Ecología de la dispersión de plantas en los bosques secos...
173
Ecología de la dispersión de plantas en los bosques secos del suroccidente Ecuatoriano Tesis Doctoral Andrea Jara Guerrero 2014
Transcript of Ecología de la dispersión de plantas en los bosques secos...
Ecología de la dispersión de plantas en los bosques secos del suroccidente Ecuatoriano
Tesis Doctoral Andrea Jara Guerrero
2014
Departamento de Biología Vegetal, Escuela Técnica
Superior de Ingenieros Agrónomos
Tesis Doctoral
Ecología de la dispersión de plantas en los
bosques secos del suroccidente
Ecuatoriano
Autor: Andrea Katherine Jara Guerrero1
Directores: Dr. Marcelino de la Cruz Rot2, Dr. Marcos
Méndez Iglesias3
1Departamento de Ciencias Naturales. Universidad Técnica Particular de Loja.
2Departamento de Biología Vegetal, Universidad Politécnica de Madrid.
3Área de Biodiversidad y Conservación. Departamento de Biología y Geología, ESCET,
Universidad Rey Juan Carlos.
Madrid, 2014
i
Marcelino de la Cruz Rot, Profesor Titular de Universidad del Departamento de Biología
Vegetal de la Universidad Politécnica de Madrid y Marcos Méndez Iglesias, Profesor
Titular de Universidad del Departamento de Biología y Geología de la Universidad Rey
Juan Carlos
CERTIFICAN:
Que los trabajos de investigación desarrollados en la memoria de tesis doctoral:
“Ecología de la dispersión de plantas en los bosques secos del suroccidente
Ecuatoriano”, son aptos para ser presentados por Andrea Katherine Jara Guerrero ante el
Tribunal que en su día se consigne, para aspirar al Grado de Doctor por la Universidad
Politécnica de Madrid.
VºBº Director Tesis VºBº Director de Tesis
Dr. Marcelino de la Cruz Rot Dr. Marcos Méndez Iglesias
ii
A mi familia, mi madre y mi hermana,
por su apoyo incondicional en todo momento.
iii
AGRADECIMIENTOS
A mi familia, por su paciencia y respaldo incondicional desde siempre. Principalmente a mi
Madre, por toda su ayuda y el tiempo invertido en elaborar trampas de semillas, sembrar
mis muestras, regar plantas, etc.
A mis profesores, Marcelino y Marcos, no solo por el apoyo científico, sino también por el
entusiasmo transmitido durante todo este proceso. Igualmente a Adrián Escudero, por su
disponibilidad para colaborar en mi proceso de aprendizaje y por sus aportes que sin duda
han contribuido a mejorar este trabajo.
Un agradecimiento especial a mi equipo de trabajo en la UTPL, principalmente.a Carlos
Iván, que informalmente ha sido mi tutor en Ecuador y un gran respaldo. A Jorge, Dalton,
Elizabeth, Ángel, Diego, Pablo y a todos mis compañeros de trabajo que en algún momento
sacrificaron parte de su tiempo para auxiliarme con el trabajo de campo y con el análisis de
datos. Igualmente a los miembros de las Fuerzas Armadas del destacamento Pintag Nuevo
por su importante colaboración logística durante las etapas más duras del trabajo de campo.
Finalmente, a la Universidad Técnica Particular de Loja, por el aporte financiero para el
desarrollo de esta investigación, pero principalmente por permitir e incentivar mi formación
como investigadora.
iv
ÍNDICE
RESUMEN ............................................................................................................................................ 1
ABSTRACT ............................................................................................................................................ 4
INTRODUCCIÓN ................................................................................................................................... 7
OBJETIVOS: ........................................................................................................................................ 13
METODOLOGÍA GENERAL Y ÁREA DE ESTUDIO ................................................................................ 16
CAPÍTULO 1 ....................................................................................................................................... 28
Seed Dispersal Spectrum of Woody Species in South Ecuadorian Dry Forests: environmental
correlates and the effect of considering species abundance. .......................................................... 28
CAPÍTULO 2 ....................................................................................................................................... 80
Does spatial heterogeneity blur the signature of dispersal syndromes on spatial patterns of woody
species? A test for a tropical dry forest ............................................................................................ 80
CAPÍTULO 3 ..................................................................................................................................... 112
Seed rain and seed bank contribution to regeneration of woody species with different dispersal
syndrome in an ecuadorian tropical dry forest ............................................................................... 112
CAPÍTULO 4 ..................................................................................................................................... 142
Legitimidad de los cérvidos como dispersores de semillas de especies leñosas en un bosque seco
tropical ............................................................................................................................................ 142
CONCLUSIONES GENERALES
......................................................................................................................................................... 165
1
RESUMEN
La importancia del proceso de dispersión de semillas en la estructura y dinámica de los
ecosistemas es ampliamente reconocida. Sin embargo, para los bosques tropicales
estacionalmente secos los estudios relacionados con este proceso son aún escasos y
dispersos en comparación con los bosques tropicales lluviosos. En este trabajo se estudió la
importancia de los síndromes de dispersión de semillas en la estructuración de
comunidades, mediante el análisis de los patrones de dispersión de semillas en el espacio y
tiempo para comunidades de leñosas en los bosques secos del suroccidente Ecuatoriano.
Esta área forma parte de la región Tumbesina, una de las áreas de endemismo más
importantes del mundo, pero también uno de los hotspots más amenazados. El clima se
caracteriza por una estación seca que va de mayo a noviembre y una estación lluviosa que
se extiende desde diciembre a abril. Para toda esta zona se estima una temperatura
promedio anual entre 20° y 26°C y una precipitación promedio anual entre 300 y 700 mm.
El trabajo de campo se desarrolló entre febrero de 2009 y septiembre de 2012. El primer
paso fue la recopilación de información sobre las especies leñosas nativas de los bosques
secos del suroccidente de Ecuador, que permitiera asignar a cada especie a un síndrome de
dispersión para determinar el espectro de síndromes de dispersión de semillas. Luego,
utilizando la información disponible de 109 parcelas establecidas previamente a lo largo de
cuatro cantones de la provincia de Loja que conservan bosques secos en buen estado, se
analizó la relación entre el síndrome de dispersión y condiciones ambientales.
La relación de los síndromes de dispersión con los patrones espaciales de las especies y con
los patrones de la lluvia y banco de semillas se estudió dentro de una parcela permanente de
2
9 ha, en la Reserva Ecológica Arenillas. Dentro de esta parcela se estableció un transecto de
aproximadamente 3,4 km, que se recorrió mensualmente para colectar excretas de cérvidos
y analizar el rol de este grupo como dispersor de semillas.
Una gran variedad de plantas en los bosques secos tropicales del suroccidente de Ecuador
requirió la asistencia de animales para la dispersión de semillas. Sin embargo, un análisis
del espectro de dispersión considerando no solo la riqueza, sino también la abundancia
relativa de especies, permitió determinar que a pesar de la alta variedad de especies
zoócoras, la mayor parte de la comunidad correspondía a individuos anemócoros, que no
proveen ninguna recompensa para la dispersión por animales. Este patrón puede deberse a
la abundancia relativa de hábitats adecuados para especies con diferente síndrome de
dispersión. Las condiciones ambientales afectaron la estructura del espectro de dispersión
en la comunidad de bosque seco neotropical estudiada.
El análisis de la importancia relativa del síndrome de dispersión y de la heterogeneidad
espacial en la formación de patrones espaciales de árboles adultos permitió determinar que
la heterogeneidad ambiental ejercía un efecto adicional (y en algunos el único) en la
formación de patrones agregados de la mayoría de especies estudiadas. Los resultados
señalaron diferencias en los patrones espaciales de las especies dependiendo del síndrome
de dispersión, pero también una gran variación en los patrones espaciales incluso entre
especies del mismo síndrome de dispersión.
El análisis simultáneo de los patrones de la lluvia de semillas y banco de semillas de una
comunidad de leñosas y su relación con la vegetación establecida indicaron que la lluvia de
semillas era temporalmente variable en número de especies y abundancia de semillas, y
3
dependía del síndrome de dispersión. El síndrome de dispersión también influyó en la
formación de bancos de semillas, siendo las especies con capacidad de dispersión limitada
(autócoras) las de mayor riqueza de especies y abundancia de semillas.
Los cérvidos también se consideraron como un elemento clave en el proceso de dispersión
de semillas. Al menos ocho especies leñosas fueron dispersadas legítimamente vía
endozoócora. La mayoría de las especies dispersadas presentaron diásporas sin
adaptaciones obvias para la dispersión, por lo que la ingestión de semillas por cérvidos se
constituye en una vía potencial para la dispersión de sus semillas a largas distancias y, con
ello, mejora la posibilidad de colonizar nuevos sitios y mantener el flujo genético.
Los resultados de este estudio aportan nuevas evidencias para el entendimiento de la
importancia de los procesos de dispersión de semillas en la estructura de los bosques secos
neotropicales. Uno de los principales hallazgos a partir de estos cuatro capítulos es que los
patrones espaciales de las especies, así como las estrategias que utilizan para dispersarse y
hacer frente a las condiciones adversas (es decir, lluvia o banco de semillas) llevan consigo
un efecto del síndrome de dispersión, y que la intensidad ese efecto depende a la vez de las
condiciones ambientales del lugar.
4
ABSTRACT
The importance of seed dispersal process in the estructuring and ecosystem dynamic is
widely recongnized. However, for seasonally tropical dry forest studies related to this
process are still scarce and scattered compared to tropical rain forests. The present research
deals with the importance of seed dispersal syndromes as a driver in the community
structure, focusing its attention to temporal and spatial patterns of seed dispersal in woody
communities of seasonally dry forest at Southwestern Ecuador. This area is part of the
Tumbesian region, one of the most important areas of endemism, but also one of the most
threatened areas around the world. Climate is characterized by a dry season from May to
November, and a rainy season from December to April. For the whole area an average
temperature between 20 ° and 26 ° C, and an average annual rainfall between 300 and 700
mm are estimated.
Fieldwork was carried out between February 2009 and September 2012. During a first step
information about native woody species of dry forests of southwestern Ecuador was
gathered, enabling to assign a dispersal syndrome to each species to determine the seed
dispersal spectrum. In a second step, available information from 109 established plots along
four municipalities in Loja province, which hold the highest and best conserved dry forest
remanants, was analyzed to establish the relationship between dispersal syndromes and
environmental conditions.
The relationships between dispersal syndromes and species spatial patterns; and between
dispersal syndromes and seed rain and seed bank patterns, were studied within a permanent
plot of 9 ha, in the Arenillas Ecological Reserve. Within this plot one transect of
5
approximately 3.4 km was set to collect monthly deer droppings, which were used to latter
analyze the rol of this group as seed dispersers.
The results showed that a large variety of plants in tropical dry forest of Southwestern
Ecuador require animal assistance to dispers their seeds. However, an analysis of seed
dispersal spectrum considering not only species richness, but also the relative abundance of
species, allowed to determine that despite the high variety of zoochorous species, most
individuals in the community corresponds to anemochoruos species. This shift may be due
to the relative abundance of habitats that are suitable for species with different dispersal
syndromes. Moreover, quantitative data analysis showed that environmental conditions
affect the structure of seed dispersal spectrum in the studied community.
The analysis of relative importance of dispersal syndrome, and the environmental
heterogeneity on formation of adult trees spatial patterns, indicated that environmental
heterogeneity exert an additional (or was the only) effect limiting the distribution of most
species in this forest. The findings showed differences in spatial patterns related to dispersal
syndrome, but also showed a large variation in spatial patterns even among species sharing
the same dispersal syndrome.
Simultaneous analysis of seed rain and seed bank patterns of a woody community, and their
relationship with established vegetation, suggested that seed rain is temporally variable in
species number and seeds abundance, and that variation is related to the dispersal
syndrome. Dispersal syndrome also influenced on the formation of seed banks, being
species with limited dispersal abilities (autochorous) the ones with highest species richness
and seed abundance.
6
Deer were found as a key element in the seed dispersal process. At least to eight woody
species were dispersed legitimately by ingestion. Diaspores of most dispersed species had
no obvious adaptations to seed dispersal, therefore, seed ingestion by deer represents a
potential pathway for long-distance dispersal, and hence, improves the chances to
colonizing new sites and to maintain gene flow.
Overall, these results provide new evidence for understanding the importance of seed
dispersal processes in the structure of Neotropical dry forests. One of the major findings
from these four chapters is that spatial patterns of species, and the strategies used to
disperse their seeds and to deal with the adverse conditions (i.e. seed rain or seed bank) are
related with dispersal syndromes, and the intensity of that relation depends in turn, on
environmental conditions.
7
INTRODUCCIÓN
El proceso de dispersión de semillas implica el movimiento de semillas lejos de la planta
madre, permitiendo a las plantas colonizar nuevos sitios, reducir la presión por competencia
y predación, y alcanzar sitios que favorezcan la regeneración (Howe y Smallwood 1982,
Wenny 2001, Herrera y Pellmyr 2002). Es así que desde hace algunas décadas la dispersión
de semillas ha sido considerada como un proceso clave en la estructuración y dinámica de
las comunidades vegetales (Howe y Smallwood 1982, Hubbell 1979, Gentry 1982, Willson
et al. 1990, Bullock et al. 1995). La concepción de la dispersión de semillas como la
plataforma sobre la cual se desarrollan los procesos subsecuentes de reclutamiento (Schupp
y Fuentes 1995, Nathan y Muller-Landau 2000) ha dado paso al estudio de este tema desde
varios enfoques, como la capacidad de las especies para colonizar nuevos sitios (Howe y
Smallwood 1982, Willson et al. 1990, Willson 1993, Schupp et al. 2002, Martínez-Garza et
al. 2011), adaptaciones a ciertas condiciones ambientales (Howe y Smallwood 1982,
Willson et al. 1990, Bullock et al. 1995) e interacciones planta-animal (Howe 1987,
Jordano 2001).
Las ventajas que una especie puede obtener de la dispersión de semillas están relacionadas
estrechamente con adaptaciones de las diásporas para aprovechar diferentes agentes
dispersores. La modificación de tejidos a alas para aprovechar el viento, el desarrollo de
estructuras comestibles sobre el fruto o la semilla para atraer animales, o estructuras
balísticas para la expulsión mecánica de las semillas, son algunas de las estrategias que
usan las plantas para dispersar sus semillas (Van der Pijl 1969, Howe y Smallwood 1982,
Cousens et al. 2008). El conjunto de estas características es conocido por los ecólogos
8
como “síndromes de dispersión” (van der Pijl 1969, Gottsberger y Silberbauer-Gottsberger
1983, Howe y Westley 1988, Cousens et al. 2008).
Para los bosques tropicales en particular, el conocimiento de los procesos de dispersión de
semillas, así como de otros procesos relacionados con el mantenimiento de la diversidad, se
ha desarrollado principalmente en torno a los bosques lluviosos, mientras que los bosques
secos han recibido poca atención (Bullock et al. 1995, Sánchez-Azofeifa et al. 2005, Miles
et al. 2006, Balvanera et al. 2011, Espinosa et al. 2012). Uno de los trabajos más conocidos
con respecto a los patrones de diversidad en bosques secos es el de Hubbell (1979), que ha
sido fundamental para esclarecer las diferencias en los patrones de distribución espacial de
las especies de bosques secos con respecto a los bosques lluviosos. La composición de
especies en los bosques secos y la estructura de las comunidades responden a condiciones
ambientales distintas de las que dominan en los bosques lluviosos. La marcada
estacionalidad en la disponibilidad de agua en los bosques secos es considerado uno de los
principales factores limitantes para los procesos de desarrollo y establecimiento de las
plantas, y por ende de su distribución dentro del ecosistema (Balvanera et al. 2011). La
topografía también está relacionada con la estructura de las comunidades de bosque seco
porque controla la disponibilidad de agua y nutrientes (Balvanera et al. 2011, Espinosa et
al. 2011, 2012). De esta manera, la variación de estos factores en el espacio genera un
mosaico de condiciones ambientales y disponibilidad de recursos (López-Martínez et al.
2013) que favorece el establecimiento de especies con características específicas. Por
ejemplo, en el contexto de dispersión de semillas, el espectro de síndromes de dispersión
puede cambiar entre comunidades de bosque seco en respuesta a la disponibilidad de agua
9
(López-Martínez et al. 2013); así, semillas dispersadas por el viento (anemocoria) tienden a
ser más frecuentes en árboles de los bosques más secos (Bullock et al. 1995).
El estudio de la dispersión de semillas como un control importante sobre la diversidad de
especies ha tomado fuerza en los últimos años (Levine y Murrell 2003, López-Martínez et
al. 2013). Trabajos como los de Hubbell (1979), Condit et al. (2000) y Seidler y Plotkin
(2006), plantean que los patrones de distribución no aleatorios están relacionados con la
eficiencia en la dispersión de semillas. Así, especies con síndromes de dispersión más
eficientes para la dispersión a largas distancias muestran patrones menos agregados que las
especies con síndromes menos eficientes. Si bien existen muchos otros factores
postdispersión que pueden restringir el reclutamiento de nuevos individuos (Nathan y
Muller-Landau 2000, Dalling et al. 2002), los efectos de los patrones de dispersión y
postdispersión han mostrado ser variables entre especies, e incluso dentro de una misma
especie a lo largo de gradientes ambientales (Nathan y Muller-Landau 2000).
Consecuentemente, la importancia de los patrones de dispersión de semillas en la estructura
de una comunidad particular no puede ser asumida a priori (Levine y Murell 2003).
En ecosistemas marcadamente estacionales, como los bosques secos, otro factor que
interviene en el éxito del evento de dispersión, así como en el desarrollo y sobrevivencia
del individuo en etapas posteriores, es el momento de dispersión de semillas (Khurana y
Singh 2001). Sin embargo, en muchas especies los requerimientos del individuo en la etapa
de dispersión difieren de los requerimientos para el desarrollo y establecimiento de la
plántula (Schupp y Fuentes 1995). Para sobrellevar estas diferencias una estrategia
alternativa luego de la dispersión es la formación de bancos de semillas en el suelo, el cual
10
puede atenuar las consecuencias de la dispersión de semillas en periodos desfavorables para
la germinación. Inclusive las desventajas de la alta densidad de semillas dispersadas cerca
de la planta madre pueden ser limitadas cuando hay un banco de semillas, debido a que los
eventos de germinación se extienden en el tiempo, reduciendo la competencia o predación
(Nathan y Muller-Landau 2000). En este sentido, tanto la dispersión a larga distancia como
la capacidad de formar bancos de semillas constituyen estrategias alternativas para evitar el
riesgo, una excesiva densidad y la competencia (Venable y Brown 1988, Bakker et al.
1996). Al ser estrategias alternativas se espera que la formación de bancos de semillas sea
limitada en especies con capacidad de dispersión a larga distancia, a diferencia de especies
con dispersión reducida, como las autócoras (Bakker et al. 1996), en las cuales la dispersión
en el tiempo puede compensar la pobre dispersión espacial (Willson 1993), mediante la
formación de bancos de semillas (Uhl et al. 1981, Putz y Appanah 1987, Alvarez-Buylla y
Martínez Ramos 1990, Moles y Drake 1999).
Para los bosques tropicales en general se señala que el síndrome de dispersión de mayor
importancia es posiblemente la zoocoria, y en particular la endozoocoria (dispersión por
ingestión). Las ventajas de la dispersión zoócora están relacionadas principalmente con la
capacidad de dispersión a largas distancias y con la dispersión dirigida a microhábitats
adecuados para el desarrollo de la plántula (Howe y Smallwood 1982). Incluso en los
bosques secos neotropicales, donde la disponibilidad de frutos está restringida en el tiempo,
se ha registrado en promedio un 46,2% (27 – 58,7%) de especies con dispersión
endozoocora (Jordando 2000). Sin embargo, y a pesar de que la información sobre animales
dispersores es limitada, estudios realizados en ecosistemas secos neotropicales mencionan
la influencia, tanto a nivel local como regional, que el comportamiento y abundancia de
11
vertebrados dispersores de semillas podría tener sobre la frecuencia de especies que
dispersan (Bullock et al. 1995, Griz y Machado 2001, Tabarelli et al. 2003).
Los grupos más estudiados desde el punto de vista de agentes dispersores de semillas han
sido las aves y los murciélagos (Nassar et al. 2014), mientras que la información disponible
acerca del rol de otros grupos animales en este proceso es dispersa. En el caso particular de
los cérvidos, los pocos estudios existentes en América se han centrado en Odocoileus
virginianus, y se han restringido a ecosistemas de Norteamerica (González-Espinosa y
Quintana-Ascencio 1986, Mandujano et al. 1997, Williams et al. 2008). En esos estudios se
ha registrado una gran variedad de semillas viables luego del paso por el tracto digestivo de
O. virginianus. En los bosques secos neotropicales los cérvidos conforman uno de los
grupos de mamíferos grandes y relativamente abundantes (Tirira 2007). Su rango de
movimiento, de varias hectáreas por día, implica que los cérvidos podrían jugar un papel
importante en el movimiento de semillas a largas distancias, y contribuir así al
mantenimiento del flujo genético, regeneración y colonización de nuevos sitios (Willson y
Traveset 2000, Traveset et al. 2007). A pesar de ello, han sido pocos los esfuerzos por
estudiar su legitimidad como dispersores de semillas en ecosistemas neotropicales (Bodmer
1991).
El presente estudio nace como una necesidad por incrementar el conocimiento de los
procesos que controlan el funcionamiento de los bosques secos de la región tumbensina y
aportar con información que permita la aplicación de medidas de manejo y conservación
adecuadas.
12
La importancia de incrementar el conocimiento de los procesos de dispersión de semillas y
su relación con el funcionamiento de los bosques secos neotropicales se debe a que estos
ecosistemas se puede plantear desde varios puntos de vista. Primero, por los elevados
niveles de endemismo que sostiene (Gentry 1995, Sánchez-Azofeifa et al. 2014),
principalmente en plantas leñosas (Portillo-Quintero y Sánchez-Azofeifa 2010) y aves (Best
y Kessler 1995), aunque para otros grupos se cuenta con poca información. Segundo, los
bosques secos tropicales ocupan el 42% de los bosques tropicales (Murphy y Lugo 1995,
Miles et al. 2006, Espinosa et al. 2012) y han sido históricamente más utilizados por el ser
humano que los bosques lluviosos (Bullock et al.1995). Tercero, a pesar de que en los
últimos años se ha incrementado el esfuerzo por estudiar los ecosistemas de bosque seco, el
área conservada sigue siendo poco perceptible, mientras que los problemas de
fragmentación y cambio de uso del suelo se intensifican (Best y Kessler 1995, Bullock et
al. 1995, Linares-Palomino et al. 2010, Espinosa et al. 2012).
13
OBJETIVOS:
Objetivo general:
La presente tesis tiene como objetivo general analizar la importancia de los síndromes de
dispersión de semillas en la estructuración de comunidades leñosas de bosques secos
tropicales.
Objetivos específicos:
1) Estudiar el espectro de dispersión de semillas y la relación del síndrome de dispersión
con la forma de crecimiento de la planta, abundancia de la especie y condiciones
ambientales.
A nivel de comunidad pueden presentarse diferentes síndromes de dispersión
influenciados por los atributos del ecosistema, circunstancias ambientales, estructura y
composición florística. El capítulo 1 “Seed dispersal spectrum of tree and shrub species
in Ecuadorian dry forests”, se centra en el análisis de la distribución de los síndromes
de dispersión en comunidades locales de vegetación leñosa para determinar si
síndromes de dispersión particulares están asociados con las formas de crecimiento de
las plantas, abundancia de las especies y condiciones ambientales. Además, se analiza
la fenología de la fructificación de cada síndrome de dispersión.
2) Examinar la importancia relativa del síndrome de dispersión y de la heterogeneidad
espacial sobre la realización del patrón espacial de árboles adultos en un bosque seco
tropical ecuatoriano.
14
El capítulo 2, “Does spatial heterogeneity blur the signature of dispersal syndromes on
spatial patterns of woody species? A test for a tropical dry forest”, contribuye al
entendimiento de los procesos funcionales y la coexistencia a través del análisis del rol
de la dispersión de semillas en la distribución espacial de especies a escalas pequeñas.
Específicamente se plantea conocer si la señal espacial esperada para especies con un
síndrome de dispersión particular es modificada por la heterogeneidad espacial y de qué
manera. Si bien estudios previos sugieren que los síndromes de dispersión dan lugar a
patrones específicos de distribución espacial (Hubbell 1979, Seidler y Plotkin 2006), en
este trabajo se propone que aquellos patrones podrían ser modificados por procesos
subsecuentes ligados a la heterogeneidad espacial.
3) Analizar en qué medida la lluvia de semillas y el banco de semillas están determinando
la composición de la vegetación leñosa.
En el capítulo 3, “Seed rain and seed bank contribution to regeneration of woody
species with different dispersal syndrome in an Ecuadorian tropical dry forest”, se
analiza si los patrones temporales en la dispersión de semillas y/o el síndrome de
dispersión controlan la relación entre la vegetación establecida y los compartimentos de
semillas (la lluvia de semillas o el banco de semillas) en un bosque seco del sur de
Ecuador. Lo que se espera es que en especies autocoras la composición de la vegetación
establecida muestre una mayor similitud con el banco de semillas, mientras que para
especies zoocoras y anemocoras se espera una mayor similitud entre la vegetación
establecida y la lluvia de semillas.
15
4) Analizar el rol que los cérvidos tienen en la regeneración de especies leñosas de un
bosque seco del suroccidente Ecuatoriano. Este objetivo se desarrolló en el capítulo 4,
titulado “Legitimidad de los cérvidos como dispersores de semillas de especies leñosas
en un bosque seco tropical”. El interés en este estudio se debe a que los resultados
obtenidos en los capítulos anteriores indican que algunas especies con características
típicas de plantas autocoras presentan patrones espaciales diferentes a los esperados
para este síndrome de dispersión. Estudios previos sugieren que los ungulados pueden
consumir frutos secos fibrosos, como algunos de los registrados en especies autocoras
del área de estudio.
16
METODOLOGÍA GENERAL Y ÁREA DE ESTUDIO
Área de estudio
El presente trabajo se llevó a cabo en los bosques estacionalmente secos del suroccidente de
Ecuador, ubicados en la región Pacífico Ecuatorial (Espinosa et al. 2012). Estos bosques
forman parte de la región biogeográfica Tumbesina, que abarca territorios del suroeste de
Ecuador y noroeste de Perú, en un rango altitudinal que va desde el nivel del mar hasta los
2000 m s.n.m., aproximadamente (Best y Kessler 1995, Espinosa et al. 2011). La región
Tumbesina abarca una gran variedad de climas, como resultado de la posición geográfica y
la topografía variada (Best y Kessler 1995). La precipitación es el factor climático más
variable, y por ende, el más importante para la definición de la vegetación a lo largo de la
región (Best y Kessler 1995). Esta región es reconocida como un centro de endemismo para
diferentes taxones (Best y Kessler 1995), así como por ser uno de los hotspots más
amenazados del mundo (Dinerstein et al. 1995, Espinosa et al. 2012).
Particularmente, los ecosistemas de bosque seco del suroccidente de Ecuador ocupan
territorios de tierras bajas, estribaciones occidentales bajas de la cordillera de los Andes y
los valles secos interandinos (Aguirre y Kvist 2005, Espinosa et al. 2012). El clima se
caracteriza por una estación seca que va de mayo a noviembre y una estación lluviosa que
se extiende desde diciembre a abril (Aguirre & Kvist 2005). Para toda esta zona se estima
una temperatura promedio anual entre 20° y 26°C y una precipitación promedio anual entre
300 y 700 mm (Aguirre y Kvist 2005).
17
Esta área comprende algunos de los remanentes de bosque seco tropical más grandes y
mejor conservados de la región Tumbesina, y una de las áreas de endemismo más
importantes del mundo (Best y Kessler 1995). Sin embargo, también representa una las
áreas más amenazadas del mundo debido a la presión antrópica, que ha reducido al bosque
seco a un mosaico de remanentes intercalados con grandes espacios dedicados a la
agricultura y ganadería (Best y Kessler 1995).
En el capítulo 1 del presente trabajo se abarcó la vegetación leñosa de toda el área de
bosques secos de la región suroccidente de Ecuador, que comprende las provincias de Loja
y El Oro, entre las latitudes 3°3’11’’ y 4°37’28’’ S, y entre las longitudes 79°14’37’’ y
80°25’46’’ O. Los capítulos 2, 3 y 4 se desarrollaron dentro de la Reserva Ecológica
Arenillas (REA), localizada en la provincia de El Oro (03° 34’ 15,44’’S; 80° 08’ 46,15’’E,
altitud 30 m). La REA fue incluida dentro del Patrimonio de Áreas Naturales del Estado
(PANE) en el año 2001 y ha estado protegida de actividades de extracción por
aproximadamente 60 años (BirdLife International 2014).
Metodología general
El presente trabajo inició con una recopilación de información sobre las especies leñosas
nativas de los bosques secos del suroccidente de Ecuador. Se obtuvo una lista de 203
especies a partir de los inventarios de Aguirre y Kvist (2005) y Aguirre et al. (2006a,
2006b). Las especies fueron clasificadas de acuerdo a la forma de crecimiento en árboles,
arbolitos y arbustos siguiendo la clasificación de Harling y Anderson (1977–2007),
Jørgensen y León-Yánez (1999) y Pennington et al. (2004). La revisión de especímenes de
herbario y la revisión de literatura permitió asignar un síndrome de dispersión a 160
18
especies de esa lista. Los síndromes de dispersión fueron asignados siguiendo la
clasificación de Van der Pijl (1969). Esta información se utilizó en el capítulo 1 para
determinar el espectro de síndromes de dispersión de semillas en los bosques secos del
suroccidente de Ecuador, y para determinar la relación del síndrome de dispersión con la
forma de crecimiento de la planta y con la fenología de la fructificación. Con un
subconjunto de especies se determinó el espectro de dispersión en base a la abundancia de
cada especie y se analizó la relación entre el síndrome de dispersión y condiciones
ambientales. Para esto se trabajó en 109 parcelas previamente establecidas en el sur del área
de estudio. Estas parcelas estuvieron establecidas en 48 parches de bosque seco con un
diseño estratificado, tratando de cubrir el rango total de condiciones físicas del área
(Espinosa et al. 2011).
