Santo Domingo, junio de 2006. p280.
Ocampo Pérez John. (2005). Estudios biogeográficos, morfológicos y moleculares de la
Passifloraceae en Colombia. Seminario Internacional sobre agrobiodiversidad y
producciún de frutales andinos de clima frío y frío moderado. Pasto, Colombia Universidad de Nariño, noviembre 2005. Resúmenes (CD).
Ocampo Pérez John, Olano Cecile, Coppens d’Eeckenbrugge Geo, Schnell Raymond.
(2004). AFLP analysis for the study genetic relationships among cultivated
Passiflora species of the subgenera Passiflora and Tacsonia. 50th Meeting of
the Interamerican Society for Tropical Horticulture, Limón, Costa Rica, October
2004. Program and abstract, 44.
231
Congress communications
________________________________________________________________________
Primot Sophie, Coppens d’Eeckenbrugge Geo, Rioux Vincent, Ocampo Pérez John,
Garcin Francois. (2003). Variación morfológica de tres especies de curubas
(Passiflora tripartita var. mollissima, P. tarminiana y P. mixta) y sus híbridos
en el Valle del Cauca (Colombia). Sexto Simpósio Brasileiro sobre a Cultura do
Maracujazeiro. November 24 27, Campos de Goytacazes (Brazil). Resumo 96
(Compact Disc).
Ocampo Pérez John, Restrepo María, Coppens d’Eeckenbrugge Geo, Caetano Creuci,
Villegas Andrés, Salazar Mike. (2004). Diversidad de Passifloraceae en la zona
cafetera de Colombia. III Congreso Colombiano de Botánica. Popayán,
noviembre 2004. p251.
232
ANNEXES
Annexes
______________________________________________________________
Annex 1. Infrageneric classification according to Killip (1938) with emends of Escobar
(1988, 1989) and MacDougal (1994).
Genus Passiflora L.
507 species
1234-
Subgenus Adenosepala Killip
Subgenus Apodogyne Killip
Subgenus Astephia Killip
Subgenus Astrophea (DC.) Masters
Section Botryastrophea (Harms) Killip
Section Cirrhipes Killip
Section Dolichostemma Killip
Section Euastrophea (Harms) Killip
Section Leptopoda Killip
Section Pseudoastrophea (Harms) Killip
5 - Subgenus Calopathanthus (Harms) Killip
6 - Subgenus Chloropathanthus (Harms) Killip
7 - Subgenus Deidamioides (Harms) Killip
8 - Subgenus Distephana (Juss.) Killip
9 - Subgenus Dysosmia (DC.) Killip
10 - Subgenus Dysosmioides Killip
11 - Subgenus Granadilla (Medic.) Masters
Series Quadrangulares Killip
Series Digitatae Killip
Series Tiliafoliae Killip
Series Marginatae Killip
Series Laurifoliae Killip
Series Serratifoliae Killip
Series Setaceae Killip
Series Pedatae Killip
Series Incarnatae Killip
Series Palmatisectae Killip
Series Kermesinae Killip
Series Imbricatae Killip
Series Simplicifoliae Killip
Series Lobatae Killip
Series Menispermifoliae Killip
12 - Subgenus Manicata (Harms) Escobar
13 - Subgenus Murucuja (Medic.) Masters
14 - Subgenus Plectostemma Masters
Section Cieca (Medic.) Mast.
Section Mayapathanthus Killip
Section Decaloba (DC.) Mast.
Series Auriculatae Killip
Series Heterophyllae Killip
Series Sexflorae Killip
Series Apetalae Killip
Series Luteae Killip
Series Organenses Killip
Series Miserae Killip
Series Punctatae Killip
Series Eudecaloba**
Section Hahniopathanthus (Harms) Killip
Section Hollrungiella**
Section Mayapathanthus (Harms) Killip
Section Pseudodysosmia (Harms) Killip
234
01
01
01
46
11
01
02
15
01
16
01
01
02
11
11
04
97
02
01
10
01
13
04
01
01
06
01
08
02
14
29
04
05
04
154
25
01
91
02
02
04
02
03
07
05
47
19
03
01
01
18
Annexes
______________________________________________________________
Section Pseudogranadilla (Harms) Killip
Section Xerogona (Raf.) Killip
15 - Subgenus Polyanthea (DC.) Killip
16 - Subgenus Porphyropathantus Escobar
17 - Subgenus Pseudomurucuja (Harms) Killip
18 - Subgenus Psilanthus, Killip
19 - Subgenus Rathea (Karsten) Killip
20 - Subgenus Tacsonia (Juss.) Triana & Planchon
Section Ampullacea Escobar
Section Boliviana Escobar
Section Bracteogama DC Prodr.
Section Colombiana Escobar
Series Leptomischae Escobar
Series Colombianae Escobar
Series Quindiensae Escobar
Section Fimbriatistipula Escobar
Section Parritana Escobar
Section Poggendorffia Triana & Planchon
Section Tacsonia Escobar
Section Tacsoniopsis Triana & Planchon
Section Trifoliata Escobar
21 - Subgenus Tacsonioides (DC.) Killip
22 - Subgenus Tacsoniopsis (Tr. & Planch) Killip
23 - Subgenus Tetrapathea*
24 - Subgenus Tryphostemmatoides (Harms) Killip
(Harms, 1925; de Candolle, 1822; Hutchinson, 1967; Green, 1972)* and (de Wilde, 1972)**.
235
06
08
01
01
04
04
02
48
01
01
12
19
08
09
02
02
02
01
05
03
02
05
01
01
03
Annexes
______________________________________________________________
Annex 2. Infrageneric classification according to Feuillet & MacDougal (2003).
Genus Passiflora L.
