USF Libraries

Family composition of vascular epiphytes varies by directional quadrant

MISSING IMAGE

Material Information

Title:
Family composition of vascular epiphytes varies by directional quadrant
Translated Title:
La composición de la familia de epifitas vasculares varia según el cuadrante direccional ( )
Physical Description:
Book
Language:
English
Creator:
Bums, David N

Subjects

Subjects / Keywords:
Epiphytes   ( lcsh )
Trade winds   ( lcsh )
Costa Rica--Puntarenas--Monteverde Zone--Monteverde   ( lcsh )
Epífitas
Vientos alisios
Costa Rica--Puntarenas--Zona de Monteverde--Monteverde
Tropical Ecology Fall 2003
Ecología Tropical Otoño 2003
Genre:
Reports   ( lcsh )
Reports

Notes

Abstract:
Vascular epiphytes are an extremely diverse and prevalent plant-form of neotropical cloud forests, and are strongly affected by abiotic factors including light and moisture. The goal of this study was to determine whether the family composition and diversity of vascular epiphytes living on pasture tree trunks differed by quadrant (northeast, northwest, southeast, and southwest). It was hypothesized that the northeast quadrant would exhibit the greatest diversity due to the mist-laden trade winds blowing from that direction. Twenty trees in a pasture surrounded by lower montane wet forest in Monteverde, Puntarenas, Costa Rica were divided into four quadrants and sampled for vascular epiphytes (N = 597 plants), which were tallied by family. A Chi-squared analysis revealed that there was a nonrandom frequency of families across quadrants (x2 = 29.445, df = 12, P = 0.0034). The family diversity of the northeast (H’ = 2.067) was significantly higher than the diversity of the southwest (H’ = 1.817; t = 2.497, df = 263.04). Additionally, there were more bromeliads in the northeast than expected (54 observed, 34.7 expected), and less than expected in the southeast (15 observed, 22.1 expected) and southwest (7 observed, 18.4 expected). Also, there were more Pteridophytes in the southwest than expected (45 observed, 36.8 expected) and less than expected in the northeast (54 observed, 69.5 expected). These differences may be due to a combination of abiotic and biotic factors related to moisture acquisition by the epiphytes.
Abstract:
Las epífitas vasculares son una forma vegetal extremadamente diversa y frecuente en los bosques nubosos neotropicales, y se ven fuertemente afectados por factores abióticos incluyendo luz y humedad. El objetivo de este estudio fue determinar si la composición de la familia y la diversidad de las epífitas vasculares que viven en los troncos de los árboles en los potreros diferenciados por cuadrante (noreste, noroeste, sureste, y sudoeste). Se hizo una hipótesis de que el cuadrante noreste mostraría la mayor diversidad debido a los vientos alisios cargados de humedad soplando de esa dirección. Veinte árboles en un potrero rodeado de bosque húmedo montano bajo en Monteverde, Puntarenas, Costa Rica fueron divididos en cuatro cuadrantes y muestreados para la existencia de epífitas vasculares (N = 597 plantas), que fueron identificadas por familia. Un análisis Chi-cuadrado reveló que había una frecuencia no aleatoria de familias a través de los cuadrantes (x2 = 29.445, df = 12, P = 0.0034). La diversidad de familias del noreste (H’= 2.067) era perceptiblemente más alta que la diversidad del sudoeste (H’ = 1,817; t = 2,497, df=263.04). Además, había más bromelias en el noreste que lo esperado (54 observados, 34,7 esperados), y menos que lo esperado en el sureste (15 observados, 22,1 esperados) y en el sudoeste (7 observados, 18,4 esperados). También, había más pteridófitos en el sudoeste que lo esperado (45 observados, 36,8 esperados) y menos que lo esperado en el noreste (54 observados, 69,5 esperados). Estas diferencias pueden ser debido a una combinación de factores abióticos y bióticos relacionados con la absorción de humedad por las epífitas.
Language:
Text in English.
General Note:
Digitized by MVI