Para cumplir con los objetivos planteados en los capítulos 2, 3 y 4, se estableció una parcela
permanente de 9 ha en la Reserva Ecológica Arenillas (REA). Esta parcela se ubicó en una
de las zonas mejor conservadas de la REA, que posee una formación vegetal de transición
entre bosque tropical seco y matorral seco de tierras bajas. Todas las plantas leñosas con
DAP > 5 cm han sido inventariadas y georreferenciadas desde el año 2009.
Adicionalmente, en la hectárea central de la parcela se registraron todos los individuos a
partir de 1cm de DAP de especies leñosas subarbustivas. El muestreo de la lluvia de
semillas y banco de semillas se realizó en las cinco hectáreas centrales de la parcela. En
ambos casos el muestreo se realizó de forma sistemática a partir de 265 puntos separados
14 m. Para la lluvia de semillas se utilizaron trampas de 0,64 m2, cubriendo un área total de
169,6 m2. Para el banco de semillas se tomaron muestras de 0.25 x 0.25 x 0.03 m,
cubriendo un área total de 16.56 m2.
19
BIBLIOGRAFÍA
Aguirre, Z., Kvist, L. 2005. Composición florística y estado de conservación de los bosques
secos del sur-occidente del Ecuador. Lyonia 8: 41-67.
Aguirre, Z., Kvist L., Sánchez O. 2006a. Bosques secos en Ecuador y su diversidad.
Botánica Económica de los Andes Centrales.
Aguirre Z., Linares-Palomino R., Kvist L. 2006b. Especies leñosas y formaciones vegetales
en los bosques estacionalmente secos de Ecuador y Perú. Arnaldoa 13: 324–350.
Álvarez-Buylla, E. R., Martínez-Ramos, M. 1990. Seed bank vs seed rain in the
regeneration of a tropical pioneer tree. Oecologia 84: 314-325.
Bakker, J. P., Poschlod, P., Strykstra, R. J., Bekker, R., Thompson, K. 1996. Seed banks
and seed dispersal: important topics in restoration ecology. Acta Botanica Neerlandica
45: 461-490.
Balvanera, P., Quijas, S. Pérez-Jiménez, A. 2011. Distribution patterns of tropical dry forest
trees along a mesoscale water availability gradient. Biotropica 43: 414–422.
Best, B. J., Kessler, M. 1995. Biodiversity and conservation in Tumbesian Ecuador and
Peru. BirdLife International, Cambridge.
BirdLife International. 2014. Important Bird Areas factsheet: Reserva Ecológica Arenillas.
Downloaded from http://www.birdlife.org on 07/10/2014
Bodmer, R. 1991. Strategies of seed dispersal and seed predation in amazonian ungulates.
Biotropica 23: 255-261.
20
Bullock, S. H., Mooney, H. A., Medina, E. 1995. Seasonally Dry Tropical Forests.
Cambridge University Press, Nueva York.
Condit, R., Ashton, P., Baker, P., Bunyavejchewin, S., Gunatilleke, S., Gunatilleke, N.,
Hubbell, S., Foster, R., Itoh, A., Lafrankie, J. V., Lee, H. S., Losos, E., Manokaran, N.,
Sukumar, R., Yamakura, T. 2000. Spatial patterns in the distribution of tropical tree
species. Science 288: 1414-1418.
Cousens, R., Dytham, C., Law, C. 2008. Dispersal in plants. A population perspective.
Oxford University Press, Nueva York.
Dalling, J. W., Muller-Landau, H. C., Wright, S. J., Hubbell, S. P. 2002. Role of dispersal
in the recruitment limitation of neotropical pioneer species. Journal of Ecology 90:
714–727.
Dinerstein, E., Olson, D. M., Gram, D. J., Webster, A. L., Primm, S. A., Brookbinder, M.
P., Ledec, G. 1995. Una evaluación del estado de conservación de las ecorregiones de
América Latina y Caribe. Banco Internacional de Reconstrucción y Fomento/Banco
Mundial, Washington, DC.
Espinosa, C. I., Cabrera, O., Luzuriaga, A., Escudero, A. 2011. What factors affect
diversity and species composition of endangered tumbesian dry forests in Southern
Ecuador? Biotropica 4: 15-22.
Espinosa, C. I., De La Cruz, M., Luzuriaga, A. L., Escudero, A. 2012. Bosques tropicales
secos de la región Pacífico Ecuatorial: diversidad, estructura, funcionamiento e
implicaciones para la conservación. Ecosistemas 21: 167-179.
21
Gentry, A. H. 1982. Patterns of Neotropical plant species diversity. Evolutionary Biology
15: 1-84.
Gentry, A. 1995. Diversity and floristic composition of neotropical dry forests. En:
Bullock, S. H., Mooney, H. A., Medina, E. (eds.). Seasonally Dry Tropical Forests. pp.
Pages 146-194. Cambridge University Press, Cambridge.
González-Espinosa, M., Quintana-Ascencio, P. 1986. Seed predation and dispersal in a
dominant desert plant: Opuntia, ants, birds, and mammals. En: Estrada, A., Fleming, T.
H. (eds.) Frugivores and seed dispersal. Junk Publishers, Dordrecht, pp. 273-284.
Gottsberger, G., Silberbauer–Gottsberger, I. 1983. Dispersal and distribution in the Cerrado
vegetation of Brazil. Sonderb. Naturwissenschaften 7: 325-352.
Griz, L. M. S., Machado, I. C. S. 2001. Fruiting phenology and seed dispersal syndromes in
Caatinga, a tropical dry forest in the northeast of Brazil. Journal of Tropical Ecology
17: 303-321.
Harling, G., Andersson, L. (eds). 1977 – 2007. Flora of Ecuador. University of Göteborg.
Sweden
Herrera, C. M., Pellmyr, O. (eds.) 2002. Plant–animal interactions. An evolutionary
approach. Blackwell Publishing. Oxford.
Howe, H. F. 1987. Seed dispersal by fruit-eating birds and mammals. En: Murray, D. R.
(ed.). Seed dispersal. Academic press, Sydney, pp. 123-189..
22
Howe, H. F., Smallwood., J. 1982. Ecology of seed dispersal. Annual Review of Ecology
and Systematics 13: 201-228.
Howe, H. F., Westley, L. C. 1988. Mechanics and ecology of mutualism. En: Howe, H. F.,
Westley, L. C. (eds.). Ecological relationships of plants and animals. pp. 107-
160.Oxford University Press, Oxford, pp. 107–160.
Howe, H. F., Miriti, M. N. 2000. No question: seed dispersal matters. Trends in Ecology
and Evolution 15: 434-436.
Hubbell, S. 1979. Tree dispersion, abundance, and diversity in a tropical dry forest. Science
203: 1299-1309.
Jordano, P. 2000. Fruits and frugivory. En: M. Fenner (ed.). Seeds: The ecology of
regeneration in plant communities (2ª ed.), CABI Publishing, Wallingford, pp. 125–
165.
Jordano, P. 2001. Conectando la ecología de la reproducción con el reclutamiento
poblacional de plantas leñosas Mediterráneas. En: Zamora, R., Pugnaire, F. (eds.).
Ecosistemas Mediterráneos. Análisis funcional. CSIC, Madrid, pp. 183-211.
Jørgensen, P. M., León-Yánez, S. (eds.). 1999. Catalogue of the vascular plants of Ecuador.
Monographs in Systematic Botanty of the Missouri Botanical Garden 75: 1–1182.
Khurana, E., Singh, J. S. 2001. Ecology of seed and seedling growth for conservation and
restoration of tropical dry forest: a review. Environmental Conservation 28: 39–52.
23
Levine, J., Murrell, D. 2003. The community-level consequences of seed dispersal patterns.
Annual Review of Ecology, Evolution and Systematics 34: 549-574.
Linares-Palomino, R., Kvist, L. P., Aguirre-Mendoza, Z., Gonzales-Inca, C. 2010.
Diversity and endemism of woody plant species in the Equatorial Pacific seasonally
dry forests. Biodiversity and Conservation 19: 169–185.
López-Martínez, J. O., Sanaphre-Villanueva, L., Dupuy, J. M., Hernández-Stefanoni, J. L.,
Meave, J. A., Gallardo-Cruz, J. A. 2013. β-Diversity of functional groups of woody
plants in a tropical dry forest in Yucatan. PLoS ONE 8(9): e73660.
Mandujano, M. C., Golubov, J., Montaña, C. 1997. Dormancy and endozoochorous
dispersal of Opuntia rastrera seeds in the southern Chihuahuan Desert. Journal of Arid
Environments 36: 259–266.
Martínez-Garza, C., Osorio-Beristain, M., Valenzuela-Galván, D., Nicolás-Medina, A.
2011. Intra and inter-annual variation in seed rain in a secondary dry tropical forest
excluded from chronic disturbance. Forest Ecology and Management 262: 2207-2218.
Miles, L., Newton, A., Defries, R., Ravilious, C., May, I., Blyth, S., Kapos, V., Gordon, J.
2006. A global overview of the conservation status of tropical dry forests. Journal of
Biogeography 33: 491–505.
Moles, A. T., Drake, D. R. 1999. Potential contributions of the seed rain and seed bank to
regeneration of native forest under plantation pine in New Zealand. New Zealand
Journal of Botany 37: 83-93.
24
Murphy, P. G., Lugo, A. E. 1986. Ecology of Tropical Dry Forest. Annual Review of
Ecology, Evolution and Systematics 17:67-88.
Murphy, P. G., Lugo. A. E. 1995. Dry forests of Central America and the Caribbean. En:
Bullock, S. H., Mooney, H. A., Medina, E. (eds.). Seasonally dry tropical forests.
Cambridge University Press, Cambridge, pp. 9-34.
Nassar, J., Stoner, K. E., Ávila-Cabadilla, L., Espírito-Santo, M. M., Aranguren, C. I.,
González-Carcacía, J. A., Lobato-García, J. M., Olívio-Leite, L., Álvarez-Añorve, M.,
Brandão, H. N. M., Dolabela-Falcão, L. A., Rodríguez, J. P. 2014. Fruit-eating bats and
birds of three seasonal tropical dry forests in the Americas. En: Sánchez-Azofeifa, A.,
Powers, J. S., Fernandes, G. W., Quesada, M. (eds.) Tropical dry forests in the
Americas: ecology, conservation, and management. CRC Press, Boca Ratón, FL, pp.
173-220.
Nathan, R., Muller-Landau, H. C. 2000. Spatial patterns of seed dispersal, their
determinants and consequences for recruitment. Trends in Ecology and Evolution 15:
278 – 285.
Pennington, T. D., Reynel, C., Daza, A. 2004. Illustrated guide to the trees of Peru. David
Hunt, Sherborne.
Ragusa-Netto, J., Silva, R. R. 2007. Canopy phenology of a dry forest in western Brazil
Brazilian Journal of Biology, 67: 569-575.
Portillo-Quintero, C. A., Sánchez-Azofeifa, G. A. 2010. Extent and conservation of tropical
dry forests in the Americas. Biological Conservation 143: 144–155.
25
Putz, F. E., Appanah, S. 1987. Buried seeds, newly dispersed seeds, and the dynamics of a
lowland forest in Malaysia. Biotropica 28: 192-209.
Sánchez-Azofeifa, G. A., Quesada, M., Rodriguez, J. P., Nassar, J. M., Stoner, K. E.,
Castillo, A., Garvin, T., Zent, E. L., Calvo-Alvarado, J. C., Kalacska, M. E. R.,
Fajardo, J. L., Gamon, J. A., Cuevas-Reyes, P. 2005. Research priorities for
Neotropical dry forests. Biotropica 37:477-485.
Sanchez-Azofeifa, A., Powers, J. S., Fernandes, G. W. Quesada, M. (eds.). 2014. Tropical
dry forests in the Americas: Ecology, conservation, and management. CRC Press,
Boca Ratón, FL.
Schupp, E., Fuentes, M. 1995. Spatial patterns of seed dispersal and the unification of plant
population ecology. Écoscience 2: 267-275.
Schupp, E., Milleron, T., Russo, S. 2002. Dissemination limitation and the origin and
maintenance of species-rich tropical forests. En: Levey, D. J., Silva, W. R., Galetti, M.
(eds.) Seed dispersal and frugivory: ecology, evolution and conservation. CAB
International, Wallingford, pp. 19-33.
Seidler, T., Plotkin, J. 2006. Seed dispersal and spatial pattern in tropical trees. PLoS
Biology 4: 2132-2137.
Tabarelli, M., Vicente, A., Barbosa, D. C. A. 2003. Variation of seed dispersal spectrum of
woody plants across a rainfall gradient in north-eastern Brazil. Journal of Arid
Environments 53: 197–210.
26
Tirira, D. 2007. Guía de campo de mamíferos del Ecuador. Ed. Murciélago Blanco. Quito.
Traveset, A., Robertson, A. W., Rodríguez-Pérez, J. 2007. A Review on the Role of
Endozoochory in Seed Germination. En: Dennis E.W. Schupp, Green, R.J. and
Westcott, D.A. (Eds.). Seed dispersal: Theory and its application in a changing world,
CAB International, Oxford, pp. 78-103.
Uhl, C., Clark, K., Clark, H., Murphy, P. 1981. Early plant succession after cutting and
burning in the upper Rio Negro region in the Amazon Basin. Journal of Ecology 69:
631-649.
Van der Pijl, L. 1969. Principles of dispersal in higher plants. Springer-Verlag Berlín.
Venable, D. L., Brown, J. S. 1988. The selective interactions of dispersal, dormancy, and
seed size as adaptations for reducing risk in variable environments. American
Naturalist 131: 360-384.
Wenny, D. G., Levey, D. J. 1998. Directed seed dispersal by bellbirds in a tropical cloud
forest. Proceedings of the National Academy of Sciences USA 95: 6204–6207.
Wenny, D. 2001. Advantages of seed dispersal: a re-evaluation of directed dispersal.
Evolutionary Ecology Research 3: 51–74.
Williams, S. C., Ward, J. S., Ramakrishnan, U. 2008. Endozoochory by white-tailed deer
(Odocoileus virginianus) across a suburban/woodland interface. Forest Ecology and
Management 255: 940–947.
27
Willson, M. F. 1993. Dispersal mode, seed shadows, and colonization patterns. Vegetatio
107-108: 261-280.
Willson, M. F., Rice, B.L., Westoby, M. 1990. Seed dispersal spectra: a comparison of
temperate plant communities. Journal of Vegetation Science 1: 547-562.
Willson, M. F., Traveset, A. 2000. The Ecology of Seed Dispersal. En: Fenner, M. (ed.).
Seeds: The ecology of regeneration in plant communities. 2ª ed. CABI Publishing.
Wallingford, pp. 85-110.
28
CAPÍTULO 1
SEED DISPERSAL SPECTRUM OF WOODY SPECIES IN SOUTH ECUADORIAN
DRY FORESTS: ENVIRONMENTAL CORRELATES AND THE EFFECT OF
CONSIDERING SPECIES ABUNDANCE.
Andrea Jara-Guerrero1, Marcelino De la Cruz
2, Marcos Méndez
3
1Instituto de Ecología, Universidad Técnica Particular de Loja, CP.: 11-01-608, Loja, Ecuador
2Departamento de Biología Vegetal, E.U.I.T. Universidad Politécnica de Madrid, Spain
3Área de Biodiversidad y Conservación, Universidad Rey Juan Carlos, Spain
Manuscrito publicado en:
BIOTROPICA 43(6): 722–730
29
ABSTRACT
This study examines the seed dispersal spectrum of the tropical dry forests of Southern
Ecuador, in an effort to contribute to the knowledge of the complex dynamics of tropical
dry forests. Seed dispersal spectrum was described for a total number of 160 species.
Relationships of dispersal syndromes with plant growth form and climatic seasonality were
explored. For a subset of 97 species, we determined whether dispersal spectrum changes
when species abundance, in addition to species number, is taken into account. The same
subset was used to relate dispersal syndromes with the environmental conditions.
Zoochorous species dominated in the studied community. However, when considering the
individual abundance of each species, anemochory was the prevalent dispersal syndrome.
We found a significant difference in the frequency of dispersal syndromes among plant
growth forms, with epizoochory only occurring in shrub species. The dispersal spectrum
was dependent on climatic seasonality. The largest proportion of anemochorous species
fructified during the dry season, while zoochorous diaspores dominated during the rainy
season. A fourth-corner analysis indicated that the seed dispersal spectrum of southern
Ecuador dry forests is controlled by environmental conditions such as annual precipitation,
annual temperature range or topography. Our results suggest that spatio-temporal changes
in the environmental conditions may affect important ecological processes for dispersal.
Thus, the predominance of one syndrome or another may depend on the spatial variation of
environmental conditions.
Key words: dispersal syndromes; growth forms; Neotropical dry forest; fruiting phenology.
30
RESUMEN
Este estudio examina el espectro de dispersión de semillas de los bosques secos del Sur de
Ecuador, como un aporte al conocimiento de la dinámica de los bosques secos tropicales.
Con información sobre 160 especies leñosas se describió el espectro de dispersión y se
exploró su relación con el hábito de las plantas y la estacionalidad climática. Con un
subconjunto de 97 especies se analizó si el espectro de dispersión cambia cuando es
estimado en base a abundancia de individuos en lugar de especies. Se usó ese mismo
subconjunto para relacionar los síndromes de dispersión con las condiciones ambientales.
Las especies zoocoras prevalecieron dentro de la comunidad estudiada. Sin embargo,
cuando se consideró la abundancia de individuos por especie, el síndrome dominante fue la
anemocoria. Se encontró una diferencia significativa en la frecuencia de síndromes de
dispersión entre los hábitos de las plantas, con una presencia de la epizoocoria únicamente
en especies arbustivas. El espectro de dispersión estuvo influenciado por la estacionalidad
climática. La mayor proporción de especies anemócoras fructificó durante la estación seca,
mientras que en la estación lluviosa dominó la zoocoria. Además, un análisis de la cuarta
esquina indicó que el espectro de dispersión está controlado por condiciones ambientales
como precipitación anual, rango de temperatura anual o topografía. Nuestros resultados
sugieren que los cambios espacio–temporales del ambiente pueden estar relacionados con
factores importantes para la dispersión. Así, la predominancia de un síndrome de dispersión
podría depender de la heterogeneidad espacial en las condiciones ambientales.
31
INTRODUCTION
Seed dispersal plays a fundamental role in the colonization of new habitats, population
dynamics as well as in species interactions, community structure and diversity (Van der Pijl
1969, Howe & Smallwood 1982, Willson et al. 1990, Hughes et al. 1994, Morales & Carlo
2006). Because of this, it is a key factor in conservation biology and restoration
management (Strykstra et al. 2002, Navarro et al. 2009).
The diaspores of many plant species have developed specialized morphological structures
which enable them to make use of biotic and abiotic dispersal vectors (Van der Pijl 1969,
Wenny 2001, Schulze et al. 2002). These morphological structures, as well as diaspore
color, chemical characteristics of the pulp and phenological events of plants associated with
a particular dispersal mode, are known as "dispersal syndromes" (Van der Pijl 1969,
Gottsberger & Silberbauer-Gottsberger 1983, Howe & Westley 1988). The proportion of
dispersal syndromes in a particular vegetation type is known as the “dispersal spectrum”
(Howe & Smallwood 1982, Hughes et al. 1994, Arbeláez & Parrado-Rosselli 2005). The
study of dispersal spectra has interested researchers for several decades (Van der Pijl 1969,
Frankie et al. 1974, Gentry 1982, Howe & Smallwood 1982, Howe & Westley 1988). It is
seen as a way of documenting fundamental aspects of the functional diversity in
ecosystems.
The dispersal spectrum of a plant community can be influenced by several factors. First,
dispersal syndromes are related to plant growth form. For example, zoochory and autochory
are widely distributed among shrubs and treelets of the understory, whereas anemochory is
more frequent among canopy trees (Wikander 1984, Justiniano & Fredericksen 2000, Griz
32
& Machado 2001, Butler et al. 2007). In general, this pattern is explained by the reduction
of wind speed in the forest understory (Wikander 1984, Justiniano & Fredericksen 2000,
Griz & Machado 2001). However, in some tropical dry forests there is a great proportion of
anemochory among shrubs, favored possibly by plant deciduousness, which allows a higher
wind circulation in the forest understory (Griz & Machado 2001).
Second, physical conditions influence the dispersal spectrum. There is a broad consensus
that anemochorous plants are relatively common in dry habitats, while those adapted for
zoochory dominate in wet habitats (Howe & Smallwood 1982, Jordano 2000, Herrera &
Pellmyr 2002). Quantitative correlations between physical conditions and dispersal
spectrum have, however, rarely been established (see, however, Butler et al. 2007). Some
studies have found that certain environmental conditions, such as precipitation,
temperature, soil nutrient status or canopy closure influence the relative presence of a
dispersal syndrome in a particular site (Willson et al. 1990, Bullock 1995). For example, an
association of zoochory and those sites with greater moisture and soil fertility has been
suggested (Gentry 1982, Willson et al. 1990). At a local level, this pattern could be related
with topography (Wikander 1984, Bullock 1995).
Third, for tropical dry forests, several authors report a seasonal pattern in dispersal spectra.
Anemochorous species fruited during the dry season, while zoochorous species fruited
throughout the rainy season (Frankie et al. 1974, Gottsberger & Silberbauer-Gottsberger
1983, Ibarra-Manríquez et al. 1991, Oliveira & Moreira 1992, Bullock 1995, Justiniano &
Fredericksen 2000, Batalha & Mantovani 2000, Griz & Machado 2001).
33
Variation in the dispersal spectrum in relation to spatio-temporal variation of environmental
conditions could indicate important ecological factors for dispersal. For example, dry
seasons lead to leaf shedding which favours anemochory (Wikander 1984, Ibarra-
Manríquez et al. 1991, Griz & Machado 2001), while zoochorous fleshy fruits require
water and suitable temperature to mature which occur in rainy seasons or wetter sites
(Herrera & Pellmyr 2002). Nevertheless, these relationships between dispersal syndromes
and physical conditions are empirically based and causality is difficult to disentangle. For
instance, Hughes et al. (1994) and Butler et al. (2007) argue that a relationship between
physical conditions and the frequency of dispersal syndromes has lesser importance, and
rather results from the indirect influence on plant growth form and seed size. Spatio-
temporal changes in dispersal syndromes can also be causally related to physical conditions
or be a by-product of species turnover. Statistical tools such as the fourth-corner analysis
currently allow relating dispersal syndromes and environmental conditions with a robust
and simple statistical procedure, and permit disentangling some of the effects usually
confounded in traditional regression approaches. Such techniques can shed light on the
issues raised above.
As in other analyses of functional diversity, a potential limitation of current generalizations
about dispersal spectra is that they usually ignore the relative abundance of each species
(Wikander 1984, Justiniano & Fredericksen 2000, Batalha & Mantovani 2000, Griz &
Machado 2001, Ragusa-Netto & Silva 2007). For example, in a semideciduous cerrado of
Brazil zoochory was the dominant dispersal syndrome at the species level, but the
anemochorous species had a higher number of individuals (Gottsberger & Silberbauer-
Gottsberger 1983). Such mismatches suggest that different causal factors could determine
34
species richness and the success of species with certain syndromes, and consequently,
dominance and species richness in plant communities.
In general, seed dispersal in tropical dry forests has received little attention compared to
rain forests. Although some studies exist on dispersal spectra for Neotropical dry forests,
data are very superficial and scattered, which is worrying considering that they hold several
of the most endangered ecosystems in the world (Best & Kessler 1995, Vázquez et al.
2001, Espinosa et al. 2010). In order to enhance our knowledge about the functional
diversity of these forests, the present study examined the seed dispersal spectrum of woody
species in tropical dry forests of Southern Ecuador. We answered the following questions:
(1) Do dispersal syndromes vary among plant growth forms (tree, treelet and shrub)? (2)
Does the relative frequency of dispersal syndromes differ between the rainy and dry
seasons? (3) Does the dominance of dispersal syndromes change if species abundance, in
addition to species number, is taken into account? (4) Are dispersal syndromes related to
environmental conditions? We also summarize the existing data on dispersal spectra for
other Neotropical dry forests, and compare them with the results obtained in this study, in
an effort to improve the generality of the knowledge of dispersal in tropical dry forests.
METHODS
STUDY AREA.— The studied dry forests are located in Loja and El Oro Provinces, between
latitudes 3°3’11” and 4°37’28” S and between longitudes 79°14’37” and 80°25’46” O
(Fig.S1). This area comprises some of the largest and best preserved remnants of tropical
dry forest of the Tumbesian biogeographic region, one of the most important areas of
endemism in the world, but at the same time, one of the most threatened (Best & Kessler
35
1995). Nowadays, anthropogenic pressure has led to this area being dominated by
agricultural patches interspersed with forest remnants. The climate is characterized by a dry
season extending from May to November and a rainy season extending from December to
April (Aguirre & Kvist 2005). Mean annual temperature is between 20-26° C, and the
annual precipitation is between 300-700 mm (Aguirre & Kvist 2005). Elevation ranges
between 50 and 800 m asl (Vázquez et al. 2001).For the studied area, Aguirre et al. (2006a)
found 238 of the 275 woody species recorded for dry vegetation types of Ecuador.
DATA SAMPLING.— We obtained a list of 203 woody native species of southern Ecuador
from the inventories of Aguirre and Kvist (2005) and Aguirre et al. (2006a, 2006b). The
species were classified according to growth form into trees, treelets or shrubs, following
Harling and Anderson (1977-2007), Jørgensen and León-Yánez (1999) and Pennington et
al. (2004) (Table 1).
A literature revision (see Appendix 1) as well as field observations and examination of
herbarium specimens (Herbario de la Universidad Técnica Particular de Loja, Herbario
Loja and Herbario Nacional del Ecuador), allowed us to assign a dispersal syndrome to 160
species of this list (which represent 121 genera and 50 families). Dispersal syndromes were
assigned to each species based on diaspore traits such as fruit type, morphology and color
of the diaspore and, for diaspores with fleshy structures, we also registered the existence or
not of protective structures that might hinder consumption by animals (Janson 1983, Link
& Stevenson 2004) (Table 1).
The species were classified into four general dispersal syndromes (Van der Pijl 1969):
autochory, anemochory, zoochory and polychory. Autochory was further divided into
36
active or passive. Active autochory includes diaspores propelled explosively, while passive
autochory refers to the so-called passive ballists, triggered by passing animals, wind or
raindrops. So-called barochorous (diaspores simply released and falling to the ground) were
also included in the latter group (Gottsberger & Silberbauer-Gottsberger 1983).
TABLE 1. Growth forms and diaspore traits considered in this study.
Parameter Category
Growth form Tree, treelet, or shrub (Harling and Anderson 1977-2007, Jørgensen and León-
Yánez 1999, Pennington et al. 2004).
Fruit type Dry indehiscent: achene, achenetum, capsule, legume or mericarp; dry dehiscent:
capsule, follicle, legume or craspedium; fleshy: berry, baccarium, drupe,
pseudodrupe, syconium or sorosis (Spjut 1994).
Diaspore type Fruit or seed.
Diaspore
morphology
Sticky anthocarp, hooked spines, arists, stiff hairs, plumed, samara, woolly,
pappus, caruncle, sarcotesta, aril, mucilage, fleshy periant, fleshy pulp, explosive
dehiscence or without structures (van der Pijl 1969).
Diaspore
color
Green, brown, yellow, orange, red, white, black, blue/purple or mixed (Janson
1983, Wheelwright & Janson 1985, Link & Stevenson 2004, Chen et al. 2004).
Diaspore
protectiona
Protected (mature pulp covered by a husk, distinct hard, non-nutritious layer as a
barrier to feeding or digestion) or unprotected (fruits with a soft, flexible skin,
with a thickness less than 10% of the thickness of the smallest fruit dimension)
(Janson 1983, Link & Stevenson 2004).
aThe diaspore protection was determined only for endozoochorus and sinzoochorus species.
37
Following Augspurger (1986), anemochorous diaspores were further divided into six
morphological groups, according to their aerodynamic structures: undulator, floater,
autogyro, rolling-autogyro, helicopter or tumbler.
TABLE 2 Variables describing environmental conditions in 109 plots of dry forest of Macará and
Zapotillo (Loja province) (for details, see Espinosa et al.2010).
Environmental variables Scale Range of values
Statistic used in the
fourth-corner analysis
Inclination Continuous 0 - 45° Pseudo F
Annual temperature range Continuous 13.80 - 16.60°C Pseudo F
Annual Precipitation Continuous 270 - 1284 mm Pseudo F
Rainfall in the driest month Continuous 0 - 4 mm Pseudo F
Soil moisture Continuous 1.05 - 28.7% Pseudo F
Soil organic matter Continuous 0.03 - 13.02 % Pseudo F
Topography Categorical Valley, slope or ridge Pearson 2
Intervention grade Ordinal 1: low; 2: medium; 3: high Pseudo F
Zoochorous species were divided into four subcategories (Van der Pijl 1969): (a)
epizoochory, defined as the passive transport through the adhesion of diaspores to feathers
or hair of animals; (b) sinzoochory, assigned to diaspores collected and stored in caches by
rodents, as well as diaspores actively transported by animals which feed on parts of them
38
but do not ingest the seeds; (c) endozoochory, when the diaspore is actively ingested and
seeds are usually evacuated intact; and (d) myrmecochory, assigned to diaspores with
elaiosomes or those actively transported by ants.
Some species have seeds that are polymorphic for dispersal structures, or have two
dispersal modes which could operate sequentially (Willson et al. 1990, Griz & Machado
2001). Diaspores of this type were assigned to a separate category called “polychory” (Van
der Pijl 1969, Cousens et al. 2008).
Furthermore, we assigned each species to one of three fruit types: fleshy, dry-dehiscent and
dry-indehiscent, as the proportion of species in these simple categories has been suggested
to correlate with environmental variables (Knight 1986, Willson et al. 1989, Tabarelli et al.
2003). To determine the relationship among dispersal syndromes and climatic seasonality,
we collected data about fruiting phenology both from field observations and from labels of
herbarium specimen collected in dry forest localities.
In addition, we assigned dispersal syndromes to all (97) woody species recorded in 109
plots of dry forest located in the south of our study area (Fig.S1). These plots were
established on 48 forest stands with a stratified sampling design, trying to encompass the
whole range of physical conditions present in the tropical dry forests of southern Ecuador
(Espinosa et al. 2010). Two or three plots of 20 x 20 m were sampled per forest stand. The
total area sampled was 5.45 ha (for further details, see Espinosa et al. 2010). In each plot,
eight variables describing environmental conditions were also recorded (Table 2). These
variables were selected based on its documented relationship with the existence of some
dispersal syndromes (Knight 1986, Willson et al. 1990, Bullock 1995, Tabarelli et al.