520 species
1- Subgenus Astrophea (DC.) Masters
Supersection Astrophea
Section Astrophea
Section Capreolata MacDougal & Feuillet
Section Leptopoda Killip ex Feuillet & Cremers
Supersection Pseudoastrophea (Harms) Feuillet & MacDougal
Section Pseudoastrophea (Harms) Killip
Section Botryastrophea (Harms) Killip
Series Botryastrophea (Harms) MacDougal & Feuillet
Series Carnae Feuillet
2- Subgenus Deidamioides (Harms) Killip
Section Polyanthea DC.
Section Deidamioides (Harms) Feuillet & MacDougal
Section Tetrastylis (Barb. Rodr.) Harms
Section Mayapathanthus MacDougal & Feuillet
Section Tryphostemmatoides Harms
3- Subgenus Decaloba (DC.) Rchb.
Supersection Pterosperma Gilbert & MacDougal
Supersection Hahniopathanthus (Harms) MacDougal & Feuillet
Supersection Disemma (Labill.) MacDougal & Feuillet
Section Octandranthus Harms
Section Disemma (Labill.) MacDougal & Feuillet
Section Hollrungiella Harms
Supersection Multiflora (Small) MacDougal & Feuillet
Supersection Auriculata MacDougal & Feuillet
Supersection Cieca (Medic.) MacDougal & Feuillet
Supersection Bryonioides (Harms) MacDougal & Feuillet
Supersection Decaloba (DC.) MacDougal & Feuillet
Section Decaloba DC.
Section Xerogona (Raf.) Killip
4- Subgenus Passiflora
Supersection Passiflora
Series Passiflora
Series Palmatisectae Feuillet & MacDougal
Series Pedatae Killip ex Cervi
Series Setaceae Killip ex Cervi
Supersection Stipulata Feuillet & MacDougal
Section Granadillastrum Triana & Planch.
Section Calopathanthus Harms
Section Tacsonioides DC.
Section Kermesinae (Cervi) Feuillet & MacDougal
Section Dysosmia DC.
Supersection Laurifolia (Cervi) Feuillet & MacDougal
Series Laurifoliae Killip ex Cervi
Series Quadrangulares Feuillet & MacDougal
Series Tiliifolia Feuillet & MacDougal
Series Marginatae Killip ex Cervi
Supersection Coccinea Feuillet & MacDougal
Supersection Distephana (DC.) Feuillet & MacDougal
Supersection Tacsonia (Juss.) Feuillet & MacDougal
Section Rathea (Karst.) Harms
Section Insignes Feuillet & MacDougal
236
57
27
10
15
02
30
17
13
06
07
13
01
01
02
02
07
214
04
05
21
17
03
01
09
08
18
20
119
106
13
236
19
13
01
01
04
95
66
01
04
04
20
42
21
06
14
01
14
05
61
03
05
Annexes
______________________________________________________________
Section Colombiana Escobar
Series Colombianae Escobar
Series Leptomischae Escobar
Series Quidiensae Escobar
Section Parritana Escobar
Section Fimbriatistipula Escobar
Section Tacsoniopsis Triana & Planch.
Section Elkea Feuillet & MacDougal
Section Tacsonia (Juss.) Harms
Section Boliviana (Harms) Feuillet & MacDougal
Section Trifoliata Feuillet & MacDougal
Section Manicata (Harms) Feuillet & MacDougal
237
19
09
08
02
02
02
02
15
05
02
01
05
Annexes
______________________________________________________________
Annex 3. Species cultivated in Colombia.
A.3.1. Passiflora edulis Sims
Plant essentially glabrous throughout (except ovary); trilobate leaves (5-25 x 5-20 cm.); sepals
white inside and green outside. This species presents two botanical form flavicarpa Degener
and edulis, originally differentiated by the color (yellow and purple) and size of the fruit (8-12
x 5-10 cm and 5-8 x 4-6 cm respectively).
The yellow maracuja, P. edulis f. flavicarpa is very probably native of Brazil and is the most
important passion fruit in the hot tropical areas, under the names of passion fruit, yellow
granadilla, yellow maracuja, maracuyá, parchita maracuyá. The yellow maracuja requires
high temperatures, between 20 and 34 ˚C, and develops better at lower altitudes. The principal
producer is Brazil with 450.000 t.y-1. Colombia reports 5.000 ha located from 0 to 1.200 m,
with a production of 17 to 20 t.ha-1.y-1, mainly in the departments of Huila and Valle del
Cauca. Between 60 and 70% of the Colombian production is processed into frozen juice,
exported to the European market and the remainder is consumed in natura.
The purple maracuja, P. edulis f. edulis is native of southern Brazil to northern Argentina and
Paraguay. It is now cultivated in most tropical areas with a mild climate (subtropics and
tropical highlands). In Colombia, it grows up to 2.500 m. This form is known under the
names of purple maracuja, gulupa, curuba redonda, maracuyá rojo, palchita. Colombia,
Australia, USA, Kenya, and Zimbabwe are the principal producers with approximately 20.000
t, although the total area is unknown. In Colombia, about 100 ha area cultivated between
1.700 and 2.500 m, with reported yields between 10 and 12 t.ha-1 .y-1.
According to Vanderplank (2000), the presently cultivated material originated from several
fruits, found in a London market, whose seeds were sent to Argentina. In 1915, the progeny
from these seeds passed through the USDA (United States Department of Agriculture) into
the United States, which redistributed them to Australia and New Zealand. Yellow maracuja
breeding has most often involved direct mass selection of this material with a narrow genetic
base and its hybridization with genotypes of the purple form obtained from materials exported
to Australia and Hawaii more than one hundred years ago.