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - M39-00445
usfldc handle - m39.445
System ID:
SFS0036390:00001


This item is only available as the following downloads:


Full Text
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 M39-00445
040
FHM
0 041
eng
049
FHmm
1 100
Bums, David N.
242
La composicin de la familia de epifitas vasculares varia segn el cuadrante direccional
245
Family composition of vascular epiphytes varies by directional quadrant
260
g 2003-12
500
Digitized by MVI
3 520
Vascular epiphytes are an extremely diverse and prevalent plant-form of neotropical cloud forests, and are strongly
affected by abiotic factors including light and moisture. The goal of this study was to determine whether the family
composition and diversity of vascular epiphytes living on pasture tree trunks differed by quadrant (northeast,
northwest, southeast, and southwest). It was hypothesized that the northeast quadrant would exhibit the greatest
diversity due to the mist-laden trade winds blowing from that direction. Twenty trees in a pasture surrounded by lower
montane wet forest in Monteverde, Puntarenas, Costa Rica were divided into four quadrants and sampled for vascular
epiphytes (N = 597 plants), which were tallied by family. A Chi-squared analysis revealed that there was a
nonrandom frequency of families across quadrants (x2 = 29.445, df = 12, P = 0.0034). The family diversity of the
northeast (H = 2.067) was significantly higher than the diversity of the southwest (H = 1.817; t = 2.497, df =
263.04). Additionally, there were more bromeliads in the northeast than expected (54 observed, 34.7 expected), and
less than expected in the southeast (15 observed, 22.1 expected) and southwest (7 observed, 18.4 expected). Also,
there were more Pteridophytes in the southwest than expected (45 observed, 36.8 expected) and less than expected in
the northeast (54 observed, 69.5 expected). These differences may be due to a combination of abiotic and biotic
factors related to moisture acquisition by the epiphytes.
Las epfitas vasculares son una forma vegetal extremadamente diversa y frecuente en los bosques nubosos neotropicales, y se ven fuertemente afectados por factores abiticos incluyendo luz y humedad. El objetivo de este estudio fue determinar si la composicin de la familia y la diversidad de las epfitas vasculares que viven en los troncos de los rboles en los potreros diferenciados por cuadrante (noreste, noroeste, sureste, y sudoeste). Se hizo una hiptesis de que el cuadrante noreste mostrara la mayor diversidad debido a los vientos alisios cargados de humedad soplando de esa direccin. Veinte rboles en un potrero rodeado de bosque hmedo montano bajo en Monteverde, Puntarenas, Costa Rica fueron divididos en cuatro cuadrantes y muestreados para la existencia de epfitas vasculares (N = 597 plantas), que fueron identificadas por familia. Un anlisis Chi-cuadrado revel que haba una frecuencia no aleatoria de familias a travs de los cuadrantes (x2 = 29.445, df = 12, P = 0.0034). La diversidad de familias del noreste (H= 2.067) era perceptiblemente ms alta que la diversidad del sudoeste (H = 1,817; t = 2,497, df=263.04). Adems, haba ms bromelias en el noreste que lo esperado (54 observados, 34,7 esperados), y menos que lo esperado en el sureste (15 observados, 22,1 esperados) y en el sudoeste (7 observados, 18,4 esperados). Tambin, haba ms pteridfitos en el sudoeste que lo esperado (45 observados, 36,8 esperados) y menos que lo esperado en el noreste (54 observados, 69,5 esperados). Estas diferencias pueden ser debido a una combinacin de factores abiticos y biticos relacionados con la absorcin de humedad por las epfitas.
546
Text in English.
650
Epiphytes
Trade winds
Costa Rica--Puntarenas--Monteverde Zone--Monteverde
4
Epfitas
Vientos alisios
Costa Rica--Puntarenas--Zona de Monteverde--Monteverde
653
Tropical Ecology Fall 2003
Ecologa Tropical Otoo 2003
655
Reports
720
MVI
773
t Monteverde Institute : Tropical Ecology
856
u http://digital.lib.usf.edu/?m39.445