39
2003), and on its relevance to other ecological patterns, such as richness and species
composition in different plant communities (Espinosa et al. 2010). These data were used to
analyze the relationship among dispersal traits and environmental conditions. In each plot,
we also recorded the number of individuals of each species in order to calculate the seed
dispersal spectrum based on individual abundance, i.e. as the percentage of individuals
having each dispersal syndrome over the total number of individuals recorded in all the
plots.
STATISTICAL ANALYSES.— Differences in the frequency of major dispersal syndromes
among categories of growth form (tree, treelet and shrub) and among climatic seasons
(rainy and dry) were analyzed by means of contingency tables (G-test). We used also G-
tests to analyze the frequency of major dispersal syndromes in the 109 plots of dry forest,
taking into account the number of individuals of each species.
We used the set of floristic abundances and the environmental data from the 109 dry forest
plots to analyze the relationship between two dispersal traits (dispersal syndromes and type
of fruit) and environmental conditions. For this, we used the new version of the fourth-
corner analysis (Legendre et al. 1997, Dray & Legendre 2008). Fourth-corner analysis
directly relates an R matrix of environmental variables (Table 2) to a Q matrix of species
traits (dispersal syndromes and fruit types), by means of an L matrix of species abundance
measured in the field (Dray & Legendre 2008, Aubin et al. 2009). A statistic is computed
for each pair of species traits and environmental variables. Depending on the type of
variable, this statistic is a Pearson 2 (for two qualitative variables); or a Pseudo-F and a
correlation ratio η2 (for one quantitative and other qualitative variable) (Legendre et al.
40
1997, Dray & Legendre 2008). Moreover, SRLQ, a global multivariate statistic that links the
complete matrices R and Q is computed as the sum of all 2 and η
2 values in the fourth-
corner matrix. For computing this statistic, quantitative variables are standardized to mean
0 and variance 1 and qualitative variables are coded using dummy variables (Dray &
Legendre 2008).
The significance of all fourth-corner statistics can be tested using different permutation
models (Dray and Legendre 2008). In this study we used a model where cell values in the L
matrix are permuted within each column. This model tests the null hypothesis that the
species are randomly distributed with respect to abiotic environmental conditions (Aubin et
al. 2009). This analysis was carried out using the ade4 package (Dray et al. 2007) in the R
software (R environment Core Team 2010).
RESULTS
DESCRIPTION OF DISPERSAL SYNDROMES.— Our description of dispersal syndromes is based
on the total data set (n = 160). Zoochory was the dominant dispersal syndrome in the
studied community (54%, 87 species) followed by anemochory (28%, 44 species) and
autochory (15%, 24 species). The remaining species were assigned to the polychory
category.
Fleshy fruits represented the 44% percent (70 species) of all the species, followed by
dehiscent fruits (33%, 53 species) and indehiscent fruits (23%, 37 species). Fruits were the
dominant dispersal units (69.4%, 111 species), while seeds with specialized dispersal
structures occurred in 30.6% of all species, mainly the anemochorous ones.
41
Among zoochorous species, 88% (76 species) were endozoochorous, while epizoochorous,
sinzoochorous and myrmecochorous species were found in very low percentages (5%, 5%
and 2%, respectively). Fruits with fleshy pulp were the most common between
endozoochorous species, especially drupes (42%) and berries (30%).
We found nine different colors among the diaspores of 80 endozoochorous and
sinzoochorous species. Brightly colored diaspores were clearly dominant. Only 9% of these
species presented some physical protection against feeding (Table 1). This attribute was
most evident in dull colored diaspores (brown, yellow and orange).
Most anemochorous species (52.3%) presented dry dehiscent fruits, and used seed
structures for dispersal. A second group (47.7%) presented fruits with structures modified
for dispersal, which reduce rates of descent. The six morphological groups of
anemochorous diaspores were present. The rolling-autogyro structures were prevalent
(31.8%, 14 species), followed by those with helicopter structures (18.2%, 8 species),
autogyro and undulator structures (both 15.9%, 7 species) and floater and tumbler
structures (both 6.8%, 3 species).
Among autochorous species, passive autochory was dominant and only seven species
(29.2%) had an active dispersal favored by fruits with explosive dehiscence.
We registered five polychorous species. Three of these species (Cnidoscolus aconitifolius,
Croton menthodorus and Croton wagneri) presented fruits with explosive dehiscence,
which is characteristic to active autochorous diaspores, and seeds with elaiosome,
characteristic of myrmecochorous diaspores. The other two species (Erythrina spp.)
exhibited characteristics of both autochory and zoochory.
42
GROWTH FORM AND DISPERSAL SYNDROMES.— The G-test revealed significant differences
in the frequency of dispersal syndromes among growth forms (G2 = 24.526, P = 0.006).
Zoochory was the dominant dispersal syndrome in the three studied growth forms, mainly
represented by endozoochorous species (35.6% trees, 18.9% treelet and 30% shrubs).
TABLE 3 Number of species showing different dispersal syndromes and growth forms among 160
woody species in the tropical dry forests of southern Ecuador.
Dispersal syndromea
Growth form
Tree Treelet Shrub
Zoochory 36 19 35
Endozoochory 32 17 27
Myrmecochory 1 1 3
Sinzoochory 3 1 0
Epizoochory 0 0 5
Anemochory 28 3 13
Autogyro 5 1 1
rolling autogyro 6 1 9
Floater 7 0 0
Helicopter 5 1 2
Undulator 3 0 0
Tumbler 2 0 1
Autochory 13 3 13
Active 3 0 7
Pasive 10 3 6
a Polychorous species have been separated in the corresponding syndromes, contributing one case to
each dispersal syndrome implied.
43
Endozoochorous, sinzoochorous and myrmecochorous species were also present in the
three categories, while epizoochory occurred only in shrub species (Table 3).
Trees presented the greatest variety of anemochorous diaspores (Table 3), with a
dominance of floater diaspores (25%), followed by rolling-autogyro diaspores (21.4%),
autogyro (17.9%) and helicopter (17.9%). By contrast, rolling-autogyro diaspores were
dominant among treelet and shrub species. Floater and undulator diaspores were found only
among tree species (Table 3).
Autochorous diaspores were registered in the three growth forms. The greatest frequency of
active autochory was found among shrub species corresponding to Euphorbiaceae,
Fabaceae and Mimosaceae families.
CLIMATIC SEASONALITY AND DISPERSAL SYNDROMES.— We collected phenological
information for 54 species. In general, fruiting phenology differed among dispersal
syndromes (G2 = 12.626, P = 0.002). The largest proportion of anemochorous species
fructified during the dry season, while zoochorous diaspores fructified during the rainy
season (Fig.1). Autochorous species also had a marked peak of fruiting during this season
(Fig.1). No polychorous species were recorded in this group.
44
De
c
Ja
n
Fe
b
Ma
r
Ap
r
Ma
y
Ju
n
Ju
l
Au
g
Se
p
Oct
No
vMonths
Nu
mb
er
of sp
ecie
s fru
itin
g
02
46
81
01
21
4Anemochory
Zoochory
Autochory
05
01
00
15
02
00
25
0
Ra
infa
ll (
mm
)
FIGURE 1. Monthly variation in the number of fruiting species belonging to the main dispersal
syndromes in the tropical dry forests of southern Ecuador (n = 54 species). Average monthly
rainfall is represented with a continuous line. Rainfall data was obtained from the Worldclim 1.4
database (Hijmans et al. 2005).
INDIVIDUAL ABUNDANCE AND DISPERSAL SYNDROMES.— A total of 3392 individuals
(corresponding to 97 species) were recorded in the 109 forest plots. Another 132
individuals could not be identified to species and therefore were not included in the
analysis. Zoochorous species were prevalent among the 97 species recorded (55%, 54
species) followed by anemochorous (28%, 28 species), autochorous (15%, 15 species) and
polychorous species (1%, 1 species). However, when the frequency of dispersal syndromes
was analyzed based on individual abundance, the dispersal spectrum changed (G2= 20.512,
P = 0.0001). The greatest proportion of individuals corresponded to anemochorous species
(51.56%, 1749 individuals), while zoochory represented only the 34.6% (1174 individuals)
followed by autochory (11.5%, 391 individuals) and polychory (2.3% 78 individuals).
45
In terms of individual abundance, endozoochory remained as the dominant strategy within
zoochory (29.4% of total individuals), while sinzoochory and epizoochory were
underrepresented (3.3 % and 1.9%, respectively).
ENVIRONMENTAL CONDITIONS AND DISPERSAL TRAITS.— The fourth-corner analysis
revealed a significant relationship among dispersal traits and all environmental variables
considered with the exception of soil organic matter (SRLQ = 0.438, P = 0.001, Table S1).
Twenty two statistics in the fourth-corner matrix were significant (Fig. 2, Table S2).
According to this analysis, both anemochory and epizoochory were associated with sites
that showed special homogeneity with respect to the range of annual temperature. The
correlation ratio showed that both were negatively correlated with this range (2 = -0.187
and -0.04, respectively), i.e., they were favored in sites with small variation in annual
temperature. Both syndromes were associated to annual precipitation but in this case
responded in opposite ways: anemochory was associated with high values (wetter sites) and
epizoochory with low values (dryer sites, 2 = 0.124 and -0.022, respectively). Epizoochory
was also associated with the sites that get the smallest rainfall during the driest month (2 =
-0.155).
A positive relation was found between endozoochorous species and slope sites while
epizoochorous species had a major presence on ridge sites. Sinzoochorous and autochorous
species were not significantly related to any environmental variables studied.
Dry dehiscent and indehiscent fruits are mostly related to anemochorous and autochorous
dispersal (Appendix 1), and showed a similar pattern with respect to environmental
46
conditions, being present mainly in sites with lower soil moisture (dry indehiscent fruits, 2
= -0.073) and having lower precipitation during the driest month and lower annual
temperature range (dehiscent fruits 2 = -0.097 and -0.089, respectively). Dry dehiscent and
fleshy fruits were more frequent than expected by the permutation model in slopes while
dry dehiscent fruits were more frequently associated to valley sites.
DISCUSSION
The seed dispersal spectrum of the studied southern Ecuador dry forests was characterized
by the dominance of zoochory, followed by anemochory, and small percentages of
autochorous and polychorous species. These results are consistent with those reported for
other Neotropical dry forests (Table 4). Among zoochorous species, endozoochory was the
prevalent strategy, while for anemochory, six morphological groups with similar
percentages were found. Thus, functional diversity was higher for anemochorous compared
to zoochorous species.
Among zoochorous species, the high proportion of unprotected diaspores with bright colors
could suggest that bird dispersal was prevalent (Van der Pijl 1969, Janson 1983, Link &
Stevenson 2004, Cousens et al. 2008), in contrast with the low proportion of protected
diaspores with dull colors, which are preferentially consumed by mammals (Van der Pijl
1969, Janson 1983, Link & Stevenson 2004, Cousens et al. 2008). Nevertheless, the
scarcity of fleshy fruits that is usually observed during the dry season in this type of
vegetation could force a wide range of animals to feed on any type of fruit that is available
(Griz & Machado 2001).
47
Surprisingly, when we examined the individual abundance of each species, we found a shift
in dominance from zoochory to anemochory. This means that, although there is a high
variety of zoochorous species, most of the studied community corresponds to individuals
which do not provide any reward for dispersing animals. Therefore, from a functional
perspective of dispersal interactions, we must consider not only the proportion of species by
dispersal syndrome, but also the abundance of each species. This shift, together with the
existing relationships between environmental conditions and dispersal syndromes (see
below) may be due to the relative abundance of habitats that are suitable for species with
different dispersal syndromes. In this particular case, it is possible that relative abundance
of suitable habitats for anemochorous species is higher than relative abundance of suitable
habitats for zoochorous species. Hence the predominance of one syndrome or another may
depend on the spatial heterogeneity in environmental conditions. In addition, consideration
of the phylogenetic signal in the study of dispersal syndromes also deserves future attention
in elucidating the relative frequency of dispersal syndromes using species or individual
abundances.
Dispersal syndromes showed an association with growth form, as shown in previous studies
(Van der Pijl 1969, Wikander 1984, Willson et al. 1990, Justiniano & Fredericksen 2000,
Griz & Machado 2001). In particular, epizoochory was found only among shrubs. This is
consistent with the need for contact between hooked or sticky diaspores with the body of
passing animals, which constrains the growth form of the species having this dispersal
syndrome (Willson et al. 1990). Contrary to what has been reported for other tropical dry
forest sites, endozoochory was not restricted to treelets and shrubs but was equitably
48
present in all growth forms (Wikander 1984, Justiniano & Fredericksen 2000). Further
research is needed to explain this finding.
Anemochory was observed mainly among tree species, but contrary to what has been
reported for other communities, it also occurred among treelets and shrubs. According to
Griz and Machado (2001), this is related to the loss of foliage that affects both trees and
shrubs in dry forests, which allows wind circulation not only at the canopy, but also
through the understory. However, there was a difference in the types of anemochorous
strategies among growth forms. Floater and undulator diaspores were prevalent among tree
species, which may be related to their rapid rates of fall (Augspurger 1986, Ibarra-
Manríquez et al. 1991, Cousens et al. 2008), so their release from greater heights would
allow better use of horizontal wind force and reaching longer distances (Cousens et al.
2008). Conversely, diaspores with rolling autogyro structures were more common among
shrub and treelet species. These diaspores have slow rates of fall that allow dispersing the
seeds away from the mother plant even when released from a moderate height (Augspurger
1986, Cousens et al. 2008).
Despite being a rare syndrome in the community, autochory was found in the three growth
forms considered. Hughes et al. (1994) note that distances achieved by autochory are
constrained by the physical mechanism itself and rarely exceed a few meters. So, effective
escape from competition in the immediate vicinity will only be achieved if the parent plant
is small. We, and other researchers before (Griz & Machado 2001), found some exceptions
to this pattern. One possible explanation for the presence of autochory between taller plants
may be that secondary dispersal agents that are not considered in this study, (e.g. water,
49
ingestion by herbivores, or handling rodents or insects) act to move seeds at farther
distances (Wenny 2005). Another possibility is that tall trees can disperse seeds at greater
distances. For example, Hura crepitans has explosive capsules that can release the seeds up
to 14 m away (Van der Pijl 1969).
anemochory
autochory
endozoochory
epizoochory
polichory
sinzoochory
dry dehiscent fruit
dry indehiscent fruit
fleshy fruit
inclin
atio
n
tem
pe
ratu
re r
an
ge
rain
fall d
rie
st m
on
th
pre
cip
ita
tio
n
inte
rve
ntio
n
so
il m
ois
ture
so
il o
rga
nic
ma
tte
r
rid
ge
slo
pe
va
lle
y
FIGURE 2. Results of fourth-corner test comparing environmental variables with dispersal
syndromes and fruit type. White cells: non-significant results, soft gray: significantly negative
deviations from the permutational null model; dark gray: significantly positive deviations. First
seven columns (slope to soil organic matter): normalized measures of within-group homogeneity
(within group sum of squares divided by the total sum of squares). Three last columns (ridge to
valley): cell values of the contingency table analyses.
50
Our results demonstrate with quantitative data, and through a direct relation, that
environmental conditions affect the structure of seed dispersal spectrum in a Neotropical
dry forest community. Thus, anemochory was associated with sites that had higher annual
precipitation and lesser variation in the annual temperature range. This trend is contrary to
what has been reported for other Neotropical dry forests (Daubenmire 1972, Gentry 1982,
Howe & Smallwood 1982, Tabarelli et al. 2003, Ragusa-Netto & Silva 2007). For example,
Howe and Smallwood (1982) found a significant negative correlation between the
percentages of wind dispersed species and annual precipitation in six different
communities. This discrepancy could be due to those studies considering only frequency of
dispersal syndromes at the species level without accounting for individual species
abundance, in opposition to our four corner statistics.
The exclusive association of endozoochorous species with slope sites is difficult to
rationalize. A relationship between dispersal by vertebrates and increasing moisture and
soil fertility has been reported for temperate plant communities (Milewski 1986, Willson et
al. 1989, 1990, Bronstein et al. 2007). Similarly, some researchers have shown a possible
association between the proportion of zoochorous species and the quantity and actual
duration of rainfall (Frankie et al. 1974, Griz & Machado 2001, Tabarelli et al. 2003,
Ragusa-Netto & Silva 2007). According to Bullock (1995) sites that are drier and moister
because of local topography, support lower and higher frequencies of zoochorous tree
species, respectively. Within our study area, the soils in slope sites vary largely in the
ranges of moisture, soil organic matter, carbon and nitrogen, whereas soils in ridge sites are
more homogeneous and, on average, have lower values for these parameters. This could
explain the observed pattern.
51
Despite being an uncommon strategy in the studied community, epizoochory showed an
association with sites of lower precipitation and lower annual temperature range. Other
studies have shown a tendency of epizoochorous species to associate with open and dry
habitats (Gottsberger & Silberbauer–Gottsberger 1983, Sørensen 1986, Willson et al. 1990,
Mori & Brown 1998). This trend, together with the association found between these species
and ridge sites may partially reflect the availability of dispersal agents in those areas
(Gottsberger & Silberbauer–Gottsberger 1983, Willson & Traveset 2000) and possibly
indicate a better adhesion ability under dry conditions.
We found an association between dispersal syndromes and fruiting phenology, which is
consistent with those reported in other Neotropical dry forests (Frankie et al. 1974, Bullock
1995, Justiniano & Fredericksen 2000, Batalha & Mantovani 2000, Griz & Machado 2001)
(Table 4). In anemochorous species, the peak of fruiting was observed during the dry
season, whereas in zoochorous species it was observed in the rainy season. This
phenological segregation of dispersal syndromes mirrored the environmental correlates
described above. This suggests that spatio-temporal changes in physical conditions may be
related to important ecological factors for dispersal. The loss of leaves of most species
during the dry season allows a higher wind circulation, even in the understory, thereby
facilitating anemochory (Wikander 1984, Ibarra-Manríquez et al. 1991, Griz & Machado
2001). Moreover, for other Neotropical dry forest the dominance of zoochory during the
rainy season has been linked to seasonality in temperature and water availability, which set
limits on development time and maturation of fruits (Herrera & Pellmyr 2002). It is also
worth considering that in these ecosystems, animal dispersal agents are more active during
52
the rainy season. These patterns can influence the evolution of fruiting seasons
(Wheelwright & Janson 1985, Griz & Machado 2001, Herrera & Pellmyr 2002).
TABLE 4 Seed dispersal spectrum for other seasonally Neotropical dry forests. Dispersal syndrome
(ane = anemochory; aut = autochory; zoo = zoochory; pas= passive autochory). Growth form (T=
tree; S= shrub; CL= climber; CT= cactus; H= herb; P= phanerophytes; CH= chamaephytes; G=
geophytes).
Altitude
(m a.s.l.)
Rainy period
(months)
Mean annual
rainfall (mm)
Dispersal syndrome
Dominant syndrome by
season (%)
Ref. Locality Growth form
Anemochory
(%)
Zoochory
(%)
Autochory
(%) n
Dry
season Rainy season
1 Pernambuco (Brazil) - 7 549 T, S, CL, CT, H 33 36 314 42 ane (47) zoo (38)
2 Guanacaste
(Costa Rica) - 4 1533 T 31 512 18 104 ane (43) zoo (90)
3 N Venezuela 200 7 957 P, CH, G, CL 42 30 284 167 - -
4 Planalto da Ibiapaba
(Brazil) 700-900 - 668-1289 T, S 6 52 424 38 - zoo
5 Foothills of Urucum
(Brazil) 150-200 6 1400 Canopy T 27 53 20 56 ane-aut zoo
6 Caatinga (Brazil) 500 5 803 - 32 26 42 19 ane zoo
7 Cerrado (Brazil) - 6 1300 - 30 52 18 271 ane zoo
8 Santa Cruz (Bolivia) - 7 1100 Canopy and
subcanopy T 44 561 - (-) 39 ane (59) zoo-pas (581)
9 Cerrado (Brazil) 660-730 9 1499 P 26 62 12 108 ane zoo
10 Caatinga (Brazil) - 3-7 240-900 T, S - (-) 39.83 - (-) 103 - -
11 S Ecuador 160-800 5 300 - 700 T, S 28 54 184,5 160 ane (59) zoo (68)
1Dispersal by zoochory and passive autochory;
2Fleshy fruits;
3Dispersal by vertebrates;
4Including
passive autochory; 5Including polychory.
References: 1. Griz and Machado (2001); 2. Frankie et al. (1974); 3. Wikander (1984); 4.
Vasconcelos (2006); 5. Ragusa-Netto and Silva (2007); 6. Machado and Barros (1997); 7.
Gottsberger and Silberbauer-Gottsberger (1983); 8. Justiniano and Fredericksen (2000); 9. Batalha
and Mantovani (2000); 10. Oliveira and Moreira (1992); 11. This study.
53
However, our results on fruiting phenology should be considered with caution because data
for some species were obtained only from herbarium specimens, and this does not
guarantee that the recorded data would cover the actual fruiting period for all species
(Borchert 1996).
Summarizing, the observed patterns suggest that the seed dispersal spectrum of dry forests
of southern Ecuador is strongly affected by spatio-temporal changes in environmental
conditions. Moreover, these forests hold a large variety of plants requiring animal
assistance to seed dispersal. However, fruit availability is limited by fruiting seasonality
and a low frequency of individuals of zoochorous species. In altered habitats, like those
studied here, maintaining seed dispersal processes is important to preserve the connectivity
among patches and foster the recolonization of degraded zones. Our identification of
general patterns will help to understand the ecology of this ecosystem, to determine which
species are more susceptible to environmental changes and to inform restoration projects.
ACKNOWLEDGEMENTS
This research was financed by the AECID project A/024796/09, REMEDINAL2 and
Universidad Técnica Particular de Loja (UTPL). The authors wish to thank C.I. Espinosa
and O. Cabrera for access to floristic composition and environmental data. Thanks to the
staff of Herbarium UTPL and thanks Bolivar Merino for their help in species
determinations and to Paul Cahén for the English language revision. Thanks to three
anonymous referees whose constructive criticism improved a previous version of the
manuscript.
54
LITERATURE CITED
AGUIRRE, Z., AND L. KVIST, 2005. Composición florística y estado de conservación de los
bosques secos del sur-occidente del Ecuador. Lyonia 8: 41-67.
AGUIRRE, Z., L. KVIST., AND O.SÁNCHEZ. 2006a. Bosques secos en Ecuador y su
diversidad. In M. MORAES R., B. ØLLGAARD, L. P. KVIST, F. BORCHSENIUS AND H.
BALSLEV (Eds.). Botánica Económica de los Andes Centrales, pp. 162-187.
Universidad Mayor de San Andrés, La Paz.
AGUIRRE, Z., R. LINARES-PALOMINO, AND L. KVIST. 2006b. Especies leñosas y formaciones
vegetales en los bosques estacionalmente secos de Ecuador y Perú. Arnaldoa 13:
324–350.
ARBELÁEZ, M.V., AND A. PARRADO-ROSSELLI. 2005. Seed dispersal modes of the sandstone
plateau vegetation of the Middle Caquetá River Region, Colombia Amazonia.
Biotropica 37: 64-72.
AUBIN, I., M. OUELLETTE, P. LEGENDRE, C. MESSIER, AND A. BOUCHARD. 2009.
Comparison of two plant functional approaches to evaluate natural restoration along
an old-field – deciduous forest chronosequence. J. Veg. Sci. 20: 185–198.
AUGSPURGER, C. 1986. Morphology and dispersal potential of wind-dispersed diaspores of
Neotropical trees. Am. J. Bot. 73: 353-363.
BATALHA, M.A., AND W. MANTOVANI. 2000. Reproductive phenological patterns of
cerrado plant species at the Pé-de-Gigante Reserve (Santa Rita do Passa Quatro, Sp,
55
Brazil): A comparison between the herbaceous and woody floras. Rev. Brasil. Biol.
60: 129-145.
BEST, B. J., AND M. KESSLER. 1995. Biodiversity and conservation in Tumbesian Ecuador
and Peru. BirdLife International, Cambridge.
BORCHERT, R. 1996. Phenology and flowering periodicity of Neotropical dry forest species:
evidence from herbarium collections. J. Trop. Ecol. 12: 65-80.
BRONSTEIN, J.L., I. IZHAKI, R. NATHAN, J.J. TEWKSBURY, O. SPIEGEL, A. LOTAN, AND O.
ALTSTEIN. 2007. Fleshy-fruited plants and frugivores in desert ecosystems. In A.J.
DENNIS, E.W. SCHUPP, R.J. GREEN, AND D.A. WESTCOTT (Eds.). Seed dispersal:
theory and its application in a changing world. pp. 148-177. CAB International,
Oxford, UK.
BULLOCK, S. H. 1995. Plant reproduction in Neotropical dry forests. In S., BULLOCK, H.,
MOONEY AND E. MEDINA (Eds.). Seasonally dry tropical forests. Cambridge
University Press, Cambridge, UK. pp. 277 – 297.
BUTLER, D. W., R. J. GREEN, D. LAMB, W. J. F. MCDONALD AND P. I. FORSTER. 2007.
Biogeography of seed-dispersal syndromes, life-forms and seed sizes among woody
rain-forest plants in Australia’s subtropics. J. Biogeogr. 34: 1736–1750.
CHEN, J., T. FLEMING, L. ZHANG, H. WANG, Y. LIU. 2004. Patterns of fruit traits in a tropical
rainforest in Xishuangbanna, SW China. Acta Oecol. 26: 157–164.
56
COUSENS, R., C. DYTHAM, AND R. LAW. 2008. Dispersal in plants. A population
perspective. Oxford University Press Inc., New York.
DAUBENMIRE, R. 1972. Phenology and other characteristics of tropical semideciduous
forest in north-western Costa Rica. J. Ecol. 60: 147-70.
DRAY, S., A.B. DUFOUR, AND D. CHESSEL. 2007. The ade4 package-II: Two-table and K-
table methods. R News 7: 47-52.
DRAY, S. AND P. LEGENDRE. 2008. Testing the species traits–environment relationships:
The fourth-corner problem revisited. Ecology 89: 3400–3412.
ESPINOSA, C.I., O. CABRERA, A. LUZURIAGA, AND A. ESCUDERO. 2010. What factors affect
diversity and species composition of endangered Tumbesian dry forest s in southern
Ecuador? Biotropica. DOI: 10.1111/j.1744-7429.2010.00665.x
FRANKIE, G. W., H.G. BAKER, AND P.A. OPLER. 1974. Comparative phenological studies of
trees in tropical wet and dry forests in the Lowlands of Costa Rica. J. Ecol. 62: 881-
919.
GENTRY, A.H. 1982. Patterns of Neotropical plant species diversity. Evol. Biol. 15: 1-84.
GOTTSBERGER, G., AND I. SILBERBAUER–GOTTSBERGER. 1983. Dispersal and distribution in
the Cerrado vegetation of Brazil. Sonderb. Naturwiss 7: 325-352.
GRIZ, L.M.S., AND I.C.S. MACHADO. 2001. Fruiting phenology and seed dispersal
syndromes in Caatinga, a tropical dry forest in the northeast of Brazil. J. Trop. Ecol.
17: 303-321.
57
HARLING, G. AND L. ANDERSON (Eds). 1977 – 2007. Flora of Ecuador. Board Editorial,
Berlings, Arlov, Sweden.
HERRERA, C. M., AND O. PELLMYR. 2002. Plant – Animal Interactions. An evolutionary
approach. Blackwell Publishing. Oxford, UK.
HIJMANS, R. J., S. E. CAMERON, J. L. PARRA, P. G. JONES, AND A. JARVIS. 2005. Very high
resolution interpolated climate surfaces for global land areas. Int. J. Climatol. 25:
1965-1978.
HOWE, H.F., AND J. SMALLWOOD. 1982. Ecology of seed dispersal. Annu. Rev. Ecol. Syst.
13: 201-228.
HOWE, H.F., AND L.C. WESTLEY. 1988. Mechanics and ecology of mutualism. In H.F.
HOWE, AND L.C. WESTLEY. (Eds.). Ecological relationships of plants and animals,
pp. 107-160. Oxford University Press, Oxford.
HUGHES, L., M. DUNLOP, K. FRENCH, M. LEISHMAN, B. RICE, L. RODGERSON, AND M.
WESTOBY. 1994. Predicting dispersal spectra: a minimal set of hypotheses based on
plant attributes. J. Ecol. 82: 933-950.
IBARRA-MANRÍQUEZ, G., B. SÁNCHEZ-GARFIAS, AND L. GONZÁLEZ-GARCÍA. 1991.
Fenología de lianas y árboles anemocoros en una selva cálido-húmeda de México.
Biotropica 23: 242-254.
JANSON, CH. 1983. Adaptation of fruit morphology to dispersal agents in a Neotropical
forest. Science 219: 187-189.
58
JORDANO, P. 2000. Fruits and Frugivory. In M. Fenner (Ed.). Seeds: The Ecology of
Regeneration in Plant Communities (second edition), pp. 125 – 165. CABI
Publishing, UK.
JØRGENSEN, P.M., AND S. LEÓN-YÁNEZ (Eds.). 1999. Catalogue of the vascular plants of
Ecuador. Monogr. Syst. Bot. Missouri Bot. Gard. 75: i–viii, 1–1182.
JUSTINIANO, M., AND T. FREDERICKSEN. 2000. Phenology of tree species in Bolivian dry
forests. Biotropica 32: 276-281.
KNIGTH, R.S. 1986. Interrelationships between fruit types in southern African trees and
environmental variables. J. Biogeogr. 13: 99–108.
LEGENDRE, P., R. GALZIN, AND M. HARMELIN-VIVIEN. 1997. Relating behavior to habitat:
solutions to the fourth-corner problem. Ecology 78: 547–562.
LINK, A., AND P.R. STEVENSON. 2004. Fruit dispersal syndromes in animal disseminated
plants at Tinigua National Park, Colombia. Rev. Chil. Hist. Nat 77: 319-334.
MACHADO, I., AND L. BARROS. 1997. Phenology of Caatinga Species at Serra Talhada, PE,
Northeastern Brazil. Biotropica 29: 57-68.
MILEWSKI, A.V. 1986. A comparison of bird–plant relationships in southern Australia and
southern Africa. In H.A. FORD, AND D.C. PATON (Eds.). The dynamic partnership:
birds and plants in Southern Africa. pp. 111–118. Flora and Fauna of South
Australia Handbooks Committee.