238
Annexes
______________________________________________________________
The purple and yellow forms have frequently been crossed and spontaneous hybrids are found
in Hawaii and Australia. The direction of crossing determines its success: P. edulis f.
flavicarpa should be used as the male (Beal, 1975). The F1 hybrids are intermediate, normal
and vigorous (Nakasone et al., 1967). Meiosis is normal with the formation of nine bivalents,
but the chiasma frequency is lower in the hybrid than in the parents, which suggests slightly
reduced chromosomal homology. In the F2, 6 to 8% of the plants were found to be abnormal
(Beal, 1975). These data, as well as differences in the phenology and acclimatization of the
two forms, indicate the beginning of differentiation, tempered by the fact that equally
important divergences may be observed within the purple form in Brazil (Oliveira et al.,
1987).
A.3.2. Passiflora ligularis Juss.
Vernacular names: sweet granadilla, granadilla.
Origin: Andes of Colombia to Peru.
Distribution: Mexico to Peru.
Plant glabrous with broad heart-shaped leaves (rarely trilobate), flowers white or pinkish
white, fruit round to ovoid, 5 to 9 cm long, 4 to 7 cm in diameter, tapering towards the
peduncle, with a thin, hard and brittle pericarp, light brown to orange, sometimes tinged of
violet, with small light spots or stripes. Its light grey pulp is aromatic, slightly tangy and
relished by consumers in raw form. This species is usually cultivated at altitudes of 1.400 to
2.200 m. near the Equator, with extremes of 800 and 3.000 m. It grows at average
temperatures of 14 to 22oC and a relative humidity of 70%. It can tolerate short and very light
frosts. It is said to be tolerant to parasites and diseases of root and collar, but is susceptible to
withering due to Nectria haematococca (anamorphic Fusarium solani) in poorly drained soils.
A plantation is productive for four to eight years. The sweet granadilla commences flowering
in the ninth month and production starts 75 to 80 days later. Colombia is the principal
producer with 2.661 ha (2004) at a density of 400 plants per hectare and yields reach 45 to 50
t ha-1 y-1. Currently, Colombia is exporting this fruit to Europe. There are no commercial
varieties of the sweet granadilla. Although some superior types may be propagated by cuttings
or grafting, multiplication is usually by seed and the plant allogamy maintains considerable
variability in plantations.
239
Annexes
______________________________________________________________
A.3.3. Passiflora tripartita var. mollissima (Kunth) Holm-Nielsen & Jørgensen
Vernacular names: Banana passion fruit, curuba de Castilla, tacso, tumbo.
Origin: Andes of Venezuela to Bolivia (2.000 - 3.500 m).
Distribution: Venezuela to Argentina (mostly cultivated), New Zealand.
Plant pubescent, leaves deeply trilobate, 5-10 x 6-12 cm; flowers pendant, corolla variable in
size, with pink to deep red sepals and petals, 2.5 to 4 cm long. The crown is reduced to a
whorl of white tubercles on a red background. The floral tube is 6 to 11 cm long and 1.5 cm
wide; fruit oblong, 6-5 cm x 3-5 cm. with rounded ends, weighing 50 to 150 g (average 80 g).
The pericarp is pale yellow, occasionally green, somewhat pubescent, thin and supple but
coriaceous. It is rich in pectin and can be processed if in a perfect state, without anthracnose
spots. The pulp, constituting 60% of the fruit, is salmon-pink to dark orange, slightly acidic,
very pleasantly scented but generally astringent. The plant grows at altitudes between 2.000
and 3.000 m, sometimes up to 3.600 m, at average temperatures of 12 to 15oC and relative
humidity of 70 to 80%. It cannot adapt to warmer climates yet it is sensitive to extended
periods of frost. Production begins at about 18 months. Commercial cultivation began in the
1950s. In Colombia, 1.795 ha are cultivated, with yield ranging from 7 t.ha-1.y-1 to 45 t, the
latter under the most suitable conditions. The major phytosanitary problems are anthracnose,
which spoils the fruit, and nematodes of the genus Meloidogyne. Being an allogamous
species, its repeated self-fertilization results in a serious loss of vigor (Schöniger, 1986).
A.3.4. Passiflora tarminiana Coppens & Barney
Vernacular names: curuba india, tacso, tumbo
Distribution: Venezuela to Argentina, as a cultigen, naturalized in the USA (Hawaii; invasive
species), New Zealand, Philippines, Ceylon, Papua New Guinea, tropical highlands of East
Africa.
Plant glabrous; leaves trilobate, 7-9 x 12-18 cm; flowers pendent, petals bright pink to light
pink; fruit fusiform 10-14 x 3.5-4.5 cm, pericap yellow to orange with dots except along the
main vascular bundles. P. tarminiana is adapted to a wide range of elevations as compared to
other species of the subgenus Tacsonia and have been introduced into numerous cool tropical
and tropical montane areas other than its original Andean range, in some cases becoming an
invasive weed. The hybrids with P. mixta and P. tripartita var. mollissima are fertile and
show intermediate phenotypes (Primot et al., 2005). This species is tolerant to Alternaria and
240
Annexes
______________________________________________________________
Colletrotrichum. In Colombia, it is cultivated between 2.000 and 2.700 m with yields around
20-25 t.ha-1.y-1, although the cultivated area is not known, as it is frequently confused with P.
tripartita var. mollissiima.
a
b
c
d
e
Figure 1. Species cultivated in Colombia: (a) P. edulis f. flavicarpa; (b) P. edulis f. edulis; (c) P. ligularis; (d)
P. tripartita var. mollissima; (e) P. tarminiana.
241
Annexes
______________________________________________________________
A.3.5. Passiflora quadrangularis L.
Vernacular names: giant granadilla, badea, barbadine.
