1 Host tree dispersal method and community composition of vascular epiphytes Kathleen White Department of Biology, Occidental College ABSTRACT The pattern of epiphyte distribution along host trees is a function of many variables. This study examined ho w the dispersal syndrome of the host tree affected the composition of epiphyte dispersal syndromes in that tree. It was expected that a host tree species bearing bird dispersed fruits would harbor more epiphytes exhibiting the bird dispersal syndrome than a tree that has fruits eaten by terrestrial mammals. Two species of host trees were surveyed, Inga sierrae a tree with terrestrial mammal dispersed fruits, and Sapium glandulosum, a tree with fruits that are bird dispersed. Nine of each tree species wer e sampled. The tree with terrestrial mammal dispersed fruits contained significantly higher numbers of epiphytes whose fruits are dispersed by birds, while there was no significant difference in numbers of epiphytes with wind dispersed seeds between the tw o species. This unexpected result may be due to the differing architecture between the two tree species studied. RESUMEN Las epfitas se distribuyen en los rboles anfitrin en patrones que varan de acuerdo a muchas variables. Este estudio examin cmo el sndrome de dispersin del rbol anfitrin afect la composicin de los sndromes de dispersin del epfitas. Se esperaba que una especie anfitrin de rboles dispersada por pjaros abrigara ms epfitas dispersadas por pjaros. Dos especies de rboles anfitrin fueron examinadas, Inga sierrae un rbol dispersado por mamferos terrestres, y Sapium glandulosum un rbol dispersado por pjaros. Nueve individuos de cada especie fueron muestreados. Los rboles dispersados por mamferos terrestres contuvier on cantidades perceptiblemente ms altas de epfitas dispersadas por pjaros. Mientras que no hubo una diferencia significativa en las epfitas dispersadas por el viento entre las dos especies. Este resultado inesperado puede ser debido a las diferentes a rquitecturas entre las dos especies estudiadas. INTRODUCTION Epiphytes, non parasitic plants that anchor themselves in trees, comprise ten percent of all plant species (Kress 1986). They vary greatly in abundance and diversity across all spatial scales, but their patterns of distribution are still not completely understood (Richards 1996). The vertical and horizontal distribution of epiphytes within a host tree have been examined as a function of many variables. A complex set of factors including alti tude, humidity (Sugden and Robins 1979; ter Steege and Cornelisson 1989), humus type, light intensity, water availability, exposure to the elements (Johansson 1975), bark type, tree architecture, branch diameter (Zimmerman and Olmstead 1992), host tree age and species (Catling and Lefkovitch 1989), and epiphyte dispersal syndrome (Edwards 1997) have all been investigated for possible effects on epiphyte distribution. However, how the dispersal method of the host tree affects the community composition of epi phytes has not been previously investigated. A tree may attract different animal species according to the type of fruit it provides, and these animals may influence the
2 composition of epiphytes in the host tree as they deposit seeds. If this is correct, t hen trees with bird dispersed fruits would exhibit a higher proportion of epiphytes with the same dispersal syndrome due to increased visitation and seed addition by birds. Conversely, host trees producing terrestrial mammal dispersed fruits would be expe cted to have a smaller proportion of epiphytes with bird dispersed fruits than tree yielding fruits eaten by birds. Finally, the relative proportion of epiphytes producing seeds dispersed by the wind would be expected to remain constant between both speci es, as they do not require an animal intermediary for dispersal. In examining this relationship, it was assumed that there were no differences in seed germination rates between species, and therefore the number of epiphytes found on a tree accurately conve y the amount of seeds that arrive on that tree. This study examined whether host tree dispersal syndrome affects the proportions of bird and wind dispersed epiphytes that inhabit the host. To accomplish this, the composition of epiphyte communities were c ompared in tree species producing either bird or terrestrial mammal dispersed fruits. MATERIALS AND METHODS This study was carried out in the pasture behind the Estacin Biolgica (elevation of 1550 m) in Monteverde, Costa Rica during October and Novembe r of 2004. The epiphytic composition of two tree species, Sapium glandulosum (Euphorbiaceae) and Inga sierrae (Mimosaceae), were compared. The fruits of S. glandulosum are dehiscent capsules with conspicuous red arils. Twenty two species of birds have bee n observed feeding on these fruits. Inga sierrae produces indehiscent woody pods containing seeds wrapped in a white aril. The aril is eaten by mammals, which presumably disperse the seeds (Haber et al. 2000). Nine trees of each species were sampled. T he number of individual epiphytes per family were censused in three, 4,800 cm 2 areas (1975) five Zones (Fig. 1). Zone Two was omitted because it was not present the m ain trunk of both species was usually too short for it to contain this zone. Zone Five was not incorporated because of inaccessibility. Trees were surveyed using both free climbing and ground observation with binoculars. The epiphytic taxa included were: F. Ericaceae (bird dispersed fruits), F. Gesneriaceae (bird), F. Piperaceae (bird), F. Bromeliaceae (wind), D. Pteridophyta (wind), and F. Orchidaceae (wind) (Haber 2000). The number of individuals in per taxa were recorded for each tree, in each zone. To determine the difference between the average number of epiphytic dispersal syndromes between the two tree species, the Mann Whitney U test was applied. The Friedman Test was also used to determine whether average numbers of individuals exhibiting eac h dispersal syndrome differed between zones regardless of species. Table 1 : Chi squared conting ency table illustrating differences in epiphyte dispersal type between saprophytes species.