59
MORALES, J.M., AND T.A. CARLO. 2006. The effects of plant distribution and frugivore
density on the scale and shape of dispersal kernels. Ecology 87: 1489–1496.
MORI, S., AND J. BROWN. 1998. Epizoochorous dispersal by barbs, hooks, and spines in a
lowland moist forest in Central French Guiana. Brittonia 50: 165-173.
NAVARRO, T., V. PASCUAL. C.L. ALADOS, AND B. CABEZUDO. 2009. Growth forms,
dispersal strategies and taxonomic spectrum in a semi-arid shrubland in SE Spain. J.
Arid Environ. 73: 103–112.
OLIVEIRA, P.E.A.M. DE AND A.G. MOREIRA. 1992. Anemocoria em espécies de Cerrado e
Mata de Galeria de Brasília, DF. Rev. Br. Bot. 15: 163-174.
PENNINGTON, T.D., C. REYNEL, AND A. DAZA. 2004. Illustrated guide to the trees of Peru.
DH Books, England.
RAGUSA-NETTO, J., AND R.R. SILVA. 2007. Canopy phenology of a dry forest in western
Brazil Braz. J. Biol. 67: 569-575.
SCHULZE, E., E. BECK, AND K. MÜLLER-HOHENSTEIN. 2002. Plant Ecology. Springer Berlin,
Heidelberg, New York.
SØRENSEN, A.E. 1986. Seed dispersal by adhesion. Ann. Rev. Ecol. Syst. 17: 443-463.
SPJUT, R. W. 1994. A systematic treatment of fruit types. The New York Botanical Garden.
U.S.A.
60
STRYKSTRA, R. J., R. M. BEKKER, AND J. VAN ANDEL. 2002. Dispersal and life span spectra
in plant communities: a key to safe site dynamics, species coexistence and
conservation. Ecography 25: 145–160.
TABARELLI, M., A.VICENTE, AND D.C. A. BARBOSA. 2003. Variation of seed dispersal
spectrum of woody plants across a rainfall gradient in north-eastern Brazil. J. Arid
Environ. 53: 197–210.
VAN DER PIJL, L. 1969. Principles of dispersal in higher plants. Springer-Verlag Berlin
Heidelberg, New York.
VASCONCELOS, S.F.D. 2006. Fenologia e síndromes de dispersao de espécies arbustivas e
arbóreas acorrentes em uma área de carrasco no planalto da Ibiapaba, Ceará. Msc
Dissertation, Universidade Federal de Pernambuco, Brasil, 61 pp.
VÁZQUEZ, M. A., M. LARREA, L. SUÁREZ, AND P. OJEDA. (Eds.). 2001. Biodiversidad en los
bosques secos del suroccidente de la provincia de Loja: un reporte de las
evaluaciones ecológicas y socioeconómicas rápidas. EcoCiencia, MAE, Herbario
Loja y Proyecto Bosque Seco. Quito.
WENNY, D. 2001. Advantages of seed dispersal: a re-evaluation of directed dispersal. Evol.
Ecol. Res. 3: 51–74.
WENNY, D. 2005. Post-dispersal seed fate of some cloud forest tree species in Costa Rica.
In: FORGET, P-M., J.E. LAMBERT, P.E. HULME AND S.B. VANDER WALL (EDS). Seed
Fate: Predation, dispersal and seedling establishment. CAB International.
61
WHEELWRIGHT, N., AND C. JANSON. 1985. Colors of fruit displays of bird-dispersed plants
in two tropical forests. Am. Nat. 126: 777-799.
WIKANDER, T. 1984. Mecanismos de dispersión de diásporas de una selva decidua en
Venezuela. Biotropica 16: 276-283.
WILLSON, M. F., A. K. IRVINE, AND N.G. WALSH. 1989. Vertebrate dispersal syndromes in
some Australian and New Zealand plant communities, with geographic
comparisons. Biotropica 21: 133-147.
WILLSON, M. F., B.L. RICE, AND M. WESTOBY. 1990. Seed dispersal spectra: a comparison
of temperate plant communities. J. Veg. Sci. 1: 547-562.
WILLSON, M. F., AND A. TRAVESET. 2000. The ecology of seed dispersal. In M. FENNER
(Ed.). Seeds: the ecology of regeneration in plant communities (second edition), pp.
85-110. CABI Publishing, UK.
62
MATERIAL SUPLEMENTARIO PARA EL CAPÍTULO 1.
FIGURE S1. Location of study area. Gray shade represents the area of distribution of dry forest in
southern Ecuador. Black dots locate the 48 forest stands (109 plots) studied. See text for details.
63
APPENDIX 1 Growth forms and diaspore traits for 160 species in South Ecuadorian Dry Forests.
Dispersal unit (F= fruit; S= seed). Diaspore morphology (1Protected diaspore;
2Flower persistent
structures which operate on dispersal). Dispersal syndrome (Z= zoochroy; AN= anemochory; AU=
autochory; P= autochory/zoochory; end=endozoochory; sin= sinzoochory; mir= mirmecochory;
epi= epizoochory; a/m= active autochory/mirmechory; p/e= passive autochory/endozoochory; ac=
active autochory; pas= pasive autochory; aut= autogyro; roll= rolling-autogyro; flo= floaters; und=
undulator; hel= helicopter; tum= tumbler).
Species Growth
form
Fruit
type
Dispersal
Unit
Diaspore
morphology
diapore
color
dispersal
syndrome References
Achatocarpaceae
Achatocarpus
pubescens shrub berry F fleshy pulp green Z - end 13, 26
Anacardiaceae
Loxopterygium
huasango tree samara F samara brown AN - roll 13, 14
Mauria
heterophylla tree drupe F fleshy pulp red Z - end 13, 14, 44
Mauria
membranifolia tree drupe F fleshy pulp orange Z - end 13, 41
Mauria suaveolens tree drupe F fleshy pulp red Z - end 14, 44
Apocynaceae
Aspidosperma sp tree follicle S samara brown AN - und 13, 14, 33
Rauvolfia
tetraphylla shrub drupe F fleshy pulp black Z - end
14, 33, Obs.
Pers.
Asteraceae
Fulcaldea laurifolia tree achene F pappus brown AN - hel 43
Verbesina
pentantha shrub achene F pappus brown AN - flo 44, 29
64
Vernonanthura
patens tree achene F pappus brown AN - hel 45
Wedelia grandiflora shrub achene F pappus brown AN - hel 33, 43
Bignoniaceae
Delostoma
integrifolium tree capsule S samara brown AN - roll 13, 14
Jacaranda sparrii tree capsule S samara brown AN - roll 13, 14
Tabebuia bilbergii treelet capsule S samara brown AN - roll 14
Tabebuia
chrysantha tree capsule S samara brown AN - roll 14
Tecoma castanifolia shrub capsule S samara brown AN - roll 14
Tecoma stans shrub capsule S samara brown AN - roll 14
Tecoma
weberbaueriana shrub capsule S samara brown AN - roll 14
Bixaceae
Cochlospermum
vitifolium tree capsule S plumed brown AN - flo 33, Obs. Pers.
Bombacaceae
Cavanillesia
platanifolia tree samara F samara brown AN - tum 31
Ceiba trichistandra tree capsule S woolly brown AN - flo 5, 13, 14, 33
Eriotheca ruizii tree capsule S woolly brown AN - flo 33, Obs. Pers.
Ochroma
pyramidale tree capsule S woolly brown AN - flo 14, 33
Pachira rupicola tree capsule S woolly brown AN - flo 13, 33, 44
Pseudobombax
millei tree capsule S woolly brown AN - flo 13, 14, 33
Boraginaceae
Cordia alliodora tree capsule F samara 2 brown AN - hel 27, Obs. Pers.
Cordia lantanoides shrub drupe F fleshy pulp orange Z - end 8, 43, 44
65
Cordia lutea shrub drupe F fleshy pulp withe Z - end 4, 13
Cordia
cylindrostachya shrub drupe F fleshy pulp red Z - end 5, 43
Cordia hebeclada tree drupe F fleshy pulp black Z - end 8, 26, 33
Cordia macrantha treelet capsule F samara 2 brown AN - hel 43, 44
Cordia
macrocephala shrub drupe F fleshy pulp red Z - end Obs. Pers.
Tournefortia
polystachya treelet drupe F fleshy pulp yellow Z - end 26, 33, 41, 44
Burseraceae
Bursera graveolens tree drupe F fleshy pulp mixed Z - end 33, Obs. Pers.
Cactaceae
Armatocereus
brevispinus treelet berry F fleshy pulp green Z - end 14, 33
Armatocereus
cartwrightianus tree berry F fleshy pulp red Z - end 14, 33
Armatocereus
godingianus treelet berry F fleshy pulp green Z - end 14, 33
Armatocereus
matucanensis treelet berry F fleshy pulp green Z - end 14, 33
Cereus diffusus treelet berry F fleshy pulp mixed Z - end 14, 33
Espostoa lanata shrub berry F fleshy pulp red Z - end 14, 33
Opuntia quitensis shrub berry F fleshy pulp green Z - end 14, 33
Opuntia
soederstromiana shrub berry F fleshy pulp red Z - end 14, 33
Caesalpiniaceae
Caesalpinia
glabrata tree legume F
without
structures brown AU - pas 40, Obs. Pers.
Cercidium praecox treelet legume F without
structures brown AU - pas 13, 26, 40
Cynometra
bauhiniifolia tree legume F
without
structures brown AU - pas 13, 26, 33
66
Senna incarnata shrub legume F without
structures brown AU - pas 33, 35, 40
Senna pistaciifolia shrub legume F without
structures brown AU - pas 21, 33, 35
Senna spectabilis shrub legume F without
structures brown AU - pas 27, 33, 35, 41
Capparidaceae
Capparis flexuosa treelet berry S sarcotesta brown Z - mir 18, 20, 40
Capparis scabrida shrub berry S fleshy pulp brown Z - end 30, 37, 44
Caricaceae
Carica parviflora shrub berry F fleshy pulp orange 1 Z - end
13, 28, 30, 41,
44
Carica microcarpa treelet berry F fleshy pulp red 1 Z - end 13, 28, 33, 41
Cecropiceae
Cecropia litoralis tree achenetum F fleshy pulp brown Z - end 16, 26, 33, 43
Celastraceae
Maytenus octogona tree capsule S aril mixed Z - end 13, 21, 28, 41,
44
Combretaceae
Terminalia
valverdeae tree samara F samara brown AN - roll 13, 33, 44
Convolvulaceae
Ipomoea carnea shrub capsule S plumed brown AN - 13, 24, 28, 33
Ipomoea
wolcottiana shrub capsule S plumed brown AN -
13, 24, 26, 28,
33, 44
Erythroxylaceae
Erythroxylum
glaucum tree drupe F fleshy pulp red Z - end 14, Obs. Pers
Euphorbiaceae
Acalypha
diversifolia shrub capsule S
explosive
dehiscence brown AU - ac
5, 33, 38, 43,
Obs. Pers.
67
Cnidoscolus
aconitifolius tree capsule S
explosive
dehiscence
/caruncle
mixed P - a/m 20, 25, 26, 32,
38
Croton
menthodorus shrub capsule S
explosive
dehiscence
/caruncle
mixed P - a/m 13, 20, 32, 33
Croton wagneri shrub capsule S
explosive
dehiscence
/caruncle
mixed P - a/m 13, 20, 28
Hura crepitans tree capsule S explosive
dehiscence brown AU - ac 13, 27, 31
Jatropha curcas shrub capsule F caruncle black Z - mir 18, 26, 33, 41,
Obs. Pers.
Phyllantus sp. shrub capsule S explosive
dehiscence brown AU - ac
2, 26, 33, Obs.
Pers.
Fabaceae
Centrolobium
ochroxylum tree samara F samara brown AN - aut 31, Obs. Pers.
Clitoria
brachystegia treelet legume F
without
structures brown AU - pas 2, 10, 26, 33
Coursetia caribaea shrub legume S without
structures brown AU - ac 26, 29
Coursetia
grandiflora shrub legume S
without
structures brown AU - ac 26, 29, 40
Cyathostegia
mathewsii tree legume S
without
structures mixed AU - pas 26, 40
Erythrina smithiana tree legume S without
structures mixed P - p/e 33, Obs. Pers.
Erythrina velutina tree legume S without
structures mixed P - p/e 20, 33
Geoffroea spinosa tree drupe F fleshy pulp green Z - sin 13, Obs. Pers.
Gliricidia breningii tree legume S without
structures brown AU - ac 13, 26, 40
Lonchocarpus
atrourpureus tree legume F
without
structures brown AU - pas 33, 43
68
Machaerium millei tree samara F samara mixed AN - aut 13, 33, Obs.
Pers.
Myroxylon
peruiferum tree samara F samara yellow AN - aut 14, 31, 33
Piscidia
carthagenensis tree legume F samara brown AN - tum 26, 31
Pterocarpus sp. tree samara F samara brown AN - und 5, 13, 26, 33,
35, 44
Flacourtiaceae
Muntingia calabura tree berry F fleshy pulp red Z - end 4, 13, 27, 33,
44, Obs. Pers.
Prockia crucis shrub berry F fleshy pulp2 mixed Z - end
26, 44, Obs.
Pers.
Lauraceae
Nectandra
acutifolia tree drupe F fleshy pulp mixed Z - end 29, 44
Lythraceae
Adenaria floribunda treelet capsule F blue/purple Z - end 26, 33, 44
Lafoensia
acuminata tree capsule S samara yellow AN - roll 33, Obs. Pers.
Malpighiaceae
Bunchosia deflexa tree drupe F fleshy pulp red Z - end 13, 26, 30, 33
Bunchosia
pseudonitida shrub drupe F fleshy pulp orange Z - end 13, 26, 33, 43
Malpighia
emarginata treelet drupe F fleshy pulp red Z - end 30, 44
Malpighia glabra treelet drupe F fleshy pulp red Z - end 1, 13, 26, 38,
41, 44
Malvaceae
Bastardia bivalvis shrub mericarp F arists brown Z - epi 20, 43
Pavonia sepium shrub mericarp S arists green Z - epi 9, 13, 33, 44
Meliaceae
69
Cedrela odorata tree capsule S samara brown AN - aut 33
Schmardaea
microphylla treelet capsule S samara brown AN - aut 29, 44
Trichilia hirta tree capsule S aril mixed Z - end 8, 13, 28, 33,
35
Mimosaceae
Acacia
macracantha tree legume F
without
structures brown AU - pas 40, 41
Acacia riparia shrub legume F without
structures brown AU - pas 40, 41
Albizia multiflora shrub legume S without
structures brown AU - pas 33, 40, 45
Leucaena trichodes tree legume S without
structures brown AU - pas 13, 26, 40, 41
Mimosa
acantholoba shrub craspedium F
without
structures brown AU - pas 26, 28, 33, 40
Mimosa caduca treelet legume S without
structures brown AU - pas
20, 26, 28, 33,
40, Obs. Pers.
Piptadenia flava tree legume S without
structures brown AU - pas 26, 28, 33
Pithecellobium
excelsum shrub legume S aril mixed Z - end 8, 26, 28
Prosopis juliflora treelet legume F fleshy pulp yellow Z - end 4, Obs. Pers.
Prosopis pallida treelet legume F fleshy pulp yellow Z - end 26, 37
Pseudosamanea
guachapele tree legume S
without
structures brown AU - pas 10, 26
Zapoteca
caracasana shrub legume S
without
structures brown AU - ac 29, 40
Monimiaceae
Siparuna eggersii treelet drupe F fleshy pulp mixed Z - end 14, 33
Moraceae
Brosimum tree drupe F fleshy pulp yellow Z - sin
12, 13, 14, 27,
70
alicastrum 33, 41
Ficus citrifolia tree syconium F fleshy pulp green Z - end 2, 14, 28, 41,
43
Ficus insipida tree syconium F fleshy pulp green Z - end 35, 43
Ficus jacobii tree syconium F fleshy pulp green Z - end 14, 43
Ficus maxima tree syconium F fleshy pulp green Z - end 14, 43
Ficus tonduzii tree syconium F fleshy pulp mixed Z - end 33, 43
Maclura tinctoria tree sorosis F fleshy pulp yellow Z - end 10, 14, 26, 41
Sorocea sprucei treelet drupe F fleshy pulp mixed Z - end 13, 14, 45
Sorocea trophoides tree drupe F fleshy pulp mixed Z - end 13, 14, 43
Myrtaceae
Psidium guajava shrub berry F fleshy pulp yellow Z - end 14, 33
Psidium guineense tree berry F fleshy pulp yellow Z - end 26, 30, 33
Nyctaginaceae
Bougainvillea
peruviana shrub achene F samara
2 brown AN - roll Obs. Pers.
Bougainvillea
spectabilis shrub achene F samara
2 brown AN - roll
Pisonia aculeata shrub achene F sticky
anthocarp brown Z - epi 13, 14, 38, 43
Opiliaceae
Agonandra excelsa tree drupe F fleshy pulp yellow Z - end 33, 43
Phytolaccaceae
Gallesia integrifolia tree samara F samara brown AN - aut 26, Obs. Pers.
Phytolacca dioica tree baccarium F fleshy pulp black Z - end 11, 28, 33, 41
Polemoniaceae
Cantua quercifolia shrub capsule S samara brown AN - roll 29, 44
Polygonaceae
Coccoloba mollis tree pseudodrupe F fleshy periant2 blue/purple Z - end 14, 19, 33
71
Coccoloba ruiziana treelet pseudodrupe F fleshy periant2 blue/purple Z - end 4, 14, 33
Ruprechtia
jamesonii tree achene F samara
2 brown AN - hel 13, 14, 40
Triplaris
cumingiana tree achene F samara
2 brown AN - hel 13
Rhamnaceae
Scutia spicata shrub drupe F fleshy pulp red Z - end 8, 26, 43
Ziziphus thyrsiflora treelet drupe F fleshy pulp yellow Z - sin 8, 41, Obs.
Pers.
Rosaceae
Prunus
subcorymbosa tree drupe F fleshy pulp red Z - sin 3, 26, 43
Rubiaceae
Randia sp shrub berry F fleshy pulp brown 1 Z - end 8, 26, 33, 41
Simira ecuadorensis shrub capsule S samara yellow AN - aut 33, Obs. Pers.
Rutaceae
Zanthoxylum fagara treelet follicle F mixed Z - end 7, 26, 28, 41,
44
Salicaceae
Salix humboldtiana tree capsule S plumed brown AN - flo 33
Sapindaceae
Cupania cinerea tree capsule S aril mixed Z - end 5, 8, 19, 33,
35
Cupania latifolia tree capsule S aril mixed Z - end 26, 33, 35
Dodonea viscosa shrub samara F samara brown AN - tum 26, Obs. Pers.
Sapindus saponaria tree drupe F fleshy pulp orange Z - end 14, 41
Sapotaceae
Pradosia montana tree drupe F fleshy pulp brown 1 Z - end 26, 33, 43
Scrophulariaceae
72
Buddleja americana shrub capsule S samara brown AN - roll 23, 29, 33
Solanaceae
Acnistus
arborescens shrub berry F fleshy pulp orange Z - end 8, 29, 30, 43
Cestrum
auriculatum shrub berry F fleshy pulp blue/purple Z - end
26, 33, Obs.
Pers.
Lycianthes lycioides shrub berry F fleshy pulp orange Z - end 42, 33
Solanum
confertiseriatum shrub berry F fleshy pulp green Z - end 8, 33, 39
Sterculariaceae
Guazuma ulmifolia tree capsule F mucilage brown Z - end 13, 15, 33,
Obs. Pers.
Byttneria parviflora shrub capsule F hooked spines brown Z - epi 2, 26, 33, 43
Theophrastaceae
Clavija euerganea shrub berry F fleshy pulp orange 1 Z - end 14, 17, 30, 33
Clavija pungens shrub berry F fleshy pulp orange 1 Z - end
8, 14, 17, 26,
30, 33
Tiliaceae
Heliocarpus
americanus tree capsule F stiff hairs brown AN - und 26, Obs. Pers.
Triumfetta
semitriloba shrub capsule F hooked spines brown Z - epi 21, 26, 33, 38
Ulmaceae
Celtis iguanaea shrub drupe F fleshy pulp yellow Z - end 33, 43, Obs.
Pers.
Celtis loxensis shrub drupe F fleshy pulp yellow Z - end 43, Obs. Pers.
Trema micrantha tree drupe F fleshy pulp red 1 Z - end
11, 12, 13, 26,
33, 41
Urticaceae
Urera caracasana treelet achene F fleshy periant 2 orange Z - end
6, 12, 19, 21,
26, 36, 44
73
Verbenaceae
Aegiphila
cuatrecasasii tree drupe F fleshy pulp
2 blue/purple Z - end 2, 19, 26, 33
Citharexylum sp tree drupe F fleshy pulp red Z - end 26, Obs. Pers.
Duranta
dombeyana shrub drupe F fleshy pulp
2 yellow Z - end 17, 26, 29, 43
Cornutia
pyramidata treelet drupe F fleshy pulp
2 green Z - end 26, 43
Lantana trifolia shrub drupe F fleshy pulp blue/purple Z - end 17, 26, 33
REFERENCES:
1. AGUIRRE, Z., AND L. KVIST. 2005. Compisición florística y estado de conservación
de los bosques secos del sur-occidente del Ecuador. Lyonia 8: 41-67.
2. BATALHA, M.A., AND W. MANTOVANI. 2000. Reproductive phenological patterns of
cerrado plant species at the Pé-de-Gigante Reserve (Santa Rita do Passa Quatro, Sp,
Brazil): A comparison between the herbaceous and woody floras. Rev. Brasil. Biol.
60: 129-145.
3. BURROWS, J.C. 1994. Fruit types and seed dispersal modes of woody plants in
Ahuriri Summit Bush, Port Hills, western Banks Peninsula, Canterbury, New
Zealand. New Zeal. J. Bot. 32: 169–181.
4. COSSÍOS, E. 2005. Dispersión y variación de la capacidad de germinación de
semillas ingeridas por el zorro costeño (Lycalopex sechure) en el Santuario
Histórico Bosque de Pómac, Lambayaque. MSc Dissertation, Universidad Nacional
Mayor de San Marcos, Perú, 74 pp.
74
5. CHAZDON, R. L., S. CAREAGA, W. WEBB, AND O. VARGAS. 2003. Community and
phylogenetic structure of reproductive traits of woody species in wet tropical
forests. Ecol. Monograph. 73: 331–347.
6. COLLEEN, K. 1995. Seed size in tropical trees: a comparative study of factors
affecting seed size in Peruvian angiosperms. Oecologia 102: 337 - 388
7. DALLING, J. W., S. HUBELL, AND K. SILVERA. 1998. Seed dispersal, seedling
establishment and gap partitioning among tropical pioneer trees. J. Ecol., 86: 674 -
689
8. DE LA TORRE, L., H. NAVARRETE, P. MURIEL M., M.J. MACÍA, AND H. BALSLEV
(EDS.). 2008. Enciclopedia de las plantas útiles del Ecuador. Herbario QCA de la
Escuela de Ciencias Biológicas de la Pontificia Universidad Católica del Ecuador
and Herbario AAU del Departamento de Ciencias Biológicas de la Universidad de
Aarhus. Quito and Aarhus .
9. DE MARZI, V. 2006. Cien plntas argentinas. Albatros, Buenos Aires, Argentina
10. FLORES, E.M. 2002. Pseudosamanea guachapele (Kunth) Harms. In Vozzo J. A.
(Ed.). Tropical Tree seed manual. Part II Species Descriptions. United States
Department of Agriculture Forest Service, Washington DC. U.S.A.
11. GALETTI, M. 2000. Frugivory by toucans (Ramphastidae) at two altitudes in the
atlantic forest of Brazil. Biotropica 32: 842-850.
75
12. GALINDO-GONZÁLEZ, J. 1998. Dispersión de semillas por murciélagos: Su
importancia en la conservación y regeneración del bosque tropical. Acta Zool. Mex.
74: 57-74.
13. GENTRY, A.H. 1993. A field guide to woody plants of northwest of South America.
Chicago.
14. HARLING , G., AND L. ANDERSON (EDS.). 1977 – 2007. Flora of Ecuador. Berlings,
Arlov, Sweden.
15. JANZEN, D.H. 1982. Natural history of Guacimo fruits, Sterculiaceae, Guazuma
ulmifolia with respect to consumption by large mammals. Am. J. Bot. 69: 1240-
1250.
16. JUSTINIANO, M., AND T. FREDERICKSEN. 2000. Phenology of tree species in Bolivian
dry forests. Biotropica 32: 276-281.
17. KUBITZKI, K. (ED.). 2004. The families and genera of vascular plants. Flowering
plants, Dicotyledons: Celastrales, Oxalidales, Rosales, Cornales, Ericales. vol. VI.
Springer-Verlag Berlin Heidelberg, Germany.
18. LEAL, I. R., R. WIRTH, AND M. TABARELLI. 2007. Seed dispersal by ants in the semi-
arid Caatinga of north-east Brazil. Ann. Bot. 99: 885-894.
19. LINK, A., AND P.R. STEVENSON. 2004. Fruit dispersal syndromes in animal
disseminated plants at Tinigua National Park, Colombia. Rev. Chil. Hist. Nat. 77:
319-334.
76
20. LIOGIER A. 1994. Descriptive flora of Puerto Rico and adjacent islands:
Spermatophyta. Vol 3. Cyrillaceae to Myrtaceae. Universidad de Puerto Rico.
Puerto Rico
21. MCMULLEN, C. K. 1999. Flowering Plants of the Galápagos. Cornell University
Press. South Korea.
22. NOIR, A., S. BRAVO, AND R. ABDALA. 2002. Mecanismos de dispersión de algunas
especies de leñosas nativas del Chaco Occidental y Serrano. Quebracho 9: 140-150
23. OCAMPO-ACOSTA, G. 2004. Flora del Valle de Tehuacán-Cuicatlán.
BUDDLEJACEAE. Universidad Nacional Autónoma de México 39: 1-12.
24. PARRA-TABLA, V., AND S. H. BULLOCK. 2005. Ecological and selective effects of
stigma-anther separation in the self-incompatible tropical tree Ipomoea wolcottiana
(Convolvulaceae). Plant Syst. Evol. 252: 85–95
25. PARRA-TABLA, V., V. RICO-GRAY, AND M. CARBAJAL. 2004. Effect of defoliation
on leaf growth, sexual expression and reproductive success of Cnidoscolus
aconitifolius. Plant Ecol. 173: 153–160
26. PENNINGTON T.D., C. REYNEL, AND A. DAZA. 2004. Ilustrated guide to the trees of
Peru. David Hunt, Sherborne, England."
27. SALAZAR, R., C. SOIHET, AND J. MÉNDEZ. 2000. Manejo de semillas de 100 especies
forestales de América Latina. Centro Agronómico Tropical de Investigación y
Enseñanza (CATIE), Turrialba, Costa Rica.
77
28. SPJUT, R. W. 1994. A systematic treatment of fruit types. The New York Botanical
Garden, U.S.A.
29. ULLOA, C., AND P. JORGENSEN. 1995. Árboles y arbustos de los Andes de Ecuador.
Abya-yala. Quito, Ecuador.
30. VAN DEN EYDEN, V., E. CUEVA, AND O. CABRERA. 1998. Plantas comestibles del sur
de Ecuador. Abya-Yala, Quito, Ecuador, 221 pp.
31. VAN DER PIJL, L. 1969. Principles of dispersal in higher plants. Springer-Verlag
Berlin Heidelberg, New York.
32. VASCONCELOS, S.F.D. 2006. Fenologia e síndromes de dispersao de espécies
arbustivas e arbóreas acorrentes em uma área de carrasco no planalto da Ibiapaba,
Ceará. Msc Dissertation, Universidade Federal de Pernambuco, Brasil, 61 pp.
33. VÁSQUEZ, R. 1997. Flórula de las reservas biológicas de Iquitos, Perú: Missouri
B.G. Vol.63.
34. VILCAPOMA SEGOVIA, G. 2007. Frutos silvestres (solanáceas) de la cuenca del Río
Chillón, provincia de Canta, Lima-Perú. Ecol. apl. 6: 23-37
35. VOZZO, J. A. (ED.). 2002. Tropical Tree seed manual. United States Department of
Agriculture, Forest Service. Washinton DC, U.S.A.
36. WIGGINS, I. L., AND D. M. PORTER. 1971. Flora of the Galapagos Islands. Stanford
University Press, Stanford, California, U.S.A.
78
WEB REFERENCES:
37. FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS (FAO). 1997.
Especies arbóreas y arbustivas para las zonas áridas y semiáridas de América
Latina. Red Latinoamericana de Cooperación Técnica en Sistemas Agroforestales.
Santiago, Chile. Accesss to: January 30, 2008. Available from:
http://www.fao.org/regional/lamerica/redes/sisag/arboles/Bra-capa.htm.
38. GONZÁLEZ, J. 2007. Flora digital de Palo Verde. Organización para Estudios
Tropicales (OTS). Access to: January 9, 2009. Available from:
http://www2.ots.ac.cr/biots/florulapv/families_desc/Euphorbiaceae.pdf.
39. KNAPP, S. 2004. Description based on taxon concept. Natural History Museum.
Accesss to: January 9, 2009. Available from: http://www.nhm.ac.uk/research-
curation/research/projects/solanaceaesource/taxonomy/description-
detail.jsp?spnumber=1863
40. MACBRIDE, F. 1943. Flora of Peru. Field Museum of Natural History. Vol. XIII, Part
3, Num. 1, Chicago, U.S.A. Access to: June 17, 2008. Available from:
http://www.archive.org/stream/floraofperufimacb/floraofperufimacb_djvu.txt
41. SÁNCHEZ, J.M. Guía de plantas ornamentales. Access to: november 13, 2008.
Available from: http://www.arbolesornamentales.com/generos.htm
42. THE FIELD MUSEUM. Neotropical Herbarium Specimens. Access to: october 17,
2008. Available from: http://fm1.fieldmuseum.org/vrrc/index.php
79
HERBARIA RECORDS:
43. Herbario Nacional del Ecuador
44. Herbario Loja
45. Herbario UTPL
Obs. Pers. (Observaciones personales)
80
CAPÍTULO 2
DOES SPATIAL HETEROGENEITY BLUR THE SIGNATURE OF DISPERSAL
SYNDROMES ON SPATIAL PATTERNS OF WOODY SPECIES? A TEST FOR A
TROPICAL DRY FOREST
Andrea Jara-Guerrero1, 4
; Marcelino De la Cruz2, 3
; Carlos I. Espinosa1; Marcos Méndez
3;
Adrián Escudero3
1Departamento de Ciencias Naturales, Universidad Técnica Particular de Loja, CP.: 11-01-
608, Loja, Ecuador.