Distribution: West Indies to South America
Plant glabrous, stem stout, quadrangular, reaching 50 m in length; leaves simple, oval or
oblanceolate, 10 to 25 cm long and 8 to 18 cm broad; the flowers may reach 12 cm in
diameter: the inner face of the sepals and petals is white, pink, red or violet; the crown may
reach 6 cm in length; the fruit is yellowish-green, sometimes with a pinkish tint, ovoid to
oblong. It is 20 to 30 cm long, 10 to 18 cm broad, and weighs 2.8 kg on average, even
reaching 4 kg. It develops in 62 to 85 days. The fruit mesocarp is 2 to 3 cm thick, soft and
edible but bland. The pulp is transparent, white to orange, sweet and slightly tangy, the taste
variable but always less pungent than that of maracuja. Manual pollination is often
recommended. This species is tolerant to Alternaria passiflorae (McMillan & Graves 1992)
and resistant to withering but highly susceptible to nematodes and Xanthomonas spp.
(Oliveira & Ferreira, 1991; Vanderplank 2000). In Colombia 60 ha are cultivated between 0
to 1.000 m, with a yield of 16 to 18 t. ha-1 y-1.
A.3.6. Passiflora maliformis L.
Vernacular names: stone granadilla, conch apple, granadilla de piedra, chulupa, coque en fer.
Origin: West Indies to Ecuador.
Distribution: West Indies to Ecuador, North Brazil.
Plants glabrous or finely pilosulous with leaves ovate, ovate-lanceolate, or sometimes
orbicular ovate; flowers petals densely mottled with dark red-purple; fruits globose, 4 to 6 cm
in diameter, green or orange green and rarely purple, pericarp hard to extremely hard.
The juice has excellent flavor, is very refreshing and valuable due its high content of ascorbic
acid. This plant occurs naturally at mid-elevations (0 to 1.600 m), but in Colombia it is
cultivated at 600 to 1.000 m on 97 ha, with yields of 11 to 14 t ha-1 y-1, mainly in the Huila
department where there is a local market.
A.3.7. Passiflora alata Curtis
Vernacular names: Fragrant granadilla, maracuja doce, Maracua.
Origin: Brazilian and Peruvian Amazon, Brazilian Planalto in forest galleries.
Distribution: Brazil, Peru and Colombia.
242
Annexes
______________________________________________________________
Plant glabrous with stem stout and quadrangular; leaves ovate or ovate-oblong 8-15 x 7-10
cm; flowers petals red inside and green outside; fruit ovoid or pyriform, 8 to 10 cm. long,
yellow. In many regions of South America, P. alata is cultivated because of its edible fruit.
Brazil is the principal producer with 50 t.ha-1. y-1 and an area of 120 ha. In Colombia, it has
been cultivated traditionally in the Amazon and commercially in the North of the Valle del
Cauca and Quindío departments from 1999, but its yield statistics are unknown. This species
has given a number of hybrids, mainly with the closely related P. quadrangularis.
A.3.8. Passiflora popenovii Killip
Vernacular names: Granadilla de Quijos, granadilla caucana, curubejo.
Distribution: South of Colombia and Ecuador.
Plant glabrous, except the ovary and the outer surface of the calyx tube; leaves unlobed,
oblong-ovate, 8.5 to16 cm long; fragrant flowers with white petals; fruit ovoid, 6 to 10 cm in
diameter, epicarp thin and yellow. The sweet pulp and juice are highly regarded and is
enjoyed for its rich aroma and taste. This species is cultivated in Colombia (Cauca and
Nariño) and Ecuador (Tungurahua, Napo, Loja) between 1.400 and 2.100 m. and its fruits are
offered in local markets in the months of April and May. This plant is regularly propagated by
cutting. In Colombia, yields of 140 kg/plant are reported, but the cultivated area is not well
known.
243
Annexes
______________________________________________________________
a
b
c
d
Figure 2. Species cultivated in Colombia: (a) P. quadrangularis; (b) P. alata; (c) P. maliformis;
(d) P. popenovii.
244
Annexes
______________________________________________________________
Annex 4. Passiflora molecular diversity. Dendrograms obtained in previous studies.
A.a. Phylogenetic tree sensu Muschner et al. (2003). Maximum-likelihood tree for the ITS spacer in Passiflora
and outgroups. Numbers above branches are bootstrap support values (when higher than 50%) based on 1000
replicates. Abbreviations indicate the subgenera as follows: PASS 5 Passiflora; DIST 5 Distephana; TACS 5
Tacsonioides; DYSO 5 Dysosmioides; DYSA 5 Dysosmia; ASTR 5 Astrophea; DECA 5 Decaloba; MURU 5
Murucuja; DEID 5 Deidamioides; PSMU 5 Pseudomurucuja; ADOP 5 Adopogyne.
245
Annexes
______________________________________________________________
A.b. Phylogenetic tree sensu Yockteng (2003). Strict consensus tree of 3,536 most parsimonious trees based on
matk) sequences. Bootstrap support values (BS) greater than
the alignment of chloroplast-expressed mutarase K (m
70% and decay indices greater than 4 are indicated above branches. Thick lines indicate BS values of 100%, dots
indicate nodes supported by 100% in the Bayesian tree. The subdivisions according to Killip (1938) are indicated
for each taxon. The sections and series names are abbreviated as follows: Cie, Cieca; Dec, Decaloba; Auri,
Auriculatae; Sexf, Sexflorae; Orga, Organenses; Mise, Miserae; Punc, Punctatae; Eud, Eudecaloba; Xero,
Xerogona; Psedy, Pseudodysosmia; Psegr, Pseudogranadilla; Hah, Hahniopathanthus; Lept, Leptomischae; Col,
Colombianae; Bract, Bracteogama; Tac, Tacsonia; Quad, Quadrangulares; Digi, Digitatae; Tili, Tiliafoliae;
Laur, Laurifoliae; Inca, Incarnatae; Simp, Simplicifoliae; Loba, Lobatae; Meni, Menispermifoliae; Doli,
Dolichostemma; Euas, Euastrophea; Pseas, Pseudoastrophea. The three clades emerging from the analysis are
indicated on the right.