3 RESULTS Inga sierrae and S. glandulosum did not differ significantly in average total number of epiphytes surveyed (Mann Whitney U test U = 41, p = 0.965). Although S. glandulosum had a higher average of epiphytes with wind dispersed seeds, this average was not significantly different from I. sierrae (Mann Whitney U test U = 48, p = 0.508). However, the two species did differ in average number of epiphytes with bird dispersed fruits (Ma nn Whitney U test U = 63, p = 0.0443) (Table 1). A higher occurrence of bird dispersed epiphytes were observed in I. sierrae (Fig. 2). Epiphytes with both wind dispersed fruits and bird dispersed fruits were found to be similarly distributed between both host species across the three Zones sampled (Mann Whitney U test, Table 2). Even so, epiphytes eaten by birds tended to favor Zone One of I. sierrae (Table 3), although the trend was not significant ( Fig. 3). Regardless of host tree species, Zones differ ed in the total number of epiphytes censused (Freidman test N = 18, p = 0.0024), with Zone Four exhibiting the highest epiphyte abundance (38.7%). Epiphytes with wind and bird dispersed fruits were both distributed unevenly between zones (Friedman test N = 18, p = 0.0008; N = 18, p = 0.0263). Epiphytes producing wind dispersed seeds were most common in Zone Four, while numbers of epiphytes with fruits eaten by birds were greatest in Zone One. Proportions of epiphytes exhibiting the bird dispersal syndrome d ecreased with higher zones (see Figure 4). DISCUSSION Differences in epiphyte dispersal syndrome between species The data do not support the hypothesis that host trees with bird dispersed fruits contain a larger proportion of epiphytes that bear bird dispersed fruits. On the contrary, the tree with terrestrial mammal dispersed fruits, I. sierrae, exhibited a larger proportion of epiphytes with fruits eaten by birds than did S. glandulosum the species bearing bird dispersed fruits. Inga sierrae and S. glandulosum did not have significantly different numbers of wind dispersed epiphytes, which was congruent with expectations. The discrepancy between expected and observed results may be a result of differing architecture between the two species studied. Inga sierrae is more widely branched and its crown is dense and protective. Sapium glandulosum, in contrast, usually has a more vertical growth form, and its foliage is not as dense. These differences may make I. sierrae a more desirable perching tree t han S. glandulosum Epiphytic bird dispersed fruits are quite small, and they are only eaten by small to medium sized birds. These birds are extremely vulnerable to predation in the open pasture where they may easily be seen. Sheltered areas have been s hown to reduced predation and mortality in birds (Cuadrado 1997). It is probable that these birds only visit S. glandulosum to feed, and travel to other trees that are more protective, such as I. sierrae, for general perching. The
4 effect of host tree dis persal type on epiphyte dispersal type, if any, may not be large enough to exhibit any differences because of these variations in architecture. As expected, epiphytes that generate wind dispersed seeds did not differ in their distribution between species This is probably due to the random and abiotic nature of wind dispersal. This dispersal method is not dependent on animals, and for this reason it should not vary between tree species that attract a differing suites of animals. The data support this h ypothesis. The assumption made that there are equal germination rates in epiphytes between the species studied should not affect the reliability of these results. Substrate qualities such as bark type and pH, in addition to numerous other factors determin e which epiphytes can actually grow on a tree (Johansson 1975). However, there is no evidence to believe that these factors are at work in preventing the germination of epiphytes on the two species studied. Both species exhibited an abundant growth of epi phytes and are not known to deter epiphytes. Even so, if these factors are present, they would also offer an alternate explanation for the difference in epiphytes with bird dispersed fruits between I. sierrae and S. glandulosum. Further studies should st rive to minimize these effects by incorporating many species of trees in each dispersal syndrome category and increasing sample sizes. Differences in epiphyte dispersal syndrome between Zones Epiphyte dispersal syndromes did not vary between host tree s pecies in each zone censused. This suggests that there is an overarching pattern in distribution of epiphyte dispersal syndromes within host trees, regardless of species. For both tree species, epiphytes that bear wind dispersed seeds were mostly prevalent in Zone Four, and epiphytes whose fruits are eaten by birds were most common in Zone One. These epiphytes with bird dispersed fruits were most abundant in Zones One, Two and Three, and epiphytes with wind dispersed seeds dominated Zones Four and Five. This study, however, did not demonstrate that epiphytes bearing fruits eaten by birds dominated Zone Three as Edwards (1997) found. This inconsistency may be due to diffe ring methods. Edwards surveyed predominately the crotch of the tree in Zone Three, while this study only included epiphytes positioned on branches in this Zone. This was done in order to minimize physical differences (e.g. substrate, microclimate) betwee n Zones. The crotch area is optimal for collecting falling debris such as leaves or seeds. Epiphytic bird dispersed fruits dominate this area because it gathers seeds that fall from above. Wind dispersed seeds are usually laterally distributed and theref ore do not accumulate in this area as well as bird dispersed seeds. Since the distribution of epiphytes has been shown to be affected by many variables, the differences between Z ones described previously could be due to variables that were not measured in this study. Seeds of a certain species or group may be able to germinate and grow better in a Zone due to exposure, sunlight and other factors, rather than as a result