PAGE 1

Family composition of vascular epiphytes varies by directional quadrant David N. Bums Department of Environmental Studies, Bates College, Lewiston, ME 04240 ABSTRACT Vascular epiphytes are an extremely diverse and prevalent plant form of neotropical cloud forests, and are strongly affected by abiotic factors including light and moisture. The goal of this study was to determine whether the family composition and diversity of vascular epiphytes living on pasture tree trunks differed by quadrant (northe ast, northwest, southeast, and southwest). It was hypothesized that the northeast quadrant would exhibit the greatest diversity due to the mist laden trade winds blowing from that direction. Twenty trees in a pasture surrounded by lower montane wet forest in Monteverde, Puntarenas, Costa Rica were divided into four quadrants and sampled for vascular epiphytes (N = 597 plants), which were tallied by family. A Chi squared analysis revealed that there was a nonrandom frequency of families across quadrants (x 2 = 29.445, df = 12, P = 0.0034). The family diversity of the 263.04). Additionally, there were more bromeliads in the northeast than expected (5 4 observed, 34.7 expected), and less than expected in the southeast (15 observed, 22.1 expected) and southwest (7 observed, 18.4 expected). Also, there were more Pteridophytes in the southwest than expected (45 observed, 36.8 expected) and less than expected in the northeast (54 observed, 69.5 expected). These differences may be due to a combination of abiotic and biotic factors related to moisture acquisition by the epiphytes. RESUMEN Las ep fitas vasculares son una forma vegetal extremadamente diver sa y frecuente en los bosques nubosos neotropicales, y son afectad as fuertemente por factores abi ticos incluyendo luz y humedad. La meta de este estudio fue determinar si la composici n de las familias y la diversidad de las e p fitas vasculares que vivan en troncos de rboles en potreros diferenciaron por cuadrante (noreste, noroeste, sureste, y sudoeste). La hip tesis fue q ue el cuadrante noreste exhibir a la mayor diversidad debido a los vientos alisios cargados de humedad en esa direccin. Veinte rboles en un potrero pasto rodeado de bosque hmedo montano bajo en Monteverde, Puntarenas, Costa Rica fueron dividido en cuatro cuadrantes y muestreados para la existencia de epfitas vasculares (N = 597 plantas), que fueron identificadas por familia. Un anlisis Chi cuadrado revel que haba una frecuencia no al azar de familias a travs de los cuadrantes (x 2 = 29.445, df = 12, P = 0.0034). La diversi = 2.067) era perceptiblemente m s alta que la = 1,817; t = 2,497, df=263.04). Adems haba m s bromelias en el noreste que lo esperado (54 observados, 34,7 esperados), y menos que lo esperado en el sureste (15 observados, 22,1 esperados) y en el sudoeste (7 observados, 18,4 esperados). Tambin haba ms pterid fitos en el sudoeste que lo esperado (45 observados, 36,8 esperados) y menos que lo esperado en el noreste (54 observados, 69,5 esperados). Estas diferencias pueden ser debido a una combinacin de factores abi tico s y bi ticos relacionados con l a absorcin de humedad por las ep fitas.