246
Annexes
______________________________________________________________
A.c. Phylogenetic tree sensu Yockteng & Nadot (2004). Strict consensus tree of 32,573 most parsimonious trees
ncpGS) sequences. Bootstrap support
based on the alignment of chloroplast-expressed glutamine synthetase (n
values (BS) greater than 70% and decay indices greater than 4 are indicated above branches. Thick lines indicate
BS values of 100%, dots indicate nodes supported by 100% in the Bayesian tree. The subdivisions according to
Killip (1938) are indicated for each taxon. The sections and series names are abbreviated as follows: Cie, Cieca;
Dec, Decaloba; Auri, Auriculatae; Sexf, Sexflorae; Orga, Organenses; Mise, Miserae; Punc, Punctatae; Eud,
Eudecaloba; Xero, Xerogona; Psedy, Pseudodysosmia; Psegr, Pseudogranadilla; Hah, Hahniopathanthus; Lept,
Leptomischae; Col, Colombianae; Bract, Bracteogama; Tac, Tacsonia; Quad, Quadrangulares; Digi, Digitatae;
Tili, Tiliafoliae; Laur, Laurifoliae; Inca, Incarnatae; Simp, Simplicifoliae; Loba, Lobatae; Meni,
Menispermifoliae; Doli, Dolichostemma; Euas, Euastrophea; Pseas, Pseudoastrophea. The three clades
emerging from the analysis are indicated on the right.
247
Annexes
______________________________________________________________
A.d. Phylogenetic tree sensu Hansen et al. (2006). Phylogram of 1 of 200,000 trees from analysis A of trnL/trnT
sequence data illustrating the distribution of chromosome numbers in the genus. Branch lengths represent
character state changes and the number of changes is listed above each branch.
248
Annexes
______________________________________________________________
Annex 5. List of morphological descriptors in the genus Passiflora L.
PASSPORT
INSTITUTE CODE:
______________________________________________
ACCESSION NUMBER:
______________________________________________
ACCESSION NAME:
______________________________________________
STATUS OF SAMPLE:
______________________________________________
SCIENTIFIC NAME:
______________________________________________
VERNACULAR NAME:
______________________________________________
CULTIVAR ORIGIN:
1. Open pollination:
2. Artificial pollination:
3. Clonal selection:
4. Seedling selection:
______________________________________________
______________________________________________
______________________________________________
______________________________________________
COLLECTING DESCRIPTION:
1. Country of origin:
2. Province / state / Deparment:
3. Country
4. Collection site:
5. Latitude:
6. Longitude:
7. Altitude:
______________________________________________
______________________________________________
______________________________________________
______________________________________________
______________________________________________
______________________________________________
______________________________________________
ENTHNOBOTANICAL DATA:
______________________________________________
______________________________________________
COLLECTOR:
______________________________________________
DATE:
______________________________________________
NOTES:
______________________________________________
______________________________________________
______________________________________________
249
Annexes
______________________________________________________________
1. STEM_________________________________________________________
1.1 HABIT (STHA)
1. Liana (eg. P. incarnata)
2. Tree (eg. P. emarginata)
3. Shrub (eg. P. macrophylla)
1.2 EXTERNAL SHAPE (STSH)
1. Round
2. Striate
3. Subangulate
4. Angulate (e.g. P. quadrangularis)
1
2
3
4
5. Other (Specify)
1.3 ANTHOCYAN (STAN)
1. Absent
2. Medium (< 80%)
3. High (> 80%)
1.4 PUBESCENCE (STPU)
1. Glabrous
2. Few density
3. Tomentoso
4. Villous
5. Pilose
1
2
3
4
1.5 INTERNODE LENGTH (STIN)
mm
mm
mm
mm
mm
mm
mm
1.6 DIAMETERS (STDI)
mm
mm
mm
2. TENDRIL______________________________________________________
2.1 TENDRILS (TEPRE)
1. Absent
2. Present
2.2 SPIRAL SHAPE (TESH)
1. Cylindrical
2. Conic
3. Compound
1
250
2
Annexes
______________________________________________________________
4. Linear
5. Indefinite
2.3 PUBESCENCE (TEPU)
1. Glabrous
2. Low density
3. Tomentose
4. Villous
5. Pilose
1
2
3
4
5