5 of where they initially fall. This is another explanation for differences in epiphy te dispersal syndromes between Zones. It is worth mentioning that numbers of D. Pteridophyta, a contributor to the wind dispersal category, may be slightly inflated. It is often difficult to determine whether a given group of plants is one individual or many, and in which Zone its seed was originally deposited. Climbing ferns that spanned Zones may have been counted multiple times. Even so, this inflation is probably marginal and should not affect the reliability of these results. Epiphytes account fo r a significant proportion of the plant kingdom. In the forests of Monteverde, epiphytes are conspicuous and are an integral part of ecological processes (Hartshorn 1983). Further understanding of the dynamics of epiphyte communities is integral to the un derstanding the ecology of the Monteverde Cloud Forest, and other forests worldwide. ACKNOWLEDGEMENTS First and foremost, many thanks to Karen Masters for her patient guidance in every level of the development of this project, and to Alan Masters for he lping me form ideas. Thanks to Javier Mendez for help with Spanish translations. Thanks to Matt Gasner and Ollie Hyman for assistance with statistical analysis, technical difficulties, and ground support while climbing in fear of attack by howler monkeys Even though I ended up not using rope climbing techniques, thanks to Rigo for sacrificing his time to come and teach us. Lastly thanks to the Estacin Biolgica de Monteverde for permission to use their beautiful property. LITERATURE CITED Catling, P. 40. Erithracus rebecula ) territorial in winter? The importance of anti predatory behavior. 78. Ficus tuerckheimii 1997. Monteverde ed. Nalini M. Nad karni and Nathaniel T. Wheelright. Oxford University Press: Oxford. Haber, W.A., W. Zuchowski, and E. Bello. 2000 An Introduction to Cloud Forest Trees in Monteverde, Costa Rica 2 nd ed.Mountain Gem Publications: Monteverde de Puntarenas. Hartshorn, G.S. Costa Rican Natural History ed. Daniel H. Janzen. The University of Chicago Press: Chicago. Society Bulletin. 44: 125 135. 22. Richards, P.W. 1996. The Tropical Rain Forest 2 nd ed. University of Cambridge Press: Cambridge. gy of vascular epiphytes in Colombian cloud 188. rain forest of Guyana 339. 24(3): 402 407.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00372
Mtodo de dispersin en los arboles anfitriones y composicin de la comunidad de epifitas vasculares
Host tree dispersal method and community composition of vascular epiphytes
The pattern of epiphyte distribution along host trees is a function of many variables. This study examined how the dispersal syndrome of the host tree affected the composition of epiphyte dispersal syndromes in that tree. It was expected that a host tree species bearing bird-dispersed fruits would harbor more epiphytes exhibiting the bird-dispersal syndrome than a tree that has fruits eaten by terrestrial mammals. Two species of host trees were surveyed, Inga sierrae, a tree with terrestrial-mammal dispersed fruits, and Sapium glandulosum, a tree with fruits that are bird-dispersed. Nine of each tree species were sampled. The tree with terrestrial mammal-dispersed fruits contained significantly higher numbers of epiphytes whose fruits are dispersed by birds, while there was no significant difference in numbers of epiphytes with wind- dispersed seeds between the two species. This unexpected result may be due to the differing architecture between the two tree species studied.
Las epfitas se distribuyen en los rboles anfitriones en patrones que varan de acuerdo a muchas variables. Este estudio examin cmo el sndrome de dispersin del rbol anfitrin afect la composicin de los sndromes de dispersin de las epfitas. Se esperaba que una especie de anfitrin de rboles dispersada por las aves abrigara ms epfitas dispersadas por las aves. Dos especies de rboles anfitrin fueron examinadas, Inga sierrae, un rbol dispersado por mamferos terrestres, y Sapium glandulosum, un rbol dispersado por aves. Nueve individuos de cada especie fueron muestreados. Los rboles dispersados por mamferos terrestres contuvieron cantidades perceptiblemente ms altas de epfitas dispersadas por aves. Mientras que no hubo una diferencia significativa en las epfitas dispersadas por el viento entre las dos especies. Este resultado inesperado puede ser debido a las diferentes arquitecturas entre las dos especies estudiadas.
Text in English.
Monteverde Biological Station (Costa Rica)
Estacin Biolgica de Monteverde (Costa Rica)
Tropical Ecology Fall 2004
Ecologa Tropical Otoo 2004
t Monteverde Institute : Tropical Ecology