2Departamento de Biología Vegetal, E.U.I.T. Universidad Politécnica de Madrid, 28040
Madrid, Spain.
3Área de Biodiversidad y Conservación, Universidad Rey Juan Carlos, E-28933 Móstoles,
Spain.
Manuscrito remitido a:
OIKOS
81
ABSTRACT
Spatial patterns of adult plants are a consequence of several ecological processes related to
seed dispersal and recruitment. Dispersal limitation, mediated by dispersal syndrome, is
considered a key factor in the formation of adult plant spatial patterns. Although this initial
pattern determined by dispersal has been large studied, the subsequently modification by
the effect of additional ecological factors, such as habitat heterogeneity is less understood.
We explored the relative importance of dispersal syndrome and spatial heterogeneity on the
realization of spatial patterns of adult trees in an Ecuadorian tropical dry forest. The spatial
distribution of 28 species was modeled with four different spatial point processes each:
homogeneous Poisson (HPP), inhomogeneous Poisson (IPP), homogeneous Poisson cluster
(HPCP), and inhomogeneous Poisson cluster process (IPCP). These models allowed us to
discern between effects of random processes, habitat heterogeneity, limited dispersal, and
joint effects of habitat heterogeneity and limited dispersal. We employed Akaike’s
information criterion (AIC) to select the model which best fitted the spatial pattern of each
species. The best model of each species was used to analyze differences in cluster size and
degree of aggregation, between dispersal syndromes. Aggregate patterns prevailed in the
studied community, however, the effect of limited dispersal, as has been proposed for
rainforests, was not enough to explain aggregation patterns in this tropical dry forest. IPCP
yielded the best fit for the spatial distribution of 50% of the species in the studied forest and
was the prevalent model for zoochorous, anemochorous and autochorous dispersal
syndromes. The effect of spatial heterogeneity was prevalent in the distribution of most
species in this dry tropical forest (75 % species showed inhomogeneous patterns). In the
three dispersal syndromes, spatial aggregation was stronger at shorter distances, and
82
decreased from autochorous, through zoochorous, to anemochorous species. Our approach,
selecting the best model for each species, can give new insights on the process that
determine coexistence at fine and middle spatial scales in these megadiverse forests.
INTRODUCTION
Seed dispersal is a critical life process determining not only persistence but also the spatial
structure of populations and communities (Wang and Smith 2002, Cousens et al. 2008).
Initial dispersal pattern sets the scene for subsequent ecological processes, such as
depredation, germination, growth modulated by plant-plant interactions and herbivory, that
lead to the realized distribution of seedlings and adults (Schupp and Fuentes 1995, Nathan
and Muller-Landau 2000, Jansen et al. 2008). Several recent studies have advanced our
understanding about how seed dispersal processes are affecting spatial patterns of plants
(Hubbell 1979, Condit et al. 2000, Schupp and Fuentes 1995, Nathan and Muller-Landau
2000, Seidler and Plotkin 2006), and determined that dispersal syndromes are important
attributes in the formation of those patterns (Hubbell 1979, Condit et al. 2000, Seidler and
Plotkin 2006).
Other abiotic and biotic processes could distort the initial dispersal pattern, such as
environmental heterogeneity, the existence of facilitative interactions on early life stages
(Callaway and Walker 1997), or perch effects (Herrera et al. 1994, Schupp and Fuentes
1995, Gómez 2003), meaning a higher density of seeds in habitats preferred by dispersal
agents, beneath perches, along travel routes, etc. (Schupp et al. 2002). For instance,
negative density-dependent processes (Janzen 1970, Hirsch et al. 2012) lead to higher
survival and germination of those seeds falling further away from the parent plant.
83
Nevertheless, even after those processes, a concordance usually remains between the spatial
patterns of seed arrival and those of seedlings (emergence) and adults (recruitment)
(Schupp and Fuentes 1995, Seidler and Plotkin 2006). In particular, in tropical rain and dry
forests the spatial aggregation of conspecific adults at different spatial scales has been
related to limited seed dispersal (Hubbell 1979, Condit et al. 2000, Nathan and Muller-
Landau 2000, Hardesty and Parker 2002, Seidler and Plotkin 2006, Wiegand et al. 2009). If
this relationship between dispersal and realized recruitment were universal, it could be
thought that, ahead of other ecological processes, dispersal syndrome would be the main
determinant of the adult realized distribution at fine spatial scales. Hubbell (1979)
pioneered this idea by showing such a relationship between dispersal syndromes and the
spatial pattern of adult tree species in tropical dry forests, and this evidence contributed to
develop his neutral theory (Hubbell 2001). More recently, Seidler and Plotkin (2006) also
found a strong correlation among seed dispersal syndromes and spatial patterns of trees in
two tropical rainforests. They showed that the adults of zoochorous or anemochorous
species, capable of relatively long-distance dispersal, appeared in larger clusters than those
of autochorous species with more limited dispersal.
Although there is an almost general consensus that dispersal is critical for explaining the
realized distribution of adults (Hubbell 1979, Condit et al. 2000, Seidler and Plotkin 2006,
Wiegand et al. 2009, Jara-Guerrero et al. 2011), a number of mechanisms can blur such
relationships (Schupp and Fuentes 1995, Seidler and Plotkin 2006). For instance,
landscapes are typically heterogeneous, with continuous variation in topography, aspect or
soil properties, which can influence not only seed shadows but germination and recruitment
of dispersed seeds (Grubb 1977, Schupp and Fuentes 1995). This environmental
84
heterogeneity could lead to a mismatch between the spatial patterns of dispersed seeds and
that of recruited seedlings or adults (Comita et al. 2007, Cousens et al. 2008). Furthermore,
even if dispersal were not limited, environmental heterogeneity could generate virtual
aggregation (Schiffers et al. 2008), i.e., spatial patterns indistinguishable from those
resulting from limited dispersal (Wiegand et al. 2007a). This fact raises doubts on the
inferences that can be extracted, for example, from the relationships between the spatial
pattern of adult plants and dispersal syndromes (Seidler and Plotkin 2006).
We explored the relative importance of both dispersal syndrome and spatial heterogeneity
on the realization of the spatial pattern of adult trees in an Ecuadorian tropical dry forest.
Our working hypothesis is that the realized pattern is the result of a two-step process where
dispersal syndrome (Hubbell 1979, Seidler and Plotkin 2006) first sets the initial spatial
template and later ecological processes linked to environmental heterogeneity (Webb and
Peart 2000) blur the original pattern of seed dispersal and determine the realized pattern of
adult trees. As we work in a dry tropical forest, we expect that the tight relationship
between dispersal syndrome and spatial pattern of adult trees found in tropical rainforests
(Seidler and Plotkin 2006) would be less evident because of the stronger filter exerted by
harsher environmental conditions. We used spatial point pattern analysis, and a set of a
priori hypotheses to discern those processes responsible for the realized pattern of trees
(McIntire and Fajardo 2009). Specifically, we fitted four kinds of spatial point process
models that can be easily interpreted in terms of biological processes and environmental
conditions (Shen et al. 2009, Lin et al. 2011, Sánchez-Pescador et al. 2014). The processes
tested consider purely random processes (homogeneous Poisson process), random
processes in heterogeneous habitat (inhomogeneous Poisson process), limited dispersal
85
(Poisson cluster process), and the joint effect of habitat heterogeneity and limited dispersal
(inhomogeneous Poisson cluster processes). Thus, if the final distribution of adults is the
consequence of dispersal alone we would expect; i) zoochorous plants to show patterns
compatible with a homogeneous Poisson process, because seeds are carried by very mobile
dispersers and can reach any point in the territory; ii) autochorous plants to show patterns
compatible with a homogeneous Poisson cluster process, in which adults are sources for
other recruits in a nested process (Cousens et al. 2008), because seeds are dispersed in the
vicinity of mature plants, and iii) anemochorous species, depending on dispersal efficiency
(i.e. plant architecture, aerodynamic properties of diaspores (Muller-Landau et al. 2008), to
show patterns compatible with either homogeneous Poisson in more efficiently dispersed
species, or homogeneous Poisson cluster processes (but with larger clusters than
autochorous species) in those of more limited dispersal. On the contrary, if environmental
heterogeneity matters, we would expect most adult patterns to be adequately described by
inhomogeneous processes as a consequence of the environmental filter exerted on the seed
pattern after primary dispersal (i.e., inhomogeneous Poisson or inhomogeneous Poisson
cluster, depending on the relative importance of dispersal limitation).
MATERIALS AND METHODS
Study area—The study was conducted in the Arenillas Ecological Reserve (REA, from its
Spanish name), located in southwestern Ecuador (03° 34’ 15.44’’S; 80° 08’ 46.15’’E, 30 m.
a.s.l.). This area belongs to the Tumbesian biogeographic region, one of the most important
areas of endemism in the world, but at the same time, one of the most threatened by
increasing agriculture and livestock activities, which are leading to a decrease in forest
86
remnants in this region (Best and Kessler 1995). Climate is characterized by a rainy season
extending from January to May and a dry season extending from June to December. Annual
mean precipitation at REA is 667 mm, and annual mean temperature is 25°C (Huaquillas
weather station, with 45 years record period). Vegetation corresponds to a seasonal tropical
dry forest. The REA is inhabited by a high diversity of vertebrates (Best and Kessler 1995),
some of which are potential seed dispersers, such as birds, bats and deer. Although the role
of these species as seed dispersers has been poorly studied in the Tumbesian region, some
species are defined as frugivorous; for example, at least three bat species (i.e. Artibeus
fraterculus, Carollia brevicauda, Sturnira lilium) (Tirira et al. 2007) and six bird species
reported for the REA are frugivorous, and at least other ten bird species are omnivorous
(Tinoco 2009). The two deer species reported for the REA have been also reported feeding
fruits (Tirira et al. 2007).
Data collection—Between January 2010 and September 2011, we located a 9-ha (300 x
300 m) permanent forest plot in a well conserved area of this forest. Topography in the plot
is relatively flat. Altitud varied between 30.9 to 38.6 m a.s.l., with an average of 35.4 m.
The plot is traversed by a seasonal stream, with water only for very short periods during
intense rainfalls. All trees and shrubs exceeding 5 cm diameter at breast height were
mapped, measured, and identified to species. Individuals with multiple stems were counted
as a single individual (Condit et al. 2000). Coordinates (x, y) and elevation (z) were
measured using a computerized total surveying station (Leica TS02-5 Power) with a
resolution of 5 cm.
87
Score of dispersal syndromes—We collected diaspores for all species in the REA forest
during 2010 and 2011. Following Van der Pijl (1969), species were classified into three
major dispersal syndromes: zoochory, anemochory and autochory. Zoochory was defined
by the presence of fleshy and edible structures that promote the ingestion and transport of
seeds by animals. Species having diaspores with membranous wings, or other aerodynamic
structures that affect the rate of fall, were scored as anemochorous. Autochory group
included species with diaspores propelled explosively, passive ballists (triggered by passing
animals, wind or raindrops), and those so-called barochorous (released and falling to the
ground).
Statistical models—We employed Ripley's K function to characterize the spatial pattern of
each species (Ripley 1976, Illian et al. 2008). For a homogeneous point pattern where λ is
pattern intensity (i.e., density), λK(r) is the expected number of points within a circle of
radius r around an arbitrary point.
We compared the spatial pattern of each species (as characterized by the K-function) and
the expected pattern for four different spatial point processes (Shen 2009, Lin et al. 2011;
Table 1): (1) a homogeneous Poisson process (HPP), which assumes that the spatial
location of points (i.e., standing individuals) is independent from each other and their
intensity is constant in the whole area: it is the simplest process and considers the pattern as
the result of purely random processes (Shen et al. 2009, Lin et al. 2011); (2) an
inhomogeneous Poisson process (IPP), which still assumes independence between points
but recognizes the existence of variations in intensity throughout the study area; it is usually
assumed that this variation is related to habitat heterogeneity (Wiegand et al. 2007b); (3) a
88
homogeneous Poisson cluster process (HPCP) assumes constant intensity and considers
aggregation of points as a result of limited dispersal (Shen et al. 2009); it assumes that the
aggregated pattern is generated in a two-step process: first, a HPP of cluster centers (i.e.,
parent points) with intensity is created and, afterwards, each parent produces a number of
offspring according to a Poisson distribution which are located around the parents
according to a radially symmetric Gaussian distribution with mean zero and standard
deviation σ (Seidler and Plotkin 2006, Shen et al. 2009, Lin et al. 2011); (4) an
inhomogeneous Poisson cluster process (IPCP), which assumes that the spatial pattern is
created by dispersal limitation (i.e., like a HPCP), but with intensity varying in response to
habitat heterogeneity (Wiegand and Moloney 2013). We employed Akaike’s information
criterion (AIC) based on the sum of residuals and the number of parameters in different
models (Webster and McBratney 1989, Shen et al. 2009) to select the model which best
fitted the spatial pattern of each species. Mathematical details about the fitted models are
provided in Supplementary Material Appendix 1.
We tested differences in the frequency of spatial patterns (HPP, IPP, HPCP and IPCP)
among the three dispersal syndromes by means of an extension of the Fisher’s exact test for
larger than 2 x 2 tables (function fisher.test in R package stats). The p value for this
extension was obtained by Monte Carlo simulation using the simulate.p.value option in the
package "stats" of R.
Additionally, for those species which where best described by an HPCP or IPCP, we tested
the existence of differences in the parameter σ among syndromes. This parameter is related
to the average cluster size of each species (Seidler and Plotkin 2006). For this, we used
89
Kruskal-Wallis tests. All statistical analyses were carried out with R (R Core Team 2013).
Model fitting and K functions were computed using the R packages spatstat (Baddeley and
Turner 2005) and ecespa (De la Cruz 2008).
RESULTS
In the period from July 2010 to November 2011 a total of 5.296 individuals with DBH > 5
cm were tagged and georeferenced in the 9 ha plot. These individuals accounted for a total
of 37 species but we only considered 28 species with more than 10 individuals, belonging
to 19 families. The most common families were Mimosaceae (4 species), Caesalpiniaceae
and Fabaceae (3 species respectively). Zoochory was the dominant syndrome (50%, 14
species), followed by autochory (36%, 10 species) and anemochory (14%, 4 species) (Table
1). Most individuals in the plot belonged to zoochorous species (43.7%, 2.300 individuals),
while autochory represented 30.8% (1.621 individuals), and anemochory 25.5% (1.345
individuals) of the total.
Table 1. Percentage of species (and number of species) in each dispersal syndrome described by
each of the four point pattern processes considered. HPP: homogeneous Poisson process; IPP:
inhomogeneous Poisson process; HPCP: homogeneous Poisson cluster process; IPCP:
inhomogeneous Poisson cluster process.
Syndrome HPP IPP HPCP IPCP
Zoochory 7% (1) 36% (5) 14% (2) 43% (6)
Anemochory 0 25% (1) 25% (1) 50% (2)
Autochory 0 10% (1) 30% (3) 60% (6)
Total 4% (1) 25% (7) 21% (6) 50% (14)
90
Spatial patterns and their relationship to dispersal syndrome—Inhomogeneous Poisson
cluster process yielded the best fit for the spatial distribution of 50% of the species in the
studied forest (Table 1). Inhomogeneous Poisson process was the best model for 25% of
species, while other 21% of species was best fitted by homogeneous Poisson cluster
process (Table 1). Only the spatial distribution of one species, the zoochorous shrub
Vasconcellea parviflora, was best fitted by a homogeneous Poisson process (Table 1).
There was no significant association between dispersal syndrome and the process that
yielded the best fit (4 x 3 Fisher’s exact test, p = 0.739). Inhomogeneous Poisson cluster
process was the prevalent model for the three dispersal syndromes (Table 1).
Spatial signature of dispersal syndromes—The K(r) functions for the three dispersal
syndromes showed clear spatial aggregation within the range of the studied distances,
between 1 and 75 m (Fig. 1). Aggregation was stronger at short scales and decreased with
distance. Aggregation patterns decreased from autochorous, through zoochorous, to
anemochorous species (Fig. 1), but aggregation varied with the scale considered. At short
scales (1-30 m) zoochorous and autochorous were more aggregated than anemochorous,
and at medium scales (31-75 m) only autochorous species showed aggregation, while
zoochorous and anemochorous spatial patterns did not differ from that of a random process
(Figure 1).
91
Figure 1. The function K(r) evaluated at a range of distances r for tree species in the three dispersal
syndromes. Curves represent the average K(r) function value + standard error for each considered r
distance. For a clearer interpretation, we show K(r)/πr2. Values > 1 indicate spatial aggregation.
Grey lines represent confidence envelopes for a homogeneous Poisson pattern.
We failed to detect a significant relationship between spatial cluster size and dispersal
syndromes for the 28 studied species (Kruskal-Wallis, 2
= 2.674, df = 2, p = 0.263),
although a tendency to increase the size of the cluster from autochory to zoochory was
evident (Fig. 2).
92
autochory anemochory zoochory
5
10
15
20
clu
ste
r p
ara
me
ter
(m
)
Figure 2. Relationship between dispersal syndrome and spatial aggregation of 20 woody species
fitted by homogeneous or inhomogeneous Poisson cluster models. Figure shows the mean spatial
cluster size (σ) obtained from the best fitted model to each species. Thick lines indicate the median
value of cluster size for each dispersal syndrome (anemochory, n = 3; autochory, n = 9; zoochory, n
= 8).
DISCUSSION
Aggregated patterns are prevalent among species—Our results confirmed the
prevalence of aggregated patterns among tree and shrub species at the dry tropical REA
forest, however, the effect of limited dispersal in the formation of those patterns, as has
been proposed for rainforests (Plotkin et al. 2002, Seidler and Plotkin 2006) was not
enough to explain the aggregation patterns in our forest. Our results, obtained with a new
methodological approach which allow discerning between "pure" and virtually aggregated
93
patterns (Wiegand and Moloney 2004), suggest that environmental heterogeneity exerted
an additional (or was the only) effect limiting the distribution of most species in this forest
(75 % species showed inhomogeneous patterns). This effect was prevalent irrespective of
their dispersal syndrome, and even Senna mollissima, an autochorous species which should
show a clustered pattern, was in fact best described by an IPP.
The joint effects of habitat heterogeneity and dispersal limitation have been reported as
dominant also for other tropical (Comita et al. 2007, Wiegand et al. 2009) and subtropical
forests (Shen et al. 2009, Lin et al. 2011), suggesting that both are critical factors in
structuring a wide array of megadiverse forest communities. Shen et al. (2009) and Lin et
al. (2011) proposed that these two factors act sequentially. First, seed dispersal generates
the basal template pattern on which seedlings will develop. Afterwards, this pattern is
altered by the effects of habitat heterogeneity on survival rates and growth of seedlings. We
propose that the effects of environmental heterogeneity on spatial patterns may be
especially important in tropical dry forests, where water availability has been shown to
strongly affect the performance at different life stages of many trees (Espinosa et al. 2010,
2012). As seen here, heterogeneity effects could modify the original dispersal pattern in
opposite directions: either diluting the (theoretically) clustered pattern of autochorous
species (as in the case of Senna mollissima) to generate virtual aggregation, or restricting
the (theoretically) wider dispersal of zoochorous and anemochorous species.
Spatial patterns and their relationship to dispersal syndrome—The observed spatial
patterns partially matched our expectations in relation to each dispersal syndrome, and
reveal the great variation in spatial patterns, even among species with the same dispersal
94
syndrome (Table 1). Nine out of ten autochorous species showed aggregated spatial
patterns related to either limited dispersal or joint effects of limited dispersal and
environmental heterogeneity. This was consistent with our hypothesis about autochorous
species in which adult individuals are the sources for new recruits in a nested process. In
addition, 43 % of zoochorous species did not show or showed only virtual aggregation
correctly described by IPP's, which is compatible with the existence of unlimited dispersal.
Nevertheless, we found some departure from the expected results, especially in the case of
zoochorous species. A large proportion of them (57%) showed spatial patterns compatible
with limited dispersal (HPCP or IPCP). This could be explained by effects of other
processes not explicitly considered here, such as perch effects mediated by non-random
movement of dispersers (Herrera et al. 1994, Schupp and Fuentes 1995). Previous studies
have suggested perch effect as a cause of aggregated spatial patterns of some particular
zoochorous species (Godoy and Jordano 2001, Muller-Landau et al. 2008). Moreover, it has
been suggested that, in many cases, movements of dispersers occur over short distances
from the parent plant (Godoy and Jordano 2001, Muller-Landau et al. 2008), which could
explain this discrepancy in the results for some zoochorous species. Additionally, in
seasonal ecosystems like our study area, dispersal distances of zoochorous species can be
influenced by the fruit time. Dispersal distances, and quantity of dispersed seeds, are
expected to be larger in species fruiting in times of relative fruit scarcity, because of
dispersers tend to move greater distances searching for food (Muller-Landau et al. 2008). In
the study area, a large proportion of zoochorous species overlapping in fruiting time,
restricted to the short period of rains (Jara-Guerrero, pers. obs.).
95
We also found significant deviations from expected patterns in the case of anemochorous
species. The two species of Tabebuia, and Eriotheca ruizii showed patterns compatible
with limited dispersal (IPCP and HPCP, respectively) while Cochlospermum vitifolium
showed a pattern compatible with virtual aggregation (IPP). These patterns seem
contradictory with dispersal potential based on aerodynamic considerations (Augspurger
1986) because winged propagules of Tabebuia are expected to show higher dispersal
potential than floater propagules (C. vitifolium and E. ruizii). Thus, pure aerodynamic
considerations seem to be poor predictors of spatial patterns for these anemochorous
species. Instead, there has been proposed additional effects of other plant attributes in the
dispersal capacity, such as plant height, seed mass (Muller-Landau et al. 2008, Thomson et
al. 2011) and wind velocity during fruit time (Muller-Landau et al. 2008). Dispersal
phenology could provide further insights into the realized patterns of Tabebuia trees, which
are expected to have low dispersal efficiency because they fruit early in the rainy season
(A. Jara-Guerrero, pers. obs.), when wind circulation is lower. In the case of E. ruizii,
limited dispersal may be due to the fact that seeds are released in groups, which reduces
propagule aerodynamic performance (Linares-Palomino 2005).
Signature of dispersal syndromes in spatial aggregation patterns—All the species
showed spatial aggregation, with an increase of this spatial signal at short distances and
independently of the dispersal syndrome type (Fig. 1). According to our expectations,
autochorous species showed the greater aggregation, which is consistent with previous
studies (Condit et al. 2000, Seidler and Plotkin 2006). By contrast, the lower aggregation
for anemochory than zoochory did not match our expectations. This result could be
probably a spurious effect due to the small number of anemochorous species in our sample
96
(4 species); however, similar results have been obtained by Muller-Landau et al. (2008) for
the lowland moist tropical forest of BCI, with anemochorous species showing less
clumping that zoochorous species. In this context, this finding support the proposal that
other attributes of plant life history, as discussed below, can be interacting in the spatial
patterns formation (Muller-Landau et al. 2008, Thomson et al. 2011).
The variation of cluster sizes among dispersal syndromes, with zoochorous species
showing larger cluster sizes than anemochorous and autochorous species (Fig. 2) is
consistent with previous studies (Seidler and Plotkin 2006). However we did not detect
statistically significant differences among syndromes. This result may be explained by the
large variation within each dispersal syndrome. For instance, the large cluster size variation
in zoochorous species could be explained by the different behavior of dispersers (Muller-
Landau et al. 2008). According to Howe (1989), zoochory may result in two contrasting
seed deposition patterns depending on disperser. Birds and bats deposit single seeds,
resulting in isolated recruits, while large frugivores, such as mammals, usually defecate
masses of seeds and may cause aggregated spatial patterns (Howe 1989, Wiegand et al.
2009). In fact, Seidler and Plokin (2006) distinguished several types of animal dispersed
species in tropical forests and found significant differences among them. In anemochorous
species, dispersal distance depends on the rate of fall, given by the interaction between
diaspore morphology (which affects aerodynamics) and the structure of the patch (which
affects air currents) (Schupp and Fuentes 1995, Muller-Landau et al. 2008). Thus, although
some species can disperse their seeds far away from the parent plant, other species can only
attain a few meters. In general, autochorous species had the smaller cluster sizes, which is
consistent with their limited dispersal, and as has also been reported in other studies
97
(Seidler and Plotkin 2006, Muller-Landau et al. 2008). However, we found the largest
variation in cluster size within this group, with some species such as Pithecellobium
excelsum and Mimosa acantholoba showing cluster sizes even larger than all the
zoochorous species. These outlier species could be caused by of accidental dispersal events,
or secondary dispersal events. For example, in M. acantholoba, although it lacks
specialized dispersal mechanisms, mature pods may become stiffly papery and often
densely hispid-setose, which may promote dispersal away from the parent plant by wind or
ectozoochory respectively (Barneby 1991). It has also been suggested that their lustrous
seeds might be taken up by birds and dispersed undamaged (Barneby 1991). In fact, M.
acantholoba is considered a pioneer species (Lebrija-Trejos and Bongers 2008) which
suggests that these ´accidental´ or secondary events should be usual in the species. In the
case of P. excelsum, seeds do not fall from the pod immediately after dehiscence but
instead remain suspended from a colored spongy aril, attractive to birds (Barneby and
Grimes 1997), which might disperse them away. On the other hand, spatial scale considered
here is the same for all species, however, as there is proposed by Thomson et al (2011),
maximum dispersal distances are dependent of individual species, thus, spatial scale should
influence the detected patterns.
In summary, predictions about spatial patters of tree species and relations to dispersal
syndrome can be improved by selecting the best model for each species from a set of
models with well known biological links, instead of using a fixed model for all species. Our
study found that spatial heterogeneity, and not only dispersal limitation, affected species
realized spatial patterns in the studied dry tropical forest. Thus, ignoring habitat
heterogeneity could bias the analysis of relationships between dispersal syndrome and
98
species patterns. Our findings also support the hypothesis that local spatial patterns are not
independent of the biological attributes of plant species (Seidler and Plotkin 2006, Shen et
al. 2009, Lin et al. 2011). Differences in seed dispersal abilities, which are related to
morphological attributes but also to phenological differences within a dispersal syndrome,
may contribute to structuring and maintain the assemblage of species in the studied tropical
dry forest. Therefore, the use of models reproducing processes with biologically sound
properties can be valuable in the interpretation of realized individual patterns and the
structure of a community as a whole and can give new insights on the process that
determine coexistence at fine and medium spatial scales. Our work also enlightens the
academic debate about the prevalence of different mechanisms in the formation of realized
assemblages in tropical dry forests (Hubbell 1979, Leibold 1995, López-Martínez et al.
2013). Thus, although we hypothesized that dispersal would not suffice for explaining
spatial patterns and that stressful conditions should give prominence to environmental
heterogeneity, we found that spatial patterns for 21% of species studied were compatible
with random mechanisms associated to dispersal around mother sources (i.e. neutrality
sensu Hubbell 2001).
ACKNOWLEDGEMENTS
This study was partially supported by projects CGL2009-13190-C03-02 (ISLAS
ESPACIO), 293 CGL2012-38427 (MOUNTAINS), REMEDINAL3,
PROY_IECOLOGIA_ 0018 and PROY_IECOLOGIA_0035 funded by Universidad
Técnica Particular de Loja. The authors wish to thank Ministerio del Ambiente del Ecuador
99
and Ministerio de Defensa del Ecuador for facilities and operational support during field
work.
REFERENCES
Augspurger, C. 1986. Morphology and Dispersal Potential of Wind-Dispersed Diaspores of
Neotropical Trees. - Am. J. Bot. 73: 353-363.
Baddeley, A. and Turner, R. 2005. Spatstat: an R package for analyzing spatial point
patterns. – J. Stat. Softw. 12: 1–42.
Barneby, R.C. 1991. Sensitivae censitae: a description of the genus Mimosa Linnaeus
(Mimosaceae) in the New World. - Mem. N. Y. Bot. Gard., v. 65.
Barneby, R.C. and Grimes, J. W. 1997. Silk tree, Guanacaste, Monkey's earring: a generic
system for the synandrous Mimosaceae of the Americas. - Mem. N. Y. Bot. Gard. v.
74, part II.
Best, B. J., and Kessler, M. 1995. Biodiversity and conservation in Tumbesian Ecuador and
Peru. - BirdLife International.
Callaway, R., and Walker, L. 1997. Competition and facilitation: A synthetic approach to
interactions in plant communities. - Ecology 78: 1958-1965.
Comita, L. S. et al. 2007. Developmental changes in habitat associations of tropical trees. –
J. Ecol. 95: 482–492.
Condit, R. et al. 2000. Spatial patterns in the distribution of tropical tree species. - Science,
288: 1414-1418.
100
Cousens, R. et al. 2008. Dispersal in plants. A population perspective. - Oxford University
Press.
De la Cruz, M. 2008. Métodos para analizar datos puntuales. - In: Maestre, F. T., Escudero,
A. y Bonet, A. (eds.), Introducción al análisis espacial de datos en ecología y
ciencias ambientales: métodos y aplicaciones. Asociación Española de Ecología
Terrestre, Universidad Rey Juan Carlos y Caja de Ahorros del Mediterráneo, pp. 76-
127.
Espinosa, C.I. et al. 2011. What factors affect diversity and species composition of
endangered tumbesian dry forests in Southern Ecuador? - Biotropica 4: 15-22.
Espinosa, C.I., et al. 2012. Bosques tropicales secos de la región Pacífico Ecuatorial:
diversidad, estructura, funcionamiento e implicaciones para la conservación. -
Ecosistemas 21: 167-179.
Godoy, J. and Jordano, P. 2001. Seed dispersal by animals: exact identification of source
trees with endocarp DNA microsatellites. – Mol. Ecol. 10: 2275-2283.
Gómez, J.M. 2003. Spatial patterns in long-distance dispersal of Quercus ilex acorns by
jays in a heterogeneous landscape. - Ecography 26: 573–584.
Grubb, P. J. 1977. The maintenance of species-richness in plant communities: The
importance of the regeneration niche. – Biol. Rev. 52: 107-145.
Hardesty, B. and Parker, T. 2002. Community seed rain patterns and a comparison to adult
community structure in a West African tropical forest. - Plant Ecol. 164: 49–64.
101
Herrera, C.M. et al. 1994. Recruitment of a mast-fruiting, bird-dispersed tree – bridging
frugivore activity and seedling establishment. Ecol. Monogr. 64: 315–344.