2.4 ANTHOCYANIN (TEAN)
1. Absent
2. Medium (< 80%)
3. High (> 80%)
3. STIPULE______________________________________________________
3.1 STIPULES (TEPR)
1. Present
2. Absent
3.2 DURATION (SPPE)
1. Permanent
2. Deciduous
3.3 COLOR (SPCO)
3.4 PUBESCENCE (SPPU)
1. Glabrous
2. Low density
3. Tomentose
4. Villous
5. Pilose
3.5 SHAPE (SPSH)
1. Setaceous
2. Linear
3. Pinnatisect
4. Lobed
5. Lanceolate
6. Oblate
7. Ovate
8. Reniform (or auricular)
1
2
3
251
4
5
6
7
8
Annexes
______________________________________________________________
9. Other (specify)
3.6 MARGIN (SPMA)
1. Entire
2. Serrate
3. Serrulate
4. Dentate
5. Doubly dentate
6. Crenate
7. Other (specify)
1
2
3
4
5
6
3.7 ANTHOCYANIN (SPAN)
1. Absent
2. Medium (< 80%)
3. Heigth (> 80%)
3.8 LENGTH (SPLE)
mm
mm
mm
mm
mm
3.9 WIDTH (SPWI) (including the arist)
mm
mm
mm
mm
mm
3.10 TERMINAL ARIST LENGTH (SPTA)
mm
mm
mm
mm
mm
4. PETIOLE______________________________________________________
4.1 ANTHOCYANIN (PEAN)
1. Absent
2. Medium (< 80%)
3. High (> 80%)
4.2 PUBESCENCE (PEPU)
1. Glabrous
2. Low density
3. Tomentose
4. Villous
5. Pilose
4.3 COLOR (PECO)
252
Annexes
______________________________________________________________
4.4 NECTARY SHAPE (PENS)
1. Discoid (e.g. P. edulis f. flavicarpa)
2. Extended (e.g. P. ligularis)
3. Spherical (e.g. P. adenopoda)
4. Other
1
4.5 STIPE OF NECTARIES (PESN)
1. Absent
2. Intermediate
3. Present
2
1
3
2
3
4.6 LENGHT (PELE)
mm
mm
mm
mm
mm
4.7 DISTANCE FROM BASE TO FIRST GLAND (PEDG)
mm
mm
mm
mm
mm
4.8 NECTARY GLAND NUMBER (PENM)
5. LEAF_______________ ____ ____________________________________
5.1 HETEROPHYLLY (LEPO)
1. Absent
2. Present
5.2 LOBE NUMBER (LELN)
5.3 MARGIN (LEMA)
1. Entire
2. Serrate
3. Serrulate
4. Dentate
5. Doubly dentate
6. Crenate
1
253
2
3
4
5
6
Annexes
______________________________________________________________
7. Other (specify)
5.4 BASE SHAPE (LEBS)
1. Cuneate
2. Rounded
3. Truncate
4. Cordate (Heart-shaped)
5. Deeply cordate
6. Articulate
7. Other (specify)
1
2
5.5 APEX SHAPE (LEAS)
1. Rounded
2. Obtuse (>90o)
3. Acute
4. Acute (<45o)
5. Other (specify)
3
1
2
5.6 ACUMEN (LEPA)
1. Absent
2. Presence
2
5.7 PUBESCENCE ADAXIAL (LEAX)
1. Glabrous
2. Low density
3. Tomentose
4. Villous
5. Pilose
5.8 PUBESCENCE ABAXIAL (LEPB)
1. Glabrous
2. Low density
3. Tomentose
4. Villous
5. Pilose
5.9 ANTHOCYANIN - LAMINA (LEAL)
1. Absent
2. Medium (< 80%)
3. High (> 80%)
5.10 ANTHOCYANIN - NERVES (LEAN)
1. Absent
2. Medium (< 80%)
3. High (> 80%)
5.11 COLOR – ADAXIAL (LECA)
254
4
5
3
6
4
Annexes
______________________________________________________________
5.12 MARGIN SERRATION DENSITY (number on 2 cm) (LEMS)
mm
mm
mm
mm
mm
5.13 ANGLE BETWEEN LATERAL LOBES (LEAB)
mm
mm
mm
mm
mm
5.14 CENTRAL LOBE LENGTH (LELC)
mm
mm
mm
mm
mm
5.15 RIGHT LOBE LENGTH (LERL)
mm
mm
mm
mm
mm
5.16 CENTRAL LOBE WIDTH (LECL)
mm
mm
mm
mm
mm
5.17 DISTANCE BETWEEN LEAF SINUS AND PETIOLE INSERTION (LESS)
mm
mm
mm
mm
mm
5.18 HETEROBLASTY (LEPH)
0. Absent
1. Present
5.19 LAMINAR NECTARIES (LENL)
1. Absent
2. Present
5.20 NECTAR GLAND NUMBER ON LAMINA (LELA)
5.21 DISTRIBUTION OF LAMINAR NECTARIES (LEDN)
1. Along central lobe
255
Annexes
______________________________________________________________
2. On lateral lobes
3. Around lamina
4. Around apex
5. At base
5.22 PRESENCE OF MARGINAL NECTARIES (LEPN)
1. Absent
2. Present
5.23 NECTAR NUMBER ON LEAF MARGIN (LENN)
5.24 DISTRIBUTION OF LAMINAR MARGIN NECTARIES (LELM)
1. Base
2. Base and sinus
3. Margin
4. Other (specify)
5.25 NECTARY SHAPE (LENS)
1. Discoid (squashed)
2. Extended
3. Spherical
4. Other
1
2
3
6. PEDUNCLE____________________________________________________
6.1 PUBESCENCE (PDPU)
1. Glabrous
2. Low density
3. Tomentose
4. Villous
5. Pilose
6.2 ANTHOCYANIN (PDAN)
1. Absent
2. Medium (< 80%)
3. High (> 80%)
6.3 BIFURCATION (PDBN)
1. Present
2. Absent
6.4 LENGTH (PDLE)
256
Annexes
______________________________________________________________
mm
mm
mm
mm
mm
mm
mm
6.5 DIAMETER (PDDI)
mm
mm
mm
6.6 PEDICEL LENGTH (PDPL)
mm
mm
mm
mm
mm
6.7 LENGTH TO FIRST BIFURCATION (PDLF)
mm
mm
mm
mm
mm
6.8 LENGTH TO SECOND BIFURCATION (PDBS)
mm
mm
mm
mm
mm
7. BRACT________________________________________________________
7.1 UNION (BRUN)
1. Free
2. United at less than half
3. United at half
4. United at more than half
7.2 BRACT (BRPR)
1. Present
2. Absent
7.3 PERMANENCE (BRPE)
1. Permanent
2. Deciduous
7.4 PUBESCENCE (BRPU)
1. Glabrous
2. Low density
3. Tomentose
4. Villous
5. Pilose
257
Annexes
______________________________________________________________
7.5 POSITION OF PUBESCENCE (BRPP)
1. Absent
2. Adaxial
3. Abaxial
4. Both sides
7.6 COLOR (BRCO)
7.7 ANTHOCYANIN (BRAN)
1. Absent
2. Medium (< 80%)
3. High (> 80%)
7.8 SHAPE (BRSH)
1. Setaceous
2. Linear
3. Pinnatisect
4. Lobate
5. Lanceolate
6. Oblate
7. Ovate
9. Other (specify)
1
2
3
4
5
6
7.9 MARGIN (BRMA)
1. Entire
2. Serrate
3. Serrulate
4. Dentate
5. Doubly dentate
6. Crenate
7. Other (specify)
7.10 APEX SHAPE (BRAS)
1. Round
2. Obtuse (>90 o)
3. Acute
4. Very acute (<45 o)
5. Other
1
2
7.11 MARGIN NECTARIES (BRNM)
1. Present
2. Absent
7.12 LENGTH (BRLE)
mm
mm
mm
mm
mm
258
3
4
7
Annexes
______________________________________________________________
7.13 WIDTH (BRWI)
mm
mm
mm
mm
mm
8. FLOWER_______________________________________________________________
8.1 CORONA TYPE (FLCY)
1. Tuberculous
2. Filamentous
8.2 COROLLA TYPE (FLCT)
1. Campanulate (e.g. P. mixta)
2. Intermediate (eg. P. manicata)
3. Reflex (eg. P. tarminiana)
4. Other (specify)
1
8.3 ORIENTATION (FLOF)
1. Pendular (eg. P. antioquiensis)
2. Intermediate (eg. P. mixta)
3. Erect (eg. P. manicata)
8.4 ORIENTATION (in Degrees to vertical) (FLOG)
8.5 PUBESCENCE ON COROLLA (FLPU)
1. Glabrous
2. Low density
3. Tomentose
4. Villous
5. Pilose
8.6 HYPHANTHIUM PUBESCENCE (FLHP)
1. Glabrous
2. Low density
3. Tomentose
4. Villous
5. Pilose
8.7 DOMINANT PETAL COLOR (FLCP)
259
2
3
Annexes
______________________________________________________________
8.8 CHLOROPHYLL ON EXTERIOR OF SEPAL (FLCS)
1. Absent
2. Present
8.9 KEEL-SHAPED SEPALS (FLKS)
1. Absent
2. Present
8.10 SEPAL AWN (CAS)
1. Absent
2. Present
8.11 UNION OF SEPALS (FLUS)
1. Free
2. United
8.12 PETALS (FLPP)
1. Present
2. Absent
8.13
PETAL LENGTH (FLPL)
mm
mm
mm
mm
mm
8.14 PETAL WIDTH (FLPW)
mm
mm
mm
mm
mm
8.15 SEPAL LENGTH (FLSL)
mm
mm
mm
mm
mm
8.16 SEPALS WIDTH (FLSW)
mm
mm
mm
mm
mm
8.17 DIAMETER OF NECTARY CHAMBER (FLNC)
mm
mm
mm
mm
mm
8.18 HYPANTHIUM DIAMETER ABOVE NECTARY CHAMBER (FLHD)
mm
mm
mm
mm
mm
260
Annexes
______________________________________________________________
8.19 HYPANTHIUM DIAMETER – DISTAL (FLHS)
mm
mm
mm
mm
mm
8.20 FLOWER LENGTH (FLLE)
mm
mm
mm
mm
mm
8.21 HYPHANTIUM LENGHT (FLHL)
mm
mm
mm
mm
mm
8.22 LENGHT OF NECTARY CHAMBER (FLCN)
mm
mm
mm
mm
mm
8.23 NUMBER OF CORONA SERIES (FLNS)
8.24 COLOR OF FILAMENTS AT BASE (FLCB)
8.25 COLOR OF FILAMENTS AT APEX (FLCA)
8.26 FILAMENT LENGTH (FLFL)
mm
mm
mm
mm
mm
8.27 DISTRIBUTION OF ANTHERS (FLDA)
1. Symmetry radial
2. Symmetry bilateral
3. Other
8.28 COLOR OF STAMINAL FILAMENTS (CFE)
261
Annexes
______________________________________________________________
8.29 STAMINAL FILAMENTS LENGTH (FLSF)
mm
mm
mm
mm
mm
8.30 OVARY PUBESCENCE (FLOP)
1. Glabrous
2. Low density
3. Tomentose
4. Villous
5. Pilose
8.31 COLOR OF OVARY (FLCO)
8.32 OVARY LENGTH (FLOL)
mm
mm
mm
mm
mm
8.33 COLOR OF STYLE (FLCS)
8.34 STYLE LENGHT (FLSL)
mm
mm
mm
mm
mm
8.35 COLOR DISTRIBUTION ON STYLES (FLDS)
1. Uniform
2. Specked
3. Apex
4. Base
5. Other
8.36 COLOR OF STIGMA (FLCG)
8.37 COLOR OF ANDROGYNOPHORE (FLCN)
262
Annexes
______________________________________________________________
8.38 COLOR DISTRIBUTION ON ANDROGYNOPHORE (FLDN)
1. Uniform
2. Speckled
3. Other
8.39 PUBESCENCE OF ANDROGYNOPHORE (FLPN)
1. Glabrous
2. Few density
3. Tomentose
4. Villous
5. Pilose
8.40 GYNOPHORE LENGTH (FLGL)
mm
mm
mm
mm
mm
8.41 ANDROGYNOPHORE LENGTH (FLAL)
mm
mm
mm
mm
mm
8.42 OPERCULUM LENGTH (FLOP)
mm
mm
mm
mm
mm
8.43 LIMEN LENGTH (FLLL)
mm
mm
mm
mm
mm
8.44 LIMEN MARGIN (FLML)
1. Flat-entire
2. Flat-serrate
3. Wavy-entire
4. Wavy-serrate
8.45 NECTARY CHAMBER RING (FLNR)
1. Absent
2. Present
8.46 HYPANTHIUM TYPE (FLHY)
1. Flat (eg. P. suberosa)
2. Campanulate (P. ligularis)
3. Tubular (P. tarminiana)
263
Annexes
______________________________________________________________
8.47
INTERNAL COLOR OF HYPHANTIUM (FLCI)
8.48 CLOROPHYLLA ON EXTERIOR OF HYPANTHIUM (FLCE)
1. Absent
2. Partial
3. Global
8.49 ANTHOCYANIN ON EXTERIOR OF HYPANTHIUM (FLAE)