Hirsch, B. et al. 2012. Quantifying seed dispersal kernels from truncated seed-tracking data.
- Methods Ecol. Evol. 3: 595–602.
Howe, H. F. 1989. Scatter and clump-dispersal and seedling demography: hypothesis and
implications. - Oecologia 79: 417–426.
Hubbell, S. 1979. Tree dispersion, abundance, and diversity in a tropical dry forest. -
Science 203: 1299-1309.
Hubbell, S. 2001. The unified neutral theory of biodiversity and biogeography. - Princeton
University Press.
Illian, J. et al. 2008. Statistical analysis and modelling of spatial point patterns. - Wiley.
Jansen, P. et al. 2008. Is farther seed dispersal better? Spatial patterns of offspring mortality
in three rainforest tree species with different dispersal abilities. - Ecography 31: 43-
52.
Janzen, D.H. 1970. Herbivores and the number of tree species in tropical forests. – Am.
Nat. 104: 501–527.
Jara-Guerrero, A. et al. 2011. Seed dispersal spectrum of woody species in South
Ecuadorian dry forests: Environmental correlates and the effect of considering
species abundance. - Biotropica 43: 722–730.
102
Lebrija-Trejos, E. and Bongers, F. 2008. Successional change and resilience of a very dry
tropical deciduous forest following shifting agriculture. - Biotropica 40: 422–431.
Leibold, M. A. 1995. The niche concept revisited: mechanistic models and community
context. - Ecology 76: 1371–1382.
Lin, Y.C. et al. 2011. Point patterns of tree distribution determined by habitat heterogeneity
and dispersal limitation. - Oecologia 165: 175–184.
Linares-Palomino, R. 2005. Spatial distribution patterns of trees in a seasonally dry forest
in the Cerros de Amotape National Park, northwestern Peru. - Rev. Peru. Biol. 12:
317-326.
López-Martínez, J. O. et al. 2013. β-Diversity of functional groups of woody plants in a
tropical dry forest in Yucatan. - PLoS ONE 8(9): e73660.
Mcintire, E. and Fajardo, A. 2009. Beyond description: the active and effective way to infer
processes from spatial patterns. - Ecology 90: 46–56.
Muller-Landau, H. C. et al. 2008. Interspecific variation in primary seed dispersal in a
tropical forest. – J. Ecol. 96: 653-667.
Nathan, R. and Muller-Landau, H. C. 2000. Spatial patterns of seed dispersal, their
determinants and consequences for recruitment. - Trends Ecol. Evol. 15: 278 – 285.
Pescador, D. S. et al. 2014. Maintaining distances with the engineer: patterns of coexistence
in plant communities beyond the patch-bare dichotomy. - New Phytol. 204: 140–
148.
103
Plotkin, J.B. et al. 2000. Species-area curves, spatial aggregation, and habitat specialization
in tropical forests. – J. Theor. Biol. 207: 81–99.
R Core Team. 2013. R: A language and environment for statistical computing. - R
Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-
project.org/.
Ripley, B.D. 1976. The second-order analysis of stationary point processes. - J. Appl.
Prob.13: 255-266.
Schiffers, K. et al. 2008. Dealing with virtual aggregation a new index for analyzing
heterogeneous point patterns. - Ecography 31: 545-555.
Schupp, E. and Fuentes, M. 1995. Spatial patterns of seed dispersal and the unification of
plant population ecology. - Écocience 2: 267-275.
Schupp, E., Milleron, T. and Russo, S. 2002. Dissemination limitation and the origin and
maintenance of species-rich tropical forests. In: Levey, D.J., Silva, W.R. and
Galetti, M. Seed dispersal and frugivory: ecology, evolution and conservation. CAB
International.
Seidler, T. and Plotkin, J. 2006. Seed dispersal and spatial pattern in tropical trees. - PLoS
Biol. 4: 2132-2137.
Shen, G. et al. 2009. Species–area relationships explained by the joint effects of dispersal
limitation and habitat heterogeneity. - Ecology 90: 3033-3041.
104
Tinoco, B. 2009. Estacionalidad de la comunidad de aves en un bosque deciduo tumbesino
en el sur occidente de Ecuador. Ornitología Neotropical 20: 157–170.
Tirira, D. 2007. Guía de campo de mamíferos del Ecuador. Ed. Murciélago Blanco.
Publicación especial de los Mamíferos del Ecuador 6.
Thomson, F.J., Moles, A.T., Auld, T.D. and Kingsford, R.T. Seed dispersal distance is
more strongly correlated with plant height than with seed mass. J. Ecol. 99: 1299-
1307.
Van der Pijl, L. 1969. Principles of dispersal in higher plants. - Springer-Verlag, Berlin.
Wang, B. and Smith, T. 2002. Closing the seed dispersal loop. - Trends Ecol. Evol. 17:
379-385.
Webb, C. and Peart, D. 2001. High seed dispersal rates in faunally intact tropical rain
forest: theoretical and conservation implications. – Ecol. Lett. 4: 491-499.
Webster, R. and Mcbratney, A.B. 1989. On the Akaike information criterion for choosing
models for variograms of soil properties. - J. Soil Sci. 40: 493–496.
Wiegand, T. and Moloney, K. A. 2004. Rings, circles, and null models for point pattern
analysis in ecology. - Oikos 104: 209–229.
Wiegand, T. et al. 2007a. Species associations in a heterogeneous Sri Lankan dipterocarp
forest. – Am. Nat. 170: E77-E95.
Wiegand, T. et al. 2007b. How individual species structure diversity in tropical forests. -
Proc. Natl. Acad. Sci. USA 104:19029–19033.
105
Wiegand, T. et al. 2009. Recruitment in tropical tree species: revealing complex spatial
patterns. – Am. Nat. 174: 4.
Wiegand, T. and Moloney, K. A. 2013. Handbook of spatial point-pattern analysis in
ecology. - Chapman and Hall/CRC.
106
MATERIAL SUPLEMENTARIO PARA EL CAPÍTULO 2.
Appendix 1. Mathematical details about the fitted models
Spatial analysis
To characterize and fit point process models, we used Ripley’s K-function (Ripley 1976,
Illian et al. 2008). Then, we compared (using AIC, see below) the observed K-function for
each of the studied species with the expected K function for each of the fitted models.
Homogeneous Poisson Process (HPP)
For a homogeneous point pattern where λ is pattern intensity (i.e., density), and λK(r) is the
expected number of points within a circle of radius λ around an arbitrary point, the K
function is estimated as:
)(1
)(1
2rdI
w
ArK
n
i ij
ij
ij
Where A and n are respectively the area and the number of points in the plot, wij is
an edge-correction factor, and I is the indicator function. Homogeneous Poisson processes
(HPP) have constant intensity throughout the study area and the distribution of points is
independent so the expected K(r) = πr2 (Illian et al. 2008: 80).
Inhomogeneous Poisson Process (IPP)
Inhomogeneous Poisson processes share with HPP the independence of points but their
intensity is not constant; instead intensity varies from place to place according to an
intensity function λ(u) which, in ecological studies, it is assumed to depend on
107
environmental heterogeneity (e.g., Wiegand et al. 2007, Getzin et al. 2008, Wiegand and
Moloney 2014). To characterize IPP's we employed the inhomogeneous K-function
(Baddeley et al. 2000). It is estimated as:
)()()(
1)(
1
rdIxx
w
ArK
n
i ij
ij
ji
ij
I
The expected value of the inhomogeneous K-function for an IPP is:
2)( rrK I
The intensity function can be estimated in different ways. We used kernel
smoothing (Baddeley 2010), which estimates the intensity at location u as
n
i
ixuueu1
)()()(
where κ(u) is an arbitrary kernel function and e(u) is an edge
correction term. We employed the Gaussian kernel as smoothing function.
Homogeneous Poisson cluster processes (PCP)
Poisson cluster processes generate non-independent (clustered) points in a two-step
process. First, an HPP of "parent" points is generated with intensity ρ. Then, each parent
point produces "offspring". The number of offspring per parent follows a Poisson
distribution, and their locations are independent and isotropically normally distributed
around the parent tree, with mean zero and standard deviation σ. The expected K function
for a PCP is:
)
4(
2
2
2
1)(
r
errK
108
Inhomogeneous Poisson cluster processes (IPCP)
Inhomogeneous Poisson cluster processes are an extension of the PCPs, where it is
assumed that the distribution of points, in addition to clustered, is inhomogeneous. Thus,
expectation for the K-function of an IPCP is the same than for PCP, but the intensity is
estimated by a spatially inhomogeneous function.
Both homogeneous and inhomogeneous K-functions were estimated using Ripley’s
isotropic edge correction (Ripley 1978). All the functions were estimated up to 75 m (i.e.,
rmax = 75 m), with steps of 1 m.
Estimation of intensity functions for IPP and IPCP models
In the case of IPP and IPCP, the intensity functions were estimated with spatstat function
density.ppp(). This function implements kernel smoothing of a point pattern based on a
Gaussian kernel. In ecological studies, choosing the bandwidth of the kernel for estimation
of an intensity surface is based on biological criteria, i.e., selecting a bandwidth larger than
the scale at which second order effects (i.e., spatial structures derived from individuals
interactions or short scale dispersal) show up (Wiegand et al. 2007, Getzin et al. 2008).
With this in mind, we estimated for each species a set of 13 different intensity surfaces,
using Gaussian kernels varying from σ =15 to σ=75 m in 5 m steps (these are equivalents to
"bandwidths" from 30 to 150 m; Bivand et al. 2008: 168).
Model fitting and selection
109
Models were fitted with the R packages spatstat (Baddeley & Turner 2005) and ecespa (De
la Cruz 2008). For each species, we fitted a total of 28 models (1 HPP + 13 IPP + 1 HPCP
+ 13 IPCP). To identify the most parsimonious model we used Akaike’s information
criterion (AIC) based on the sum of residuals and the number of parameters in each model
(Shen et al. 2009, Wang et al. 2013). For each species, we selected the model which
rendered the smaller AIC.
REFERENCES
Baddeley, A.J. et al. 2000. Non- and semiparametric estimation of interaction in
inhomogeneous point patterns. - Statistica Neerlandica 54: 329–350.
Baddeley, A. 2010. Analysing spatial point patterns in R. Workshop notes. Version 4.1. –
CSIRO online technical publication. URL: www.uwa.edu.au/resources/pf16h.html
Baddeley, A. and Turner, R. 2005. Spatstat: an R package for analyzing spatial point
patterns. – J. Stat. Softw. 12: 1–42.
Bivand, R.S. et al. 2008. Applied spatial data analysis with R. - Springer.
De la Cruz, M. 2008. Métodos para analizar datos puntuales. - In: Maestre, F. T., Escudero,
A. y Bonet, A. (eds.), Introducción al análisis espacial de datos en ecología y
ciencias ambientales: métodos y aplicaciones. Asociación Española de Ecología
Terrestre, Universidad Rey Juan Carlos y Caja de Ahorros del Mediterráneo, pp. 76-
127.
110
Getzin, G. et al. 2008. Heterogeneity influences spatial patterns and demographics in forest
stands. - J. Ecol. 96: 807-820.
Illian, J. et al. 2008. Statistical analysis and modelling of spatial point patterns. - Wiley.
Ripley, B.D. 1976. The second-order analysis of stationary point processes. - J. Appl.
Prob.13: 255-266.
Ripley, B.D., 1978. Spectral analysis and the analysis of pattern in plant communities. - J.
Ecol. 66: 965-981.
Shen, G. et al. 2009. Species–area relationships explained by the joint effects of dispersal
limitation and habitat heterogeneity. - Ecology 90: 3033-3041.
Wang, X. et al. 2013. Phylogenetic and functional diversity area relationships in two
temperate forests. - Ecography 36: 001-011
Wiegand, T. et al. 2007. How individual species structure diversity in tropical forests. -
Proc. Natl. Acad. Sci. USA 104: 19029-19033.
Wiegand, T. and Moloney, K. 2014. Handbook of Spatial Point-Pattern Analysis in
Ecology. - CRC Press.
111
112
CAPÍTULO 3
SEED RAIN AND SEED BANK CONTRIBUTION TO REGENERATION OF WOODY
SPECIES WITH DIFFERENT DISPERSAL SYNDROME IN AN ECUADORIAN
TROPICAL DRY FOREST
Andrea Jara-Guerrero1, Marcos Méndez
2, Carlos I. Espinosa
1, Marcelino De la Cruz
3
1Departamento de Ciencias Naturales, Universidad Técnica Particular de Loja, CP.: 11-01-
608, Loja, Ecuador.
2Área de Biodiversidad y Conservación, Universidad Rey Juan Carlos, E-28933 Móstoles,
Spain.
3Departamento de Biología Vegetal, E.U.I.T. Universidad Politécnica de Madrid, 28040
Madrid, Spain.
113
ABSTRACT
Long-distance seed dispersal and the ability to form seed banks are alternative strategies
to plant regeneration. Simultaneous analysis of both strategies can provide valuable
information about the extent to which seed rain and seed bank are determining the
vegetation composition. We explored whether temporal patterns of seed dispersal and / or
dispersal syndrome control the relationship between vegetation and seed compartments
(seed rain or seed bank) in a seasonally tropical dry forest at Southern Ecuador. Within a 9
ha plot we recorded stablished woody plants. In the central 4.8 ha we sampled seed rain
and seed bank from 265 points. We analyzed monthly and seasonal variation in seed
abundance in the seed rain, for all species and by dispersal syndrome. We also examined
similarity in species composition and abundance between the three compartments -
established plants, seed rain and seed bank-, for all species and for each dispersal
syndrome. Species with different dispersal syndrome showed differences in temporal
patterns of seed dispersal. This result can be explained as adjustment of different
dispersal syndromes to particular environmental conditions for an efficient dispersal.
Species richness of each dispersal syndrome did not show significant differences among
compartments –established plants, seed rain or seed bank–. By contrast, each dispersal
syndromes showed a significantly differedence in the distribution of abundance of
individuals among compartments. Consideration of dispersal syndromes revealed
significant differences in the match in species richness and seed abundance between
compartments. In particular, few zoochorous and anemochorous species formed seed
banks; however, it can be an effective supply of seeds for some species of plants, mainly
for autochorous species.
114
INTRODUCTION
Seed dispersal is a key process for the regeneration of plant communities (Howe and
Smallwood 1982), because it is the basis for the recruitment of new individuals (Schupp
and Fuentes 1995, Nathan and Muller-Landau 2000). Recruitment may occur inmediately
upon the arrival of seeds to the ground after the seed rain, or be delayed in time due to the
formation of seed banks (Uhl et al. 1981, Putz and Appanah 1987, Alvarez-Buylla and
Martínez Ramos 1990, Moles and Drake 1999). Different plant species rely to a different
extent on direct recruitment from the seed rain or permanence in the seed bank, as part of
its regeneration strategy (Brown and Venable 1986, Dungan et al., 2001).
In species relying directly on the seed rain for regeneration, dispersal events can be timed to
certain environmental conditions. In seasonal ecosystems, zoochorous species ripen their
fruits during the rainy season, when dispersers and water for immediate germination are
readily available (Howe and Smallwood 1982, Bullock 1995, Jara-Guerrero et al. 2011).
Autochorous and anemochorous species have dry fruits which usually ripen during the dry
season, when lower canopy density, and higher wind circulation allow greater dispersal
distances (Howe and Smallwood 1982, Bullock 1995, Jara-Guerrero et al. 2011).
Furthermore, seed dispersal in this season reduces predation risk, because insect and fungal
activity is higher in moist periods (Maza-Villalobos et al. 2010). These seasonal patterns
are highly predictable and have been documented in several studies of seasonally dry
tropical forests (sDTF) (Janzen 1967, Garwood 1983, Singh and Singh 1992, Bullock 1995,
Justiniano and Fredericksen 2000, Griz and Machado 2001, Jara-Guerrero et al. 2011).
115
Both long-distance dispersal and the ability to form seed banks are alternative strategies to
avoid risk, excessive density and competition (Venable and Brown 1988, Bakker et al.
1996). Thus, many species with potential long-distance dispersal (zoochorous and
anemochorous) do not form seed banks, unlike species with reduced dispersion (i.e.
autochorous species) (Bakker et al. 1996). Long-distance dispersal is an advantage when
conditions near to the mother plant are unfavorable and recruitment probabilities improve
farther away from the seed source. In spatially homogeneous and temporally predictable
environments, seed dispersal provides no advantage for the individual, and kin selection
can instead favor dormancy and formation of seed banks that ensure a greater dispersal in
time (Venable and Brown 1988, Bakker et al. 1996) and to reduce the risk of germination in
adverse periods (Garwood 1983, Reubens et al., 2007). Thus, it is expected that in
communities hosting species with different dispersal strategies, the match between seed
rain and seed bank will be limited and will play an important role to understand the
different regeneration strategies of species and the dynamics of community structure (Rico-
Gray and García-Franco 1992, Drake 1998, Drake and Moles 1999).
Most studies examining the role of seed bank and seed rain in regeneration have been
carried out in areas of secondary vegetation, in order to determine the regenerative capacity
of the vegetation after disturbances (Hopkins and Graham 1984, Moles and Drake 1999).
The few studies carried out in areas of mature vegetation show a decline in the importance
of the seed bank (Guevara and Gomez-Pompa 1972, Hall and Swaine 1980, Hopkins and
Graham 1984, Saulei and Swaine 1988, Teketay and Granström 1995, Drake 1998) and an
increase in the importance of seed rain for regeneration compared to secondary vegetation
(Dungan et al. 2001). Other studies have shown the importance of seed banks for
116
regeneration after disturbance events in sTDF (Garwood 1989); however, most studies have
focused only on the seed bank (Rico-Gray and García-Franco 1992, Teketay and Granström
1995, Fornara and Dalling 1995, Reubens et al. 2007, Uasuf et al. 2009). This approach
may underestimate the role of seeds recently dispersed and that do not form seed banks
(Dungan et al. 2001).
Simultaneous analysis of patterns of seed rain and seed bank of a community, as well as the
similarity of species between these two components and established vegetation, can provide
valuable information about the extent to which seed rain and seed bank are determining the
vegetation composition (Henderson et al. 1988, Hopfensperger 2007). In this work we
addressed whether temporal patterns of seed dispersal and / or dispersal syndrome control
the relationship between vegetation and seed compartments (seed rain or seed bank) in a
sTDF at southern Ecuador. To address this objective, we raised the following questions: (1)
how the overall pattern of seed rain does change along the year and how does this pattern
differ between dispersal syndromes? (2) What is the species richness and abundance in the
different compartments and different dispersal syndromes? (3) To what extent is richness,
abundance and identity of species consistent across the three compartments -seed rain, seed
bank and established plants-? (4) Can the dispersal syndrome explain the degree of
similarity between the established plant community and seed rain or seed bank?
MATERIALS AND METHODS
Study site—The study was conducted in the Arenillas Ecological Reserve (REA for its
acronym in Spanish), located in southwestern Ecuador (03° 34’ 15.44’’S; 80° 08’ 46.15’’E,
30 m a.s.l.). Annual mean precipitation is 667 mm, with a dry season (< 40 mm) extending
117
from June to December and a rainy season extending from January to May, and annual
mean temperature is 25°C (Huaquillas weather station, n = 45 years). The REA holds one
of the last remnants of sTDF in the Ecuadorian Pacific coast, as well as lowland dry scrub
communities and mangroves in the lowest areas (Sierra 1999). It was included within the
Patrimony of the State Natural Areas (PANE) in 2001 (BirdLife International 2013) and
has been protected from extractive activities for about 60 years.
Sampling of the established vegetation—A 9 ha plot was set in the center of REA, in one
of the best conserved areas, consisting of transitional vegetation between dry deciduous
forest and lowland scrub. In this plot, all woody plants > 5 cm diameter at breast height
(DBH) have been recorded since 2009. Additionally, in the 1 ha central subplot all
individuals of sub-shrubby woody species with DBH > 1 cm were recorded. In total, 3780
individuals from 41 species have been recorded (Appendix 1). According to fruit and seed
traits, and direct observation in the field, species were classified into three dispersal
syndromes: zoochory, anemochory and autochory. These data on the established woody
species were used as benchmarkfor comparison with the seed rain and the seed bank.
Sampling of the seed rain—To sample the seed rain, seed traps were used. Traps were
made of fabric < 0.5 mm mesh light to allow water runoff while preventing seed loss. The
fabric was placed on a square metal frame 80 cm long (0.64 m2) located 80 cm above the
ground (Stevenson and Vargas 2008). A heavy object was placed in the center of each trap
to prevent seed loss by wind.
A total of 265 traps were placed in a regular grid in an area of 4.8 ha of forest, in the center
of the 9 ha plot. Traps were 14 m apart. Seed rain was collected monthly for 13 months
118
between July 2011 and July 2012. All seeds falling into the traps were identified (see
below) and counted. When the dispersal units were whole fruits or its parts, the number of
seeds contained in them was counted.
Sampling of the soil seed bank—The soil seed bank was sampled once in November-
December 2011, at the end of the dry season, when most seeds had been dispersed and
before the germination that occurs in the rainy season. As in the case of the seed rain, 265
soil samples were collected in 0.25 x 0.25 x 0.03 m squares located next to the seed traps,
for a total sampled area of 16.56 m2.
Seed bank composition was assessed using the germination method, which allows
estimating viable seeds in the soil (Bigwood & Inouye 1988, Holzapfel et al. 1993). Soil
samples were placed in trays with a basal layer of pumice to control the moisture. Trays
were kept under controlled watering for six months and emerging seedlings were identified,
counted and removed at 3-4 day intervals.
Species identification in the seed rain and bank—Taxonomic identity of seeds and
seedlings was verified by using specimens collected in the area or, in some cases, literature.
When possible, seeds and seedlings were identified to species; otherwise to genus or
family. Unidentified seeds in the seed rain were planted for subsequent identification. All
non-woody species recorded in the seed rain and bank were excluded from analysis.
Statistical analyses—Monthly variation in seed abundance in the seed rain for all species
and by dispersal syndrome was analyzed by using generalized linear models (GLMs). An a
posteriori Duncan test (Duncan 1955) was performed to determine which monthly averages
119
were significantly different. Differences between dry and rainy seasons in species richness
and abundance in the seed rain were also compared by using GLMs. Depending on the
season of highest seed abundance, seed rain of each species was assigned to either the dry
or the rainy season.
Association between dispersal syndrome and compartment -established plants, seed rain
and seed bank- in species richness and abundance was tested using an extension of the
Fisher exact test for > 2 x 2 contingency tables. Significance was obtained by Monte Carlo
simulations (R Core Team 2012).
Similarity in species composition between the three compartments -established plants, seed
rain and seed bank- was assessed using the Sørensen similarity index (Oksanen et al. 2012),
with presence-absence data. Sørensen similarity index ranges from 0 to 1, where 1 indicates
identical composition. Similarity was analyzed both for all species together and for each
dispersal syndrome separately. Furthermore, we evaluated the similarity in abundance
between the three compartments, for all species and for each dispersal syndrome, using
Pearson correlation. For established plants we considered only their abundance in the
central 1 ha plot, where data for trees, shrubs and sub-shrubs were available. For tree
species were considered only individuals with DBH > 5 cm. Only five species present in
the 9 ha plot were not present in the 1 ha plot used for this analysis.
All statistical analyses were carried out with the programming environment R, version
2.15.1.
120
RESULTS
Seasonal variation in the seed rain—Species richness in the seed rain ranged 4-17 species
per month (Fig. 1). The highest species richness was recorded at the beginning of the dry
season, between May and July, and the lowest in December, just before the beginning of
the rains. Autochorous species were present in the seed rain all the year round, with a
richness peak between May and July (Fig. 1). The richness of zoochorous species was high
during rainy months and the first months of the dry season (Fig. 1). All anemochorous
species dispersed seeds in the dry season, and only two continued their dispersal process
once the rains started (Fig. 1).
Figure1. Temporal patterns of woody species number in the seed rain by dispersal syndrome. The
gray area indicates the rainy season.
121
In the overall seed rain, seed abundance showed significant differences between months (R2
= 0.016, p < 0,001). The Duncan a posteriori test revealed that February, March and July
were months with a significantly higher abundance than the other months, with a maximum
in February (Fig. 2a). For each dispersal syndrome, seed abundance in the seed rain also
was significantly different among months (p < 0.001 in all three cases). For anemochory,
seed abundance reached a maximum in October, during the dry season (Fig. 2b). Autochory
showed two peaks in seed abundance, one in February due mainly to Croton spp., at the
beginning of rainy season, and a second in July, at the beginning of the dry season (Fig.
2c). Zoochory showed a maximum of seed abundance in March, at the middle of the rainy
season (Fig. 2d). From May to January zoochorous seed abundance remained low (Fig. 2d).
We did not find significant differences in total species richness in the seed rain between dry
and rainy seasons (Fig. 3a). However, the number of anemochorous species in the seed rain
was significantly higher in the dry season compared to the rainy season, while for
autochorous and zoochorous species; we did not find significant differences between
seasons (Fig. 3a). Regarding seed abundance, the overall seed rain was significantly higher
in the rainy season (Fig. 3b). Significant differences between dry and rainy seasons in seed
abundances were found for anemochorous and zoochorous species (Fig. 3b). Ninety eight
percent of anemochorous seeds was dispersed during the dry season and 88% of
zoochorous seeds was dispersed during the rainy season. For autochory there was no
significant difference in seed abundance between seasons (Fig. 3b).
122
Figure 2. Monthly average seed abundance per trap in the whole data set (a) and for each dispersal
syndrome (b-d). Segments indicate the standard error. Within each panel, means with different
letters were significantly different according to Duncan test. The gray area indicates the rainy
season.
123
Figure 3. Monthly mean of (a) species number and (b) seed abundance in the seed rain for dry and
rainy seasons. Segments on the bars indicate the standard error. n.s.: not significant, m.s.:
marginally significant (p < 0.1); *: p < 0.05, **: p < 0.01; ***: p < 0.001.
Species richness and seed abundance in the seed rain and the seed bank—A total of 51
species were recorded among the three studied compartments: established plants, seed rain
and seed bank (Appendix 1). In the seed rain we collected 15546 seeds (91.7 seeds m-2
)
from 38 woody species, with a mean density (standard error) by species of 2.4 (+ 0.76)
124
seeds m-2
(Appendix 1). The main contribution to the seed rain came from Croton spp. and
Erythroxylum glaucum (Appendix 1). In the seed bank we recorded a total of 891 seeds
(53.4 seeds m-2
) from 20 woody species, with a mean density (standard error) of 2.7 + 0.79
seeds m-2
(Appendix 1). The main contribution to the seed bank came from Alternanthera
brasiliana y Croton rivinifolius (Appendix 1). The only taxa that showed a high abundance
in the three compartments (established plants, seed rain, seed bank) were two autochorous
shrubs from genus Croton.
Table 1. Number of species or individuals (percentage) of each dispersal syndrome in each
compartment studied: established plants, seed rain and seed bank.
Species no. Abundance
Syndrome Established plants Seed rain
Seed
bank
Established
plants
Seed rain
Seed
bank
Anemochory 12 (29) 8 (21) 4 (21) 737 (19) 1568 (10) 284 (34)
Autochory 13 (31) 16 (42) 12 (63) 1720 (46) 7250 (47) 465 (56)
Zoochory 16 (40) 14 (37) 3 (16) 1323 (35) 6728 (43) 80 (10)
Species richness of each dispersal syndrome did not show significant differences among
compartments –established plants, seed rain or seed bank– (Table 1, Fisher’s exact test, p =
0.195). By contrast, dispersal syndromes significantly differed in abundance among
compartments (Table 1, Fisher’s exact test, p < 0.001). Autochory was the dispersal
syndrome with greatest abundance of individuals in the three compartments (Table 1).
125
Anemochory showed the lowest abundance of individuals both in established plants as in
the seed rain, while zoochory showed the lowest abundance in the seed bank (Table 1,
Appendix 1).
Similarity among established plants, seed rain and seed bank—From 41 woody species
present as established plants, 13 species were recorded both in the seed rain and the seed
bank, 17 species were recorded only in the seed rain and two species (Piptadenia flava and
Alternanthera brasiliana) only in the seed bank. Nine species of established plants were
absent in both the seed rain and the seed bank, including three dominant tree species; the
two Tabebuia spp. and Cynophalla mollis.
Ten species registered in the seed rain or seed bank were absent as established plants. Of
these, three species were found both in the seed rain and the seed bank, five species were
found only in the seed rain and two species only in the seed bank.
In general, Sørensen similarity index was equal to or above 0.5 among compartments, and
decreased in the sense established plants–seed rain > seed rain–seed bank > established
plants–seed bank (Fig. 4). Similarity between established plants and seed bank, as well as
between seed rain and seed bank was lower than 0.4 only for zoochorous species.
Fig. 4. Sørensen’s similarity index for species composition between established plant community
(EP), seed rain (SR) and seed bank (SB) considering all species and each dispersal syndrome.
126
Table 2. Pearson correlation coefficient (R), significance (P) and sample size (n) of the relationship
between the abundance of species in the three compartments studied: established plants, seed rain
and seed bank.
n R P
All species
Established plants vs. Seed rain 42 0.425 0.005
Established plants vs. Seed bank 41 0.521 <0.001
Seed rain vs. Seed bank 33 0.295 0.058
Anemochory
Established plants vs. Seed rain 12 -0.124 0.485
Established plants vs. Seed bank 12 0.420 0.174
Seed rain vs. Seed bank 9 -0.207 0.520
Autochory
Established plants vs. Seed rain 13 0.809 <0.001
Established plants vs. Seed bank 13 0.797 <0.001
Seed rain vs. Seed bank 12 0.941 <0.001
Zoochory
Established plants vs. Seed rain 17 -0.206 0.427
Established plants vs. Seed bank 16 0.206 0.428
Seed rain vs. Seed bank 12 -0.167 0.522
Pearson correlation coefficient applied to the whole data set revealed that seed abundance
of a given species in the seed rain or in the seed bank significantly increased with its
abundance in the established vegetation (Table 2). However, when the correlation in
127
abundance between compartments was analyzed separately for each dispersal syndrome,
we only found a significant positive correlation for autochorous species (Table 2).
DISCUSSION
Seasonal variation in the seed rain—Species with different dispersal syndrome showed
differences in temporal patterns of seed dispersal. This result is consistent with reports for
other Neotropical dry forests (Justiniano and Fredericksen 2000, Griz and Machado 2001,
Ragusa-Netto and Silva 2007) and can be explained as adjustment of different dispersal
syndromes to particular environmental conditions for an efficient dispersal. Restriction of
anemochorous seed rain to dry season could be related to the greater wind circulation in the
canopy and lower strata of vegetation by the absence of foliage, which facilitates the
movement of anemochorous seeds at greater distances (Howe and Smallwood 1982 Griz
and Machado 2001, Jara-Guerrero et al., 2011). A higher presence and abundance of
zoochorous species in the rainy season has been suggested as a strategy to take advantage
of high humidity and germinate immediately after dispersal (Howe and Smallwood 1982,
Bullock 1995, Jara-Guerrero et al. 2011). No clear seasonal pattern was found in the seed
rain of autochorous species, suggesting a high tolerance to the characteristic environmental
variations of highly seasonal environments (Jara-Guerrero et al., 2011).
Seed dispersal of zoochorous species early in the dry season has also been reported in other
dry ecosystems (Garwood 1983, Justiniano and Fredericksen 2000, Griz and Machado
2001, Ragusa-Netto and Silva 2007). Two potential explanations could be the shortness of
the rainy season in these sites (Ragusa-Netto and Silva 2007), or the occurrence of rainfall
events during the beginning of the dry season (Garwood 1983). There are two potential
128
ecological consequences of zoochorous seed dispersal in the dry season. The first one is a
lower predation and fungal attack (Garwood, 1983), although this advantage can also apply
to other dispersal syndromes (Ragusa-Netto and Smith 2007). The second is the formation
of seed banks (see below) in some zoochorous species of hard seed coat, such as Bursera
graveolens and Cordia spp., as a strategy to survive the dry season (Garwood 1983, Baskin
and Baskin 1998).
The pattern of higher species richness during the dry season coincides with that reported for
other sTDF (Griz and Machado 2004 Ceccon and Hernández 2009, Jara-Guerrero et al.,
2011, Martinez-Garza et al., 2011). However, while other studies report also a higher
abundance of seeds during the dry season (Ceccon and Hernández 2009, Martínez-Garza et
al. 2011), we found a significantly higher seed abundance during the rainy season, despite
its short duration. This difference in the temporal pattern of abundance can be explained by
the high richness (37%) and abundance (43%) of zoochorous species compared to other
sTDF (13% of species and <1% of seed density in the seed rain: Ceccon and Hernández
2009).
Similarity among established plants, seed rain and seed bank—Consideration of dispersal
syndromes revealed significant differences in the match in species richness and seed
abundance between compartments. In particular, few zoochorous and anemochorous
species formed seed banks. This is because a significant proportion of species synchronized
their dispersal and/or germination events with rainfalls, which allows maximizing the
chances of seedling establishment (Garwood 1983, Espinosa et al. 2012), but reduces the
stay in the soil seed bank. For example, Garwood (1983) reported for a sDTF of Panama
129
that germination of anemochorous species occurs in the rainy season inmediatly following
seed dispersal, while for zoochorous species, germination occurs during the same period of
showers that are dispersed. This result supports the idea that zoochorous and anemochorous
species depend mainly on seed rain for regeneration (Bakker et al. 1996, Thompson 2000).
Thus, as reported for other forest ecosystems (Teketay & Granström 1995, Quintana-
Ascencio et al. 1996, Clark et al. 1999, Thompson 2000, Uasuf et al. 2009), in our study
area the seed bank had a limited role in regeneration of woody species.
Autochory was the only dominant syndrome in the three studied compartments.
Furthermore, this was the only dispersal syndrome that showed a correlation in abundance
between those three compartments. This result suggests that limited capacity of dispersal,
characteristic of autochorous species, could be compensated by a large influx of seeds from
the established vegetation and by the ability to form seed banks. By contrast, in zoochorous
and anemochorous species abundance of seeds in the seed rain may be altered by the entry
and exit of seeds from surrounding areas, because of their high dispersal ability. This could
explain the presence of some species in the rain or the seed bank not present as established
plants.
The high similarity in species composition between established plants and seed rain is
consistent with that reported in other studies (Drake 1998). However, the similarity of the
seed bank with the other two compartments was higher in this study compared to findings
in previous studies (Drake 1998, Uasuf et al. 2009). This difference is probably due
differences across studies in the influx of species from outside the study area. In our study
are, all species in the seed rain and seed bank were present in the established vegetation,
130
while in the study of Drake (1998) 67% of the seed bank corresponds to species that were
not present in the established vegetation. Furthermore, our study area is a mature forest, and
relatively homogeneous regarding species composition. Those few species found in seed
rain or seed bank, but absent in the established plant community, were mainly small
autochorous subshrubs established in the study site that, due to their very small stem
diameter, have not been considered in the inventory of established vegetation. Other three
species were zoochorous climbers, which were present only in the seed rain, therefore, their
seeds contribution could come from individuals in the surrounding areas to the plot.
In conclusion, our results suggest that the seed rain of woody species in the dry forest
studied shows a seasonal variation in species number and seed abundance. The higher seed
abundance was restricted to dry season in anemochorous species, and to rainy season in
zoochorous species. In autochorous species seed rain showed maximums in the both
seasons. Formation of seed banks was dependent on dispersal syndrome, being those
species with limited dispersal abilities (autochorous) the ones with highest species richness
and seed abundance in the seed bank. Simultaneous consideration of seed rain and seed
bank allows us to suggest that while most woody species depended on seed rain for
regeneration, as proposed in general for forest communities in advanced successional stage,
seed bank can be an effective supply of seeds for some species of plants, mainly for
autochorous species.
REFERENCES
ÁLVAREZ-BUYLLA, E. R., MARTÍNEZ-RAMOS, M. 1990. Seed bank vs seed rain in
the regeneration of a tropical pioneer tree. Oecologia 84: 314-325.
131
ARCHIBOLD, O. W. 1979. Buried viable propagules as a factor in postfire regeneration in
northern Saskatchewan. Canadian Journal of Botany 57: 54-58.
BASKIN C. C., BASKIN J. M. 1998. Seeds: Ecology, biogeography and evolution of
dormancy and germination. Academic Press, San Diego, California, 666 p.
BAKKER, J. P., POSCHLOD, P., STRYKSTRA, R. J., BEKKER, R., THOMPSON, K.
1996. Seed banks and seed dispersal: important topics in restoration ecology. Acta
Botanica Neerlandica 45: 461-490.
BIGWOOD, D. W., INOUYE, D. W. 1988. Spatial pattern analysis of seed banks: an
improved method and optimized sampling. Ecology 69: 497-507.
BIRDLIFE INTERNATIONAL. 2013. Important Bird Areas factsheet: Reserva Ecológica
Arenillas. Downloaded from http://www.birdlife.org on 12/03/2013
BROWN, J. S., VENABLE, D. L. 1986. Evolutionary ecology of seed-bank annuals in
temporally varying environments. American Naturalist 127: 31-47.
BULLOCK, S. H. 1995. Plant reproduction in Neotropical dry forests. In S. Bullock, H.
Mooney, and E. Medina (Eds.). Seasonally dry tropical forests, pp. 277–297.
Cambridge University Press, Cambridge, UK.
CECCON, H., HERNÁNDEZ, P. 2009. Seed rain dynamics following disturbance
exclusion in a secondary tropical dry forest in Morelos, Mexico. Journal of Tropical
Biology 57: 257-269.
132
CLARK, J. S., BECKAGE B., CAMILL P., CLEVELAND B., HILLERISLAMBERS J.,
LICHTER J., MCLACHLAN J., MOHAN, J., WYCKOFF, P. 1999. Interpreting
recruitment limitation in forests. American Journal of Botany 86: 1–16.
DRAKE, D. R. 1998. Relationships among the seed rain, seed bank and vegetation of a
Hawaiian forest. Journal of Vegetation Science 9: 103-112.
DUNCAN, D. B. 1955. Multiple range and multiple F test. Biometrics 11: 1-42.
DUNGAN, R. J., NORTON, D. A., DUNCAN, R. P. 2001. Seed rain in successional
vegetation, Port Hills Ecological District, New Zeland. New Zeland Journal of
Botany 39: 115-124.
ESPINOSA, C. I., DE LA CRUZ, M., LUZURIAGA, A. L., ESCUDERO, A. 2012.
Bosques tropicales secos de la región Pacífico Ecuatorial: diversidad, estructura,
funcionamiento e implicaciones para la conservación. Ecosistemas 21: 167-179.
FORNARA, D. A., DALLING, J. W. 2005. Seed bank dynamics in five Panamanian
forests. Journal of Tropical Ecology 21: 223-226.
GARWOOD, N. 1983. Seed germination in a seasonal tropical forest in Panama: a
community study. Ecological Monographs 53: 159–181.
GRIZ, L. M. S., MACHADO, I. C. S. 2001. Fruiting phenology and seed dispersal
syndromes in Caatinga, a tropical dry forest in the northeast of Brazil. Journal of
Tropical Ecology 17: 303–321.
133
GUEVARA, S., GÓMEZ-POMPA, A. 1972. Seeds from surface soils in a tropical region
of Veracruz, Mexico. Journal of the Arnold Arboretum 53: 312-335.
HALL, J. B., SWAINE, M. D. 1980. Seed stocks in Ghanaian forest soils. Biotropica 12:
256-263.
HENDERSON C. B., PETERSEN K. E., REDAK R. A. 1988. Spatial and temporal
patterns in the seed bank and vegetation of a desert grassland community. Journal of
Ecology 76: 717-728.
HOLZAPFEL, C. SCHMIDT, W., SHMIDA, A. 1993. The role of seed bank and seed rain
in the recolonization of disturbed sites along an aridity gradient. Phytocoenologia
23: 561-580.
HOPFENSPERGER, K. N. 2007. A review of similarity between seed bank and standing
vegetation across ecosystems. Oikos 116: 1438-1448.
HOPKINS, M. S., GRAHAM, A. W. 1984. Viable soil seed banks in disturbed lowland
tropical rainforest sites in north Queensland. Australian Journal of Ecology 9: 71-
79.
HOWE, H. F., SMALLWOOD, J. 1982. Ecology of seed dispersal. Annual Review of
Ecology and Systematics 13: 201-228.
JANZEN, D. H. 1967. Synchronization of sexual reproduction of trees within dry season in
Central America. Evolution 21: 620-637.
134
JARA-GUERRERO, A., DE LA CRUZ, M., MÉNDEZ, M. 2011. Seed dispersal spectrum
of woody species in South Ecuadorian dry forests: environmental correlates and the
effect of considering species abundance. Biotropica 43: 722-730.
JUSTINIANO, M., FREDERICKSEN, T. 2000. Phenology of tree species in Bolivian dry
forests. Biotropica 32: 276-281.
KHURANA, E., SINGH, J. S. 2001. Ecology of seed and seedling growth for conservation
and restoration of tropical dry forest: a review. Environmental Conservation 28: 39–
52.
MARTÍNEZ-GARZA, C., OSORIO-BERISTAIN, M., VALENZUELA-GALVÁN, D.,
NICOLÁS-MEDINA, A. 2011. Intra and inter-annual variation in seed rain in a
secondary dry tropical forest excluded from chronic disturbance. Forest Ecology
and Management 262: 2207-2218.
MAZA-VILLALOBOS, S., LEMUS-HERRERA, C., MARTÍNEZ-RAMOS, M. 2010.
Successional trends in soil seed banks of abandoned pastures of a Neotropical dry
region. Journal of Tropical Ecology 27: 35-49.
MOLES, A. T., DRAKE, D. R. 1999. Potential contributions of the seed rain and seed bank
to regeneration of native forest under plantation pine in New Zealand. New Zealand
Journal of Botany 37: 83-93.
NATHAN, R., MULLER-LANDAU, H. 2000. Spatial patterns of seed dispersal, their
determinants and consequences for recruitment. Trends in Ecology and Evolution
15: 278-285.
135
OKSANEN, J., GUILLAUME BLANCHET, F., KINDT, R., LEGENDRE, P., MINCHIN,
P., O’HARA, R. B., SIMPSON, G., SOLYMOS, P., STEVENS, M., WAGNER, H.,
2012. Package “vegan”. Community Ecology Package. URL http://cran.r-
project.org, http://vegan.r-forge.r-project.org/
PARTRIDGE, T. R. 1992: Successional interactions between bracken and broom on the
port Hills Canterbury’ New Zealand. Journal of Applied Ecology 29: 85-91.
PUTZ, F. E., APPANAH, S. 1987. Buried seeds, newly dispersed seeds, and the dynamics
of a lowland forest in Malaysia. Biotropica 28: 192-209.
QUINTANA-ASCENCIO, P., GONZÁLEZ-ESPINOSA, M., RAMÍREZ-MARCIAL, N.,
DOMÍNGUEZ-VÁZQUEZ, G., MARTÍNEZ-ICO, M. 1996. Soil seed banks and
regeneration of tropical rain forests from milpa fields at the Selva Lacandona,
Chiapas, México. Biotropica 28: 192-209.
R CORE TEAM. 2012. R: A language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria.
RAGUSA-NETTO, J., SILVA, R. 2007. Canopy phenology of a dry forest in western
Brazil. Brazilian Journal of Biology 67: 569-575.
REUBENS, B., HEYN, M., GEBREHIWOT, K., HERMY, M., MUYS, B. 2007. Persistent
soil seed banks for natural rehabilitation of dry tropical forests in northern Ethiopia.
Tropicultura 25: 204-214.
136
RICO-GRAY, V., GARCÍA-FRANCO, J. G. 1992. Vegetation and soil seed bank of
successional stages in tropical lowland deciduous forest. Journal of Vegetation
Science 3: 617-624.
SAULEI, S. M., SWAINE, M. D. 1988. Rainforest seed dynamics during succession at
Gogol, Papua New Guinea. Journal of Ecology 76: 1133-1152.
SCHUPP, E., FUENTES, M. 1995. Spatial patterns of seed dispersal and the unification of
plant population ecology. Écoscience 2: 267-275.
SEM, G., ENRIGHT, N. J. 1996. The relationship between seed rain and the soil seed bank
in a temperate rainforest stand near Auckland, New Zealand. New Zealand Journal
of Botany 34: 215-226.
SIERRA, R. (Ed.). 1999. Propuesta preliminar de un sistema de clasificación de vegetación
para el Ecuador Continental. Proyecto INEFAN/GEF-BIRF y EcoCiencia. Quito,
Ecuador.
SINGH, J. S., SINGH, V. K. 1992. Phenology of seasonally dry tropical forest. Current
Science 63: 684-689.
STEVENSON, P., VARGAS, I. 2008. Sample size and appropriate design of fruit and seed
traps in tropical forests. Journal of Tropical Ecology 24: 95–105.
STRICKLER, G. S., EDGERTON, P. J. 1976. Emergent seedlings from coniferous litter
and soil in eastern Oregon. Ecology 57: 801-807.
137
TEKETAY, D., GRANSTRÖM, A. 1995. Soil seed banks in dry Afromontane forests of
Ethiopia. Journal of Vegetation Science 6: 777-786.
THOMPSON, K. 2000. The functional ecology of soil seed banks. In: FENNER (Ed.).
Seeds: the ecology of regeneration in plant communities. 2nd
Ed. CAB International.
UK, pp. 215 – 235.
UASUF, A. TIGABU, M. ODÉN, P. C. 2009. Soil seed banks and regeneration of
Neotropical dry deciduous and gallery forests in Nicaragua. Bois et Forêts des
Tropiques 299: 49–62.
UHL, C., CLARK, K., CLARK, H., MURPHY, P. 1981. Early plant succession after
cutting and burning in the upper Rio Negro region in the Amazon Basin. Journal of
Ecology 69: 631-649.
VAN DER PIJL, L. 1969. Principles of dispersal in higher plants. Springer-Verlag, Berlin.
VENABLE, D. L., BROWN, J. S. 1988. The selective interactions of dispersal, dormancy,
and seed size as adaptations for reducing risk in variable environments. American
Naturalist 131: 360-384.
138
MATERIAL SUPLEMENTARIO PARA EL CAPÍTULO 3.
APPENDIX 1. Woody species present in the established plant community, seed rain and/or soil seed bank. Relative abundance (RA), mean +
standard error basal area (BA) and DBH, total number of seeds, seed density seed /m2 (s/m
2), dispersal season (Season), growth form (GF.: T, tree;
S, shrub; C, climber) and dispersal syndrome (Zo, zoochory; An, anemochory; Au, autochory). NP: not present.
Established plant community Seed rain Seed bank
Species RA BA DBH No. seeds s/m-2
No. seeds s/m-2
Season GF Syndrome
Achatocarpaceae
Achatocarpus pubescens 1.0 13.7 + 2.9 3.6 + 0.4 639 3.77 NP NP Rainy T Zo
Amaranthaceae
Alternanthera brasiliana 2.5 2.0 + 0.1 1.5 + 0.0 NP NP 210 12.58 S An
Amaranthaceae 2979 17.56 Continuous S Au
Iresine sp. 0.03 4.8 + 0.0 2.5 + 0.0 3 0.02 NP NP Dry S An
Asteraceae
Asteraceae sp. 0.1 2.6 + 1.4 1.7 + 0.4 10 0.06 NP NP Dry S An
Verbesina sp. 0.7 3.2 + 0.9 1.8 + 0.2 133 0.78 NP NP Dry S An
Viguiera sp. 1.4 1.1 + 0.0 1.2 + 0.0 NP NP NP NP S An
Boraginaceae
Cordia macrocephala 3.9 2.9 + 0.3 1.8 + 0.1 50 0.29 15.0 0.90 Rainy S Zo
Cordia rosei 3.7 4.5 + 0.5 2.1 + 0.1 885 5.22 NP NP Rainy S Zo
Bignoniaceae
Tabebuia billbergii 2.1 674.4 + 71.2 25.1 + 1.6 NP NP NP NP T An
139
APPENDIX 1. Continuation.
Established plant community Seed rain Seed bank
Species RA BA DBH No. seeds s/m-2
No. seeds s/m-2
Season GF Syndrome
Bignoniaceae (continuation)
Tabebuia chrysantha 1.2 357.6 + 92.2 14.6 + 2.3 NP NP NP NP T An
Bixaceae
Cochlospermum vitifolium 1.3 353.5 + 65.8 16.6 + 1.8 345 2.03 2 0.12 Dry T An
Burseraceae
Bursera graveolens 0.3 16.7 + 303.8 23.4 + 5.8 131 0.77 8.0 0.48 Rainy T Zo
Cactaceae
Armatocereus sp. 7.0 24.2 + 0.3 5.5 + 0.0 49 0.29 NP NP Rainy S Zo
Caesalpiniaceae
Caesalpinia glabrata 0.7 236.9 + 67.5 13.9 + 2.0 10 0.06 NP NP Dry T Au
Senna bicapsularis 0.8 1.9 + 0.2 1.5 + 0.1 NP NP NP NP S Au
Senna mollisima 0.1 8.3 + 3.8 2.9 + 0.8 116 0.68 2.0 0.12 Dry T Au
Capparaceae
Colicodendron scabridum 1.0 159.2 + 51.9 9.5 + 1.7 NP NP NP NP T Zo
Cynophalla mollis 4.0 63.1 + 8.2 6.9 + 0.5 NP NP NP NP T Zo
Caricaceae
Vasconcellea parviflora 0.6 9.3 + 2.7 2.8 + 0.4 NP NP NP NP S Zo
Convolvulaceae
Ipomoea carnea 6.7 2.3 + 0.1 1.6 + 0.0 19 0.11 63 3.78 Dry C An
Erythroxylaceae
Erythroxylum glaucum 0.9 188.6 + 63.4 11.0 + 1.8 2851 16.81 NP NP Rainy T Zo
Euphorbiaceae
Croton rivinifolius 19.1 1.9 + 0.1 1.5 + 0.0 2919 17.21 170 10.18 Rainy S Au
Croton micans 6.5 19.2 + 1.3 4.3 + 0.2 27 0.16 10 0.60 Rainy S Au
140
APPENDIX 1. Continuation.
Established plant community Seed rain Seed bank
Species RA BA DBH No. seeds s/m-2
No. seeds s/m-2
Season GF Syndrome
Fabaceae
Aeschynomene scoparia 1.1 2.3 + 0.4 1.6 + 0.1 41 0.24 9 0.54 Dry S An
Canavalia ensiformis 11 0.06 43 2.58 Dry C Au
Chloroleucon mangense 2.0 136.5 + 22.2 11.4 + 0.8 259 1.53 8 0.48 Dry T Au
Cyathostegia sp. NP NP 116 6.95 S Au
Erythrina velutina 0.2 288.6 + 91.4 17.5 + 2.8 19 0.11 3 0.18 Dry T Au
Geoffroea spinosa 1.0 46.9 + 14.3 5.9 + 0.8 NP NP NP NP T Zo
Leucaena trichodes 2.3 12.3 + 3.3 3.0 + 0.3 56 0.33 52 3.12 Dry T Au
Mimosa acantholoba 0.1 10.9 + 7.7 3.0 + 1.3 1 0.01 NP NP Dry T Au
Piptadenia flava 1.1 7.8 + 1.8 2.6 + 0.3 NP NP 1 0.06 T Au
Pithecellobium excelsum 0.7 33.9 + 10.0 5.1 + 0.7 24 0.14 NP NP Dry T Zo
Lamiaceae
Hyptis sp. 1.2 3.7 + 0.4 2.0 + 0.1 852 5.02 NP NP Dry S An
Malpighiaceae
Malpighia emarginata 4.7 21.9 + 1.7 4.6 + 0.2 140 0.83 NP NP Rainy S Zo
Malvaceae
Briquetia spicata NP NP NP 5 0.03 54 3.25 Dry S Au
Byttneria flexuosa 11.7 2.4 + 0.1 1.6 + 0.0 150 0.88 3 0.18 Dry S Au
Eriotheca ruizii 1.0 190.6 + 35.0 12.4 + 1.5 165 0.97 NP NP Dry T An
Malvaceae sp. 1 0.8 1.2 + 0.1 1.2 + 0.0 7 0.04 NP NP Dry S Au
Malvaceae sp. 4 NP NP NP 255 1.50 NP NP Rainy S Au
Pavonia mutisii NP NP NP 16 0.09 3 0.18 Dry S Au
Urena sp. 0.03 0.9 + 0 1.1 + 0 420 2.48 NP NP Dry S Au
Moraceae
Ficus sp. NP NP NP 1200 7.08 NP NP Rainy C Zo
141
APPENDIX 1. Continuation.
Established plant community Seed rain Seed bank
Species RA BA DBH No. seeds s/m-2
No. seeds s/m-2
Season GF Syndrome
Polygonaceae
Coccoloba ruiziana 1.0 15.3 + 3.0 3.8 + 0.4 202 1.19 NP NP Dry T Zo
Rubiaceae
Randia aurantiaca 1.5 5.1 + 1.1 2.1 + 0.2 118 0.70 NP NP Dry S Zo
Primulaceae
Jacquinia sprucei 0.5 20.9 + 8.1 4.1 + 0.7 NP NP NP NP T Zo
Verbenaceae
Lantana sp. 3.4 1.7 + 0.1 1.4 + 0.0 8 0.05 57 3.41 Rainy S Zo
Vitaceae
Cissus sp. NP NP NP 207 1.22 NP NP Dry C Zo
Cissus syscioides NP NP NP 224 1.32 NP NP Dry C Zo
Indeterminate NP NP NP NP NP 62 3.71 S NA
Total 15546 91.66 891 53.37
142
CAPÍTULO 4
LEGITIMIDAD DE LOS CÉRVIDOS COMO DISPERSORES DE SEMILLAS DE
ESPECIES LEÑOSAS EN UN BOSQUE SECO TROPICAL
Andrea Jara-Guerrero1, Marcos Méndez
2, Carlos I. Espinosa
1, Marcelino de la Cruz
3
1Departamento de Ciencias Naturales, Universidad Técnica Particular de Loja, CP.: 11-01-
608, Loja, Ecuador.
2Área de Biodiversidad y Conservación, Universidad Rey Juan Carlos, E-28933 Móstoles,
Spain.
3Departamento de Biología Vegetal, E.U.I.T. Universidad Politécnica de Madrid, 28040
Madrid, Spain.
143
INTRODUCCIÓN
La dispersión de semillas por animales es una de las estrategias predominantes en los
bosques tropicales (Howe & Smallwood 1982, Jordano 2000, Link & Stevenson 2004). La
zoocoria puede ser el medio de dispersión de hasta el 90% de especies de plantas en
algunos bosques tropicales lluviosos (Link & Stevenson 2004). Incluso en ecosistemas
secos, donde la disponibilidad de frutos es restringida en el tiempo (Bullock 1995), se ha
registrado hasta un 58,7% de especies dispersadas por endozoocoria (Jordano 2000).
Dentro del estudio de dispersión de semillas por animales, probablemente el grupo que
mayor atención ha recibido es el de las aves (Link y Stevenson 2004, Stiles 2000), seguido
por los mamíferos, ambos considerados como los dispersores de semillas más importantes
en los trópicos (Fleming 1986, Howe y Smallwood 1982, Jordano 2000, Stiles 2000, Link y
Stevenson 2004).
La dispersión de semillas por cérvidos, y por ungulados en general, ha sido relacionada a
plantas con frutos de color verde o café, con pulpa fibrosa seca, una fuerte protección de la
semilla y con un peso mayor a 50 g (Gauthier-Hion et al. 1984). El consumo de frutos
pesados puede estar relacionado con sus necesidades energéticas. Dada la fuerte asociación
entre el peso de la fruta y contenido de fibra, la elección de frutos grandes es indispensable
para la dieta de ungulados (Gauthier-Hion et al. 1984). Sin embargo, el papel de dispersores
de los cérvidos ha recibido poca atención (González-Espinosa y Quintana-Ascencio 1986,
Malo y Suárez 1996, Mandujano et al. 1997). La importancia de los cérvidos como
dispersores ha sido estudiada en algunos ecosistemas considerando dos atributos; la
legitimidad y la eficiencia. La legitimidad está relacionada con su capacidad de dispersar
144
semillas viables (Herrera 1989, Silva et al. 2005). Myers et al. (2004), en un estudio
realizado en el Este de Norteamérica, encontraron que el venado de cola blanca
(Odocoileus virginianus) dispersó vía endozoócora semillas de más de 70 especies en un
período de un año, entre ellas muchas especies de semillas pequeñas sin adaptaciones
obvias para dispersión. Igualmente, en ecosistemas áridos de Norteamérica se ha
demostrado la dispersión de semillas viables de algunas especies de cactáceas luego del
paso por el tracto digestivo de O. virginianus (González-Espinosa y Quintana-Ascencio
1986, Mandujano et al. 1997). Por otro lado, la eficiencia del dispersor, relacionada con la
capacidad de mover las semillas lejos de la planta madre y depositarlas en sitios adecuados
para su desarrollo (Reid 1989, Silva et al. 2005), ha sido evaluada también para O.
virginianus en el noroeste de Norteamérica. Modelos generados a partir de los patrones de
movimiento de O. virginianus, y los tiempos de retención de semillas en el intestino,
indican que el 95% de semillas que son exitosamente dispersadas podrían ser llevadas a una
distancia de más de 100 m y que el 30% podría ser llevada a más de un kilómetro (Vellend
et al. 2003). La importancia de esta especie para llevar semillas a largas distancias es
esencial, mucho más si no hay fuentes de semillas cerca (Cain et al. 2000), incluso si el
número de semillas viables dispersadas es pequeño (Myers et al. 2004).
La función de cérvidos en la dispersión de semillas, sin embargo, está prácticamente
restringida al estudio de O. virginianus en Norteamérica (González-Espinosa y Quintana-
Ascencio 1986, Mandujano et al. 1997, Vellend et al. 2003, Myers et al. 2004), a la
dispersión de semillas de algunas especies puntuales en Europa (Malo y Suárez 1996, 2000)
y a algunos otros ungulados de ecosistemas tropicales del viejo mundo (Gauthier-Hion et
al. 1984). La atención que ha recibido este tema en los bosques del neotrópico es todavía
145
muy escasa (pero ver: Bodmer 1991). En los bosques secos del suroccidente de Ecuador se
han reportado dos especies de cérvidos Mazama americana y Odocoileus peruvianus
(Tirira 2007). El interés en estudiar procesos de dispersión por cérvidos en los bosques
secos se debe a que estos ecosistemas han sido altamente degradados desde hace varias
décadas por actividades antrópicas, lo que ha transformado al ecosistema en pequeños
fragmentos de bosque dentro de una matriz de tierras degradadas (Aguirre & Kvist 2005,
Jara-Guerrero et al. 2011, Espinosa et al. 2012). Los bosques secos han experimentado
transformaciones de hasta un 70%, convirtiéndose en uno de los casos más dramáticos de
extinción masiva de especies de plantas a causa de la deforestación (PNUMA 2010). La
pérdida de animales que desempeñan un papel clave en la dispersión de semillas tendrá
efectos negativos en la situación de numerosas especies de plantas y esta disminución, a su
vez, puede tener efectos ecosistémicos diversos (Bennett 2004). Al ser estas dos especies de
cérvidos relativamente abundantes en los bosques secos (Tirira 2007), y considerando el
amplio rango de movimiento que tienen, podrían ser una pieza clave en la regeneración de
estos ecosistemas.
El presente trabajo analiza el rol que un grupo de mamíferos comunes en los bosques secos
de Ecuador, los cérvidos, tienen en la regeneración de especies leñosas de un bosque seco
del suroccidente Ecuatoriano. Específicamente se plantean las siguientes cuestiones: (1) su
importancia como dispersores de semillas en términos de cantidad y riqueza de semillas
dispersadas. (2) Cuál es el efecto que la ingestión por cérvidos tiene sobre la capacidad
germinativa de las semillas. En muchos casos el paso por el tracto digestivo puede acelerar
la germinación o incrementar el porcentaje de germinación final (Howe y Smallwood 1982,
Traveset 1998). Estos efectos pueden ser el resultado de dos procesos; la remoción de la
146
pulpa, que puede contener inhibidores para la germinación, o del efecto abrasivo sobre la
testa de la semilla, haciéndola más permeable (Traveset 1998). (3) Si la legitimidad del
venado como dispersor es dependiente de la especie de semilla dispersada.