1. Absent
2. Partial
3. Global
8.50 STYLES NUMBER PER FLOWER (NEF)
8.50 NECTARS ON SEPALS (FLNS)
1. Absent
2. Present
8.51 DOMINANT SEPAL COLOR (FLCP)
8.52 ANTHOCYANIN ON EXTERIOR OF SEPALS (FRSH)
1. Absent
2. Partial
3. Global
9. FRUIT________________________________________________________
9.1 TYPE (FRTY)
1. Berry
2. Capsule
9.2 PIGMENTATION OF INMATURE FRUITS (FPFIM)
1. Uniform
2. Dotted
3. Lined
4. Other
9.3 DOMINANT COLOR OF THE RIPE FRUIT (FRCF)
264
Annexes
______________________________________________________________
9.4 SHAPE (FRSH)
1. Spherical / Round
2. Ovoid
3. Oblate
4. Oblong
5. Ellipsoid
6. Fusiform
7. Other
1
2
3
4
5
9.5 PUBESCENCE (FRPU)
1. Absent
2. Present
9.6 WEIGHT (FRWH)
g
g
g
g
g
9.7 LENGTH (FRLE)
mm
mm
mm
mm
mm
mm
mm
9.8 DIAMETER (FRDM)
mm
mm
mm
9.9 TRANSVERSAL SECTION (FRTS)
1. Round
2. Hexagonal
3. Triangular
4. Other (specify)
1
9. PERICARP DIAMETERS (FREM)
mm
mm
mm
mm
mm
9.11 MESOCARP TEXTURE (FRMT)
1. Hard
2. Soft rough
3. Soft
4. Other
265
2
3
6
Annexes
______________________________________________________________
9.12 ARIL TASTE (FRAT)
1. Acid
2. Acid sweet
3. Sweet
4. Insipid
5. Other
9.13 SKIN WEIGHT (FRSW)
g
g
g
g
g
9.14 SEEDS WEIGHT (FRSW)
g
g
g
g
g
9.15 WEIGHT OF JUICE AND PULP (FRJP)
g
g
g
g
g
9.16 SKIN TEXTURE (FRST)
1. Soft
2. Resistant to the pressure
3. Brittle
9.17 JUICE PH (FRPH)
9.18 TITRABLE ACIDITY (FRAT)
mg/100ml
9.19 TOTAL SOLUBLE SOLIDS (FRTS)
9.20 ASCORBIC ACID (FRVC)
mg/100ml
9.21 JUICE AROMA (FRAJ)
1. Weak
2. Intermediate
3. Strong
266
Annexes
______________________________________________________________
10. SEED_________________________________________________________
10.1 SURFACE TYPE (SEST)
1. Smooth
2. Reticulate
3. Other
10.2 ARIL COLOR (SEAC)
10.3 COAT COLOR (SECC)
10.4 SEED BRIGHTNESS (SEBS)
1. Mate
2. Intermediate
3. Brilliant
10.5 SEED SHAPE (SESS)
1. Obovate
2. Cordate
3. Pyramidal
4. Obcordate
5. Cuneate
6. Other (specify)
1
2
10.6 WEIGHT (100 seeds) (SEWH)
mm
mm
mm
mm
mm
mm
mm
mm
mm
10.7 LENGTH (SELE))
mm
mm
mm
10.8 WIIDTH (SEWI)
mm
mm
mm
10.9 NUMBER OF SEEDS PER FRUIT (SENF)
g
g
g
g
g
267
3
4
5
Annexes
______________________________________________________________
11. POLLINATION SYNDROME____________________________________
1.
2.
3.
4.
5.
6.
7.
Hummingbirds (e.g. P. mixta, P. vitifolia).
Bats (e.g. P. mucronata, P. lobata).
Hummingbirds and bats (e.g. P. penduliflora).
Bees (e.g. P. alnifolia, P. filipes).
Wasps (e.g. P. edulis f. flavicarpa, P. quadrangularis).
Bees and Wasps (e.g. P. foetida, P. sphaerocarpa).
Others (specify).
12. PEST AND DISEASE SUSCEPTIBILITY___________________________
12.1 PEST
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________
12.2. FUNGI
_____________________________________________________________________________________________
_____________________________________________________________________________________________
_____________________________________________________________________________________________
12.3 BACTER
_____________________________________________________________________________________________
_____________________________________________________________________________________________
_____________________________________________________________________________________________
12.4 VIRUS
_____________________________________________________________________________________________
_____________________________________________________________________________________________
_____________________________________________________________________________________________
12.5 ABIOTIC STRESS SUSCEPTIBILITY
_____________________________________________________________________________________________
_____________________________________________________________________________________________
_____________________________________________________________________________________________
12.6 OTHERS DISORDERS
_____________________________________________________________________________________________
_____________________________________________________________________________________________
_____________________________________________________________________________________________
268
View publication stats