MATERIALES Y MÉTODOS
Área de estudio- Este estudio tuvo lugar en la Reserva Ecológica Arenillas (REA). Esta
reserva se localiza al suroccidente de Ecuador (17059485 N, 9605347 E). La reserva ha
estado protegida de actividades de extracción por aproximadamente 60 años, aunque fue
incluida dentro del Patrimonio de Áreas Naturales del Estado recién en el año 2001
(BirdLife International 2013). El clima se caracteriza por una estación seca marcada, que va
de junio a diciembre. La media anual de precipitación varía entre 500 y 1000 mm y se da
principalmente entre enero y mayo. La temperatura media anual es de 25°C. La REA
sostiene uno de los últimos relictos de bosque tropical estacionalmente seco en la costa
pacífico ecuatoriana. Incluye, además, formaciones de matorral seco de tierras bajas y
manglar en las zonas más bajas (Sierra 1999).
En el centro de la reserva, en una de las zonas mejor conservadas, se encuentra una
formación vegetal de transición entre bosque tropical seco y matorral seco de tierras bajas.
En esta zona se estableció una parcela de nueve hectáreas, en la que se han inventariado
desde el año 2009 todas las plantas leñosas con DAP > 5 cm.
Historia natural de los cérvidos- Al igual que el resto de cérvidos, las dos especies que
habitan los bosques secos del suroccidente de Ecuador, Mazama americana y Odocoileus
peruvianus son clasificados como herbívoros y considerados como verdaderos rumiantes
(Eisenberg y Redford 1999, Tirira 2007). Mazama americana es de tamaño mediano
147
(Eisenberg y Redford 1999). Es un animal solitario, prefiere habitar áreas de bosque y
puede moverse fácilmente en la vegetación densa, por lo que es difícil de observar en áreas
abiertas. El área de campeo puede alcanzar hasta 100 ha en un macho adulto (Tirira 2007).
Odocoileus peruvianus es de tamaño grande y suele ser solitario, aunque también puede
vivir en grupos pequeños. Debido a su cuerpo robusto y cuernos ramificados es más común
en espacios abiertos. Ambas especies se caracterizan por alimentarse principalmente del
ramoneo (hojas, ramas y brotes tiernos de árboles y arbustos), aunque también consumen
frutos durante la estación seca (Tirira 2007).
Muestreo de excretas- En las cinco hectáreas centrales de la parcela permanente se
estableció un transecto de aproximadamente 3,4 km con un ancho de 2 m. Este transecto
fue recorrido mensualmente entre octubre de 2011 y septiembre de 2013. Durante los
recorridos se recolectaron todas las excretas de venado observadas. Las excretas de los
venados son múltiples, por lo que cada muestra estuvo compuesta por un grupo de excretas
depositadas de forma agregada en un mismo sitio. Cada muestra fue etiquetada y colocada
en fundas plásticas para su transporte. Debido a la falta de información morfométrica no
fue posible distinguir entre excretas de O. peruvianus y M. americana. Por tanto, las
muestras se analizaron en conjunto.
Registro de semillas y germinación- Todas las semillas presentes en las excretas fueron
extraídas por trituración de la muestra. Se registró la identidad de la semilla, abundancia y
el estado, es decir, si la semilla presentaba algún tipo de daño que pudiera afectar la
germinación (como semillas incompletas o con presencia de hongos o bacterias).
148
Las semillas encontradas en las excretas fueron sembradas en bandejas de plástico llenas de
turba humedecida. Los restos de la excreta también se colocaron en la bandeja con el fin de
registrar la presencia de otras semillas que no pudieran observarse con la trituración. La
germinación fue controlada por un período de 90 días, en el cual se registró la identidad de
la plántula, abundancia y porcentaje de germinación por excreta.
Efectos de la ingestión de semillas por venado sobre el proceso de germinación- La
capacidad germinativa de semillas luego del paso por el tracto digestivo del venado se
analizó en semillas de cuatro especies: Chloroleucon mangense, Caesalpinia glabrata,
Leucaena trichodes y Senna mollissima. Semillas de estas cuatro especies fueron colectadas
directamente de las plantas y sembradas bajo las mismas condiciones que las semillas
procedentes de las excretas. Por cada especie se sembraron 100 semillas, separadas en
cuatro bandejas de 25 semillas cada una. Para este análisis se consideraron únicamente las
semillas procedentes de las excretas colectadas entre junio y septiembre de 2013, y los
testigos fueron colectados y sembrados durante el mismo período. Con estos datos se
analizó además, el efecto del paso por el tracto digestivo del venado sobre el tiempo de
germinación de las semillas.
La viabilidad de las semillas no germinadas dentro de los 90 primeros días de siembra fue
evaluada con un segundo período de siembra para el cual se utilizó un tratamiento de
escarificación de la testa. Considerando el tipo de semilla de las especies registradas, que en
general presentan una testa dura e impermeable indicadora de dormancia física, la
escarificación consistió en un lijado de la testa que facilite la imbibición del embrión.
149
Los análisis de germinación y viabilidad de semillas se utilizaron como referencia de la
legitimidad del venado como dispersor de semillas.
Análisis Estadísticos- La importancia de los venados como agentes dispersores se estimó
adaptando el índice de importancia del dispersor (IID) (Galindo-González et al. 2000):
IID= B*S / 1000
Donde, B es la abundancia relativa de la especie, que en este caso está representada por la
abundancia de excretas, y S es el porcentaje de excretas con semillas. B se calculó
dividiendo el número de excretas de cada mes entre el número total de excretas registradas
durante todo el período de muestreo y multiplicado por 100. S se calculó dividiendo el
número de excretas con presencia de semillas, entre B y multiplicado por 100. El índice IID
tiene un rango entre 0 y 10, donde cero indica ausencia total de excretas con semillas y 10
indica la presencia de todas las semillas registradas (Galindo-González et al. 2000). Este
índice proporciona un valor de importancia del dispersor para cada mes. En agosto de 2012
no se pudo realizar el muestreo de excretas, por tanto, para este análisis se excluyeron los
datos de septiembre de 2012, que tenían registros acumulados de dos meses.
Para evaluar la diferencia en el patrón de germinación entre semillas procedentes de las
excretas y semillas testigo se utilizó un test de supervivencia, el cual comprueba si hay una
diferencia entre dos o más curvas de supervivencia (en este caso, germinación a partir de
excretas vs. testigo) desarrolladas mediante un análisis Kaplan-Meier. Esta función
implementa la familia G-rho de Harrington y Fleming (1982), con los pesos en cada
germinación de S (t) ^ rho, donde S es la estimación de Kaplan-Meier de supervivencia en
el tiempo t (Harrington y Fleming 1982).
150
Las diferencias en la viabilidad de las semillas presentes en las excretas y el testigo, se
analizó mediante un test ANOVA (Sokal y Rohlf 1981). La viabilidad se calculó para cada
especie como la suma del número de semillas germinadas durante los primeros 90 días, más
el número de semillas germinadas luego de la escarificación.
RESULTADOS
Durante los 23 meses de muestreo se registraron en total 386 excretas de venado. La
presencia de excretas se observó desde finales de la estación lluviosa hasta finales de la
estación seca, con abundancias de entre cuatro y 87 excretas por mes. En los meses
lluviosos la presencia de excretas fue nula (Tabla 1).
Riqueza de especies y abundancia de semillas dispersadas por venado- De las 386
excretas de venado registradas, el 65,3% contenían semillas. El número de semillas por
excreta varió de una a 54 semillas, con un promedio de 7,8 ± 0,62 semillas/excreta (Tabla
1). Todas las semillas registradas estuvieron completas y sin daño aparente en la testa.
En total se identificaron semillas de 11 especies leñosas, correspondientes al menos a siete
familias (Tabla 2). La especie más abundante en las excretas fue Chloroleucon mangense.
Otras tres especies abundantes fueron Caesalpinia glabrata, Senna mollisima y Piptadenia
flava (Tabla 2). Estas cuatro especies tienen frutos tipo legumbre y presentan una pulpa
seca fibrosa. Otras especies menos frecuentes en las excretas de venado comparten estas
características (Leucaena trichodes y Vigna sp.), aunque también se registraron semillas de
una Convolvulaceae, con una cápsula seca y dehiscente y tres bayas (Cactaceae, Randia
aurantiaca y Jacquinia sprucei).
151
Tabla 1. Registro mensual de semillas presentes en las excretas de venado.
Mes N°
excretas
N° excretas
con semillas
Riqueza de
especies
N° total de
semillas
Promedio
semillas / excreta
± error típico
oct-11 13 9 4 34 3,78 ± 1,30
nov-11 40 30 5 398 13,27 ± 2,76
dic-11 24 24 4 291 12,13 ± 2,92
ene-12 4 4 3 76 19,00 ± 9,60
feb-12 0
mar-12 0
abr-12 0
may-12 0
jun-12 8 6 6 13 2,17 ± 0,31
jul-12 19 6 5 12 2,00 ± 0,63
sep-12 87 72 7 641 8,90 ± 1,01
oct-12 36 31 4 135 4,35 ± 0,84
nov-12 24 21 4 138 6,57 ± 1,99
dic-12 11 8 5 44 5,50 ± 2,05
ene-13 9 5 2 24 4,80 ± 3,07
feb-13 0
mar-13 0
abr-13 0
may-13 18 3 2 3 1,00 ± 0,00
jun-13 26 2 2 3 1,50 ± 0,50
jul-13 30 13 3 62 4,77 ± 1,60
ago-13 16 9 1 39 4,33 ± 0,67
sep-13 21 9 4 48 5,33 ± 1,45
Total 386 252 11 1961 7,78 ± 0,62
152
La importancia del venado como dispersor de semillas de las 11 especies de leñosas fue
variable a lo largo del año (Fig. 1). El número de excretas con semillas más bajo se registró
durante la estación lluviosa, entre enero y mayo. A partir de junio el número de semillas por
excreta aumentó hasta alcanzar un máximo entre octubre y noviembre (Fig. 1).
Figura 1. Índice de importancia del dispersor estimado mensualmente. Las áreas grises señalan los
meses de la estación lluviosa.
Germinación de semillas- De las once especies encontradas en las excretas de venado
ocho especies germinaron a partir de 67 de las 252 excretas que presentaron semillas. En
total germinaron 172 semillas (8,8% de las semillas presentes), de las cuales el 79%
correspondió a C. mangense. El número de semillas germinadas por excreta varió entre
cero y 14 semillas (promedio de 0,7 + 0,1 semillas/excreta), procedentes de una a cinco
especies.
153
Tabla 2. Especies de plantas presentes en las excretas de venado, porcentaje de germinación total y
porcentaje de germinación medio por excreta ± error típico.
Familia Especie
No. Semillas
presentes
No. Semillas
germinadas
Germinación
(%)
Germinación
por excreta
Fabaceae Chloroleucon mangense 1183 136 11,5 10 ± 1,66
Vigna sp 16 3 18,8 18 ± 12,20
Caesalpiniaceae Caesalpinia glabrata 75 6 8,0 16 ± 6,52
Senna mollissima 544 13 1,8 4 ± 1,81
Mimosacee Piptadenia flava 104 11 10,6 14 ± 6,35
Leucaena trichodes 12 0 0,0 0 ± 0,00
Convolvulaceae Convolvulaceae 10 0 0,0 0 ± 0,00
Rubiaceae Randia aurantiaca 7 0 0 0 ± 0,00
Theophrastaceae Jacquinia sprucei 1 1 100,0 100 ± 0
Cactaceae Cactaceae 1 1 100,0 100 ± 0
Desconocida 8 1 12,5 14 ± 14,29
En general la proporción de semillas germinadas a partir de las excretas fue baja. Los
porcentajes de germinación por especie variaron entre 8 y 19%. Solo para dos especies (R.
aurantiaca y J. sprucei) que presentaron una sola semilla, el porcentaje de germinación fue
100% (Tabla 2). Por otro lado, mientras que para C. mangense y Vigna sp el promedio de
154
germinación por excreta fue similar al porcentaje total de germinación, para el resto de
especies el mayor porcentaje de germinación estuvo restringido a unas pocas excretas
(Tabla 2).
Efectos de la ingestión de semillas por venado en la germinación- En tres de las cuatro
especies evaluadas; C. mangense, S. mollissima y L. trichodes, los porcentajes de
germinación a partir de los testigos también fueron bajos. Solo en semillas testigo de C.
glabrata se registró un alto porcentaje de germinación en comparación con las semillas
procedentes de las excretas (Tabla 3).
Las diferencias en el patrón de germinación entre semillas de las excretas y semillas testigo
pudieron testarse solo para C. mangense, que fue la especie más abundante en las excretas.
El resultado del test de sobrevivencia
no señaló diferencias significativas en la germinación
de los dos grupos de semillas. (2= 0,2; p= 0,667). El porcentaje de germinación de
semillas de C. mangense fue 13,6 % ± 4,3 para las semillas procedentes de excretas, y 14 %
± 2 para las semillas testigo.
Tabla 3. Porcentaje total de germinación de semillas a partir de las excretas y del testigo de cuatro
especies.
Especie
Germinación
Excretas Testigo
Chloroleucon mangense 11,5 14 + 2
Senna mollissima 2,4 5 + 1,91
Caesalpinia glabrata 8,0 90 + 2,58
Leucaena trichodes 0,0 18 + 2
155
Luego de los 90 días de siembra, las semillas no germinadas de C. mangense, procedentes
tanto de las excretas como del testigo, fueron utilizadas para evaluar el estado de viabilidad.
A los 15 días a partir de la escarificación germinaron 78 semillas. El número de semillas
viables por excreta varió entre 0 y 10 semillas, alcanzando en promedio 61,7% ± 6,41 de
viabilidad / excreta. Para los testigos se obtuvo un promedio de viabilidad de 81% ± 6. El
test ANOVA no indicó diferencias significativas en la viabilidad de semillas entre excretas
y testigo (F = 1,294, DF = 33, p = 0,264).
DISCUSIÓN
Importancia de los cérvidos como dispersores de semillas- Durante todo el período de
muestreo se registró la dispersión de semillas de 11 especies a través de las excretas de
venado. Las semillas de estas especies estuvieron presentes en el 65% de las excretas
muestreadas. Todas las especies presentaron semillas de características similares; pequeñas
y de testa dura, a excepción de las semillas de Cactaceae que poseen una testa blanda. Estas
características coinciden con lo propuesto por Janzen (1982) para semillas adaptadas a
endozoocoria. Además, las características de los frutos, como la pulpa fibrosa seca, la
fuerte protección de la semilla, el color opaco, coinciden con las características propuestas
por Gauthier-Hion et al. (1984), para frutos dispersados por ungulados.
Las 10 especies que pudieron identificarse están ampliamente distribuidas en los bosques
secos del suroccidente Ecuatoriano (Aguirre et al. 2006). Las semillas de C. mangense
fueron las más abundantes. Esto puede deberse a que en julio y agosto, que son los meses
de mayor producción de frutos de C. mangense, la abundancia de frutos de las otras
especies presentes en las excretas fue baja (Jara-Guerrero datos no publicados). Por otro
156
lado, el número de semillas por fruto en C. mangense fue mayor que en las otras especies
(unas 20 semillas por vaina) (Obs. Pers.), lo que aumenta las posibilidades de encontrar un
mayor número de semillas en una misma excreta.
Las vainas de C. mangense, S. mollissima, C. glabrata y L. trichodes se observaron en gran
cantidad, y en buen estado, en el suelo del bosque por periodos mayores a un mes, por lo
que constituyen un recurso frecuente durante los meses más secos. Por otro lado, la
presencia de semillas de Cactaceae es consistente con lo reportado en zonas áridas de
México, en donde ha determinado la dispersión de semillas viables del género Opuntia
luego del paso por el tracto digestivo de O. virginianus (González-Espinosa y Quintana-
Ascencio 1986, Mandujano et al. 1997).
La importancia de los venados como dispersores de semillas varió a lo largo del año, con
mayores niveles de importancia durante los meses de la estación seca (Fig. 1). Existe
información acerca del cambio en el uso del hábitat por O. virginianus en bosques secos de
México, relacionados en parte con la disponibilidad de alimento (Mandujano et al. 2004,
Mandujano 2006), lo que podría explicar esta variación y la ausencia de registros de
excretas en la estación lluviosa. Sin embargo, la falta de información acerca de la biología
de los cérvidos en los bosques secos tumbesinos y del uso del hábitat, no permiten tener
resultados concluyentes en este caso. Además, hay que tener en cuenta que durante la
estación lluviosa la densidad de vegetación dificulta la observación, por lo que podría haber
un efecto de muestreo. Además, está la posibilidad de una degradación más rápida de las
excretas.
157
Efectos de la ingestión de semillas por venado sobre la capacidad germinativa- De las
once especies presentes en las excretas germinaron ocho especies. A pesar de que todas las
semillas presentaron un buen estado, los porcentajes de germinación fueron bajos en todas
las especies (< 20%). Sin embargo, hay que considerar que en tres de las cuatro especies
para las que sembró un testigo, el porcentaje de germinación del testigo fue igualmente
bajo. Para el caso particular de C. mangense se pudo determinar que el paso por el tracto
digestivo del venado no afectó de germinación de semillas; incluso luego de aplicar el
proceso de escarificación, el tiempo de germinación y el porcentaje de germinación
alcanzado en las semillas de las excretas y testigo fueron similares. Estudios previos
señalan que los ciervos, particularmente O. virginianus en América, son capaces de
dispersar semillas viables en otros ecosistemas, principalmente hierbas y especies exóticas
(Vellend et al. 2003, Mouissie et al. 2005, Williams et al. 2008), pero también de especies
nativas (González-Espinosa y Quintana-Ascencio 1986, Mandujano et al. 1997), incluidas
leñosas (Janzen 1982, Williams et al. 2008). Nuestros resultados respaldan las
observaciones de dichos estudios y demuestran que los cérvidos también son capaces de
dispersar por endozoocoria semillas viables de especies leñosas de los bosques secos del
suroccidente de Ecuador.
Por otro lado, el hecho de que la ingestión de semillas por venado no tuviera efectos sobre
la cubierta de las semillas es una ventaja para estas especies, considerando que son
dispersadas durante la estación seca. El conservar una testa dura e impermeable permite a
las semillas soportar las condiciones de sequía y mantener la viabilidad hasta el inicio de
las lluvias (Mandujano et al. 1997), cuando la disponibilidad de agua es adecuada para la
germinación y desarrollo de plántulas (Khurana y Singh 2001). Esto implica que el rol del
158
venado en la dispersión de las semillas se centra en el movimiento de las semillas viables
lejos de la planta madre, sin afectar su capacidad germinativa, por tanto, se puede
considerar un dispersor legítimo de semillas (Reid 1989, Herrera 1989).
En conclusión, los resultados del presente trabajo ayudan a estimar la habilidad de las
plantas leñosas para dispersar sus semillas vía endozoócora a través de los cérvidos. La
principal función de los cérvidos se centró en el movimiento de las semillas lejos de la
planta madre, sin afectar su capacidad germinativa. La morfología de los frutos y semillas
de las especies más abundantes en las excretas las define como especies “autocoras”
(Leguminosas, Tabla 2); esta observación sugiere que los cérvidos pueden ser vectores
clave en la dispersión de semillas sin adaptaciones obvias para la dispersión, con las
respectivas implicaciones en la colonización de nuevos sitios y en el mantenimiento del
flujo genético.
Este trabajo es un primer aporte al conocimiento del rol de cérvidos en la dispersión de
semillas de los bosques secos de la región Tumbesina. Las conclusiones de esta
investigación emergen del análisis de la diversidad de especies y cantidad de semillas
dispersadas, de los efectos de la ingestión de semillas sobre la germinación y del análisis
del tiempo en el que se dan los eventos de dispersión. Sin embargo, aún es necesario un
trabajo más profundo para mejorar el entendimiento de la importancia de los cérvidos en la
regeneración de estas especies. Un análisis de los patrones espaciales de deposición de las
semillas y sus posibles efectos sobre la sobrevivencia de las plántulas podría dar nuevas
luces este sentido.
159
REFERENCIAS
Aguirre, Z. y Kvist, L.P. 2005. Floristic composition and conservation status of the dry
forests in Ecuador. Lyonia 8: 41-67.
Aguirre, Z., R. Linares-Palomino, y L. Kvist. 2006. Especies leñosas y formaciones
vegetales en los bosques estacionalmente secos de Ecuador y Perú. Arnaldoa 13: 324–
350.
BirdLife International. 2014. Important Bird Areas factsheet: Reserva Ecológica Arenillas.
Downloaded from http://www.birdlife.org on 07/10/2014
Bodmer, R. 1991. Strategies of seed dispersal and seed predation in amazonian ungulates.
Biotropica 23: 255-261.
Bullock, S. H. 1995. Plant reproduction in neotropical dry forests. En S., Bullock, H.,
Mooney y E. Medina (Eds.). Seasonally dry tropical forests. Cambridge University
Press, Cambridge, UK. pp. 277 – 297.
Cain, M.L., Milligan, B.G. y Strand, A.E. 2000. Long-distance seed dispersal in plant
populations. Am. J. Bot. 87: 1217 – 1227.
Eisenberg, J. F. y Redford, K. H. 1999. Mammals of the neotropics. Vol. 3. The Central
neotropics: Ecuador, Perú, Bolivia, Brasil. The University of Chicago Press. Chicago
609.
160
Espinosa, C.I., De La Cruz, M., Luzuriaga L., y Escudero, A. 2012. Bosques tropicales
secos de la región Pacífico Ecuatorial: diversidad, estructura, funcionamiento e
implicaciones para la conservación. Ecosistemas 21 (1-2):167-179.
Fleming, T. H. 1986. Opportunism versus specialization: the evolution of feeding strategies
in frugivorous bats. En: Estrada, A. y Fleming, T. H. (eds). Frugivores and seed
dispersal. Junk, Dordrecht, The Netherlands, pp. 105-118.
Fulbright and Ortega 2006. White-tailed deer habitat: ecology and management on
rangelands. - United States, 256 pp.
Galindo-González, J., Guevara, S., y Sosa, V. 2000. Bat- and bird-generated seed rains at
isolated trees in pastures in a Tropical Rainforest. Conserv. Biol. 14: 1693-1703.
Gautier-Hion, A., Duplantier, JM., Quris, R., Feer, F., Sourd, C., Decoux, JP., Doubost, G.,
Emmons, L., Erard, C., Hecketsweiler, P., Moungazi, A., Rossilhon, C. y Thiollay,
J.M. 1985. Fruit characters as a basis of fruit choice and seed dispersal in a tropical
forest vertebrate community. Oecologia 65: 324-337.
González-Espinosa, M. y Quintanta-Ascencio, P. 1986. Seed predation and dispersal in a
dominant desert plant: Opuntia, ants, birds, and mammals. En: Estrada, A. y Fleming,
T.H. (eds.) Frugivores and seed dispersal. Junk Publishers, Dordrecht.
Harrington, D. P. y Fleming, T. R. 1982. A class of rank test procedures for censored
survival data. Biometrika 69: 553-566.
161
Herrera, C. M. 1989. Frugivory and seed dispersal by carnivorous mammals, and associated
fruit characteristics in undisturbed Mediterranean habitats. Oikos 55: 250–62.
Howe, H.F., y Smallwood, J. 1982. Ecology of seed dispersal. Annu. Rev. Ecol. Syst. 13:
201-228.
Janzen, D.H. 1982. Dispersal of small seeds by big herbivores: foliage is the fruit. Am. Nat.
123, 338-353.
Jara-Guerrero, A., Méndez, M., y De La Cruz, M. 2011. Seed Dispersal Spectrum of
Woody Species in South Ecuadorian Dry Forests: Environmental Correlates and the
Effect of Considering Species Abundance. Biotropica: 43:722-730.
Jordano, P. 2000. Fruits and frugivory. In M. Fenner (Ed.). Seeds: The Ecology of
Regeneration in Plant Communities (2nd edition), pp. 125–165. CABI Publishing,
Wallingford, UK.
Khurana, E., y Singh, J. S. 2001. Ecology of seed and seedling growth for conservation and
restoration of tropical dry forest: a review. Environ. Conserv. 28: 39–52.
Link, A., y Stevenson, P.R. 2004. Fruit dispersal syndromes in animal disseminated plants
at Tinigua National Park, Colombia. Rev. Chil. Hist. Nat 77: 319-334.
Malo, J.E. y Suárez, F. 1996. Cistus ladanifer recruitment - Not only fire, but also deer.
Acta Oecol. 17:55-60.
Malo, J.E. y Suárez, F. 2000. Herbivore dunging and endozoochorous seed deposition in a
Mediterranean dehesa. J. Range Manag 53:322-328.
162
Mandujano, M.C., Golubov, J. y Montaña, C. 1997. Dormancy and endozoochorous
dispersal of Opuntia rastrera seeds in the southern Chihuahuan Desert. J. Arid
Environ. 36: 259–266.
Mandujano, S., Gallina, S., Arceo, G. y Pérez-Jiménez, L.A. 2004. Variación estacional
del uso y preferencia de los tipos vegetacionales por el venado cola blanca en un
bosque tropical de Jalisco. Acta Zool. Mex. 20: 45-67.
Mandujano, S. 2006. Preliminary evidence of the importance of ENSO in modifying food
availabilityfor white-tailed deer in a Mexican tropical dry forest. Biotropica 38: 695–
699.
Mouissie, A.M., Van der Veen, C.E.J., Veen, G.F., VAN DIGGELEN, R. 2005. Ecological
correlates of seed survival after ingestion by Fallow Deer. Funct Ecol 19: 284–290.
Myers, J.A., Vellend, M., Gardescu, S. y Marks, P. L. 2004. Seed dispersal by white-tailed
deer: implications for long-distance dispersal, invasion, and migration of plants in
eastern North America. Oecologia 139: 35 – 44.
Programa de Naciones Unidas para el Medio Ambiente (PNUMA). 2010. América Latina y
El Caribe atlas de un ambiente en transformación. pp. 312.
Reid, N. 1989. Dispersal of mistletoes by honeyeaters and flowerpeckers: components of
seed dispersal quality. Ecology 70: 137–45.
Sierra, R. 1999. Propuesta preliminar de un Sistema de Clasificación de Vegetación para el
Ecuador Continental. INEFAN / GEF – BIRF y Ecociencia, Quito, Ecuador.
163
Silva, S., Bozinovic, F., y Jaksic, F. 2005. Frugivory and seed dispersal by foxes in relation
to mammalian prey abundance in a semiarid thornscrub. Austral Ecol. 30: 739–746.
Sokal,R. R. y Rohlf, F. J. 1981. Biometría. H. Blume, Madrid.
Stiles, E. W. 2000. Animals as seed dispersers. En: Fenner, M. (Ed.). Seeds: the ecology of
regeneration in plant communities. 2nd edition. CAB International.
Tirira, D. 2007. Guía de campo de mamíferos del Ecuador. Ed. Murciélago Blanco.
Publicación especial de los Mamíferos del Ecuador 6. Quito pp. 576.
Traveset, A. 1998. Effect of seed passage through vertebrate frugivores’ guts on
germination: a review. Perspect. Plant Ecol. Evol. Syst. 1: 151-190.
Vellend, M., Myers, J.A., Gardescu, S. y Marks, P. L. 2003. Dispersal of Trillium seeds by
deer: Implications for long-distance migration of forest herbs. Ecology 84: 1067 –
1072.
Williams, S. C., Ward, J. S., Ramakrishnan, U. 2008. Endozoochory by white-tailed deer
(Odocoileus virginianus) across a suburban/woodland interface. Forest Ecol. Manag.
255: 940–947.
164
165
CONCLUSIONES GENERALES
Los resultados obtenidos en la presente investigación han permitido dilucidar algunos
patrones de dispersión de semillas de los bosques tropicales estacionalmente secos, a partir
de los cuales se extraen las siguientes conclusiones generales:
Los bosques secos tropicales del suroccidente de Ecuador sostienen una gran
variedad de plantas que requieren la asistencia de animales para la dispersión. Si
bien estos resultados son consistentes con lo propuesto para los bosques secos
tropicales, un análisis del espectro de dispersión considerando no solo la riqueza,
sino también la abundancia relativa de especies, permitió determinar que a pesar de
la alta variedad de especies zoocoras, la mayor parte de la comunidad corresponde a
individuos anemócoros, que no proveen ninguna recompensa para la dispersión por
animales. Por otro lado, los resultados demuestran con datos cuantitativos, y a
través de una relación directa, que las condiciones ambientales afectan la estructura
del espectro de dispersión en la comunidad de bosque seco neotropical estudiada.
Estos hallazgos son muy interesantes tanto desde el punto de vista funcional de las
interacciones de dispersión, como para el manejo y conservación de estos bosques.
Se utilizó un nuevo enfoque metodológico para analizar la importancia relativa del
síndrome de dispersión y de la heterogeneidad espacial en la formación de patrones
espaciales de árboles adultos de un bosque tropical estacionalmente. Este método
permitió sugerir que la heterogeneidad ambiental ejerce un efecto adicional (y en
166
algunos el único) limitando la distribución de la mayoría de especies de la
comunidad estudiada. Los resultados señalan diferencias en los patrones espaciales
de las especies dependiendo del síndrome de dispersión, pero también revelan una
gran variación en los patrones espaciales incluso entre especies del mismo síndrome
de dispersión. Este enfoque puede dar nuevos indicios acerca de los procesos que
determinan la coexistencia de especies a escalas locales y aporta al debate acerca de
la prevalencia de diferentes mecanismos en la estructuración de comunidades de
bosque tropical estacionalmente seco.
El análisis simultáneo de los patrones de la lluvia de semillas y banco de semillas de
una comunidad y su relación con la vegetación establecida, sugieren que la lluvia de
semillas de especies leñosas en el bosque seco de la REA es temporalmente variable
en número de especies y abundancia de semillas, y depende del síndrome de
dispersión. El síndrome de dispersión también influyó en la formación de bancos de
semillas, siendo las especies con capacidad de dispersión limitada (autócoras) las de
mayor riqueza de especies y abundancia de semillas. El estudio en conjunto de la
lluvia de semillas y el banco del suelo nos permitió proponer que si bien la mayoría
de especies leñosas dependieron de la lluvia de semillas para su regeneración, como
se propone en general para las comunidades forestales en estado sucesional
avanzado, en el bosque seco estudiado el banco de semillas puede ser un aporte
efectivo para la regeneración de algunas especies, principalmente para aquellas con
dispersión autocora.
167
El estudio del rol que cumplen los cérvidos en la regeneración de especies leñosas
del bosque seco permite concluir que estos son capaces de dispersar por
endozoocoria semillas viables, y son dispersores legítimos de al menos ocho
especies. La mayoría de las especies dispersadas presentaron diásporas sin
adaptaciones obvias para la dispersión, por lo que la ingestión de semillas por
cérvidos se constituye en una vía potencial para la dispersión a largas distancias y,
con ello, la posibilidad de colonizar nuevos sitios y mantener el flujo genético.