PAGE 2

INTRODUCTION An important feature of neotropical forests is the abundance and diversity of vascular epiphytes. Worldwide, there are over 23, 000 species of vascular epiphytes in 876 genera and 84 families, representing some ten percent of all vascular plants (Lesica & Antibus 1990; Zimmerman & Olmsted 1 992). Epiphytes account for up to 35% of vascular flora in some neotropical wet forests, and in some montane rain forests, their biomass can equal half of all tree leaf biomass (Ingram & Nadkarni 1993). In the neotropical lower montane wet forest of Montev erde, Costa Rica, epiphytes are the most species rich growth form, with 878 identified species (Haber 2000). There has been increasing attention given to canopy communities, especially in cloud forests. They are important subsystems in ecosystem level int eractions between terrestrial and atmospheric processes because they can retain atmospheric nutrients and pollutants (Ingram & Nadkarni 1993). Additionally, epiphytes are useful climatic indicators and may aid ecological assessment of forests (Hietz Seifer t et al. 1995). One highly studied phenomenon is that epiphytic growth is strongly influenced by abiotic factors, including substrate texture and size, light, nutrients taken from the atmosphere, and moisture. These resources are limiting and epiphytes ma y compete for them (Hopcus 1999; Haber 2000). Studies have shown that epiphyte growth in wet forests is influenced most strongly by light and moisture availability (Veblen 1996). Hietz Seifert et al. (1995) found that trees with greater exposure to cool an d humid winds from the north had more epiphytes than trees receiving hot and dry winds from the south. While ter Steege and Cornelissen (1989) have proposed that tree bases and humid understory areas are typically poorer in epiphyte diversity than the can opy, Veblen (1996) argues that the canopy conditions of abundant wind, light, and moisture conditions are experienced by trees in the hillside pasture behind the Estacin Biolgica Monteverde in Costa Rica. Two of the above conditions, wind and moisture, h ave been studied in this pasture, and it is known that mist laden trade winds blow southwest ( Meisner Bagdahn 2003). As more neotropical forests are converted to cropland or pastures, it is becoming clearer that a better understanding of these altered eco systems is needed. Often, a few isolated trees are left following deforestation and these may play an important role in maintaining biodiversity. Hietz Seifert et al. (1995) found that the numbers of epiphytic species per tree in pastures were within the r ange of those found in studied forests. Thus, they may serve as nuclei for reestablishment of forest species (Hietz Seifert et al. 1995). The goal of this experiment was to determine whether the family composition and diversity of vascular epiphytes livin g on pasture tree trunks differed by quadrant (northeast, northwest, southeast, and southwest). It was hypothesized that the northeast facing quadrant would exhibit the greatest diversity of epiphytic families, as the trade winds from the northeast bring t he most mist to that quadrant during the dry season (Meisner Bagdahn 2003, Ingram & Nadkarni 1993). This would provide the most favorable growing conditions for

PAGE 3

vascular epiphytes. Conversely, the southwest side receives the least mist in the dry season, a nd will support the least diversity of vascular epiphytes. This study, relating vascular epiphytic family diversity to tree quadrant based on fine scale mist frequencies is a novel concept, and thus far, has received little attention from the tropical scie ntific community. METHODS Study Site The trees sampled were located in a hillside pasture of approximately 2.5 ha, behind the Estacin Biolgica Monteverde, Puntarenas, Costa Rica. This site lies at approximately 1570m in elevation, and is surrounded by Lo wer Montane Wet Forest on all sides except the southwest (Levine 1999). Total annual precipitation is about 2 2.5m of rain and an additional 0.5 2m is contributed by mist (Ingram & Nadkarni 1993). The majority of this mist is delivered between November and May by trade winds from the northeast (Ingram & Nadkarni 1993). The majority of this mist is delivered between November and May by trade winds from the northeast (Ingram & Nadkarni 1993, Meisner Bagdahn 2003). Tree Selection Eighteen of the 20 host trees used in this study were those previously sampled in Meisner abundance for the trees has been recorded (Meisner Bagdahn 2003). These trees were o riginally chosen, regardless of identity, on the basis that they contained over ten pleurothalid individuals and were at least 30 cm in dbh (Meisner Bagdahn 2003). The two newly sampled trees were chosen on the bases of having over 30 vascular epiphytes an d a dbh of at least 30cm. Tree Preparation & Experimental Design Each of the 20 trees was divided into four quadrants, representing the northeast, northwest, southeast, and southwest facing sides of those trees. These quadrants were created by hanging string from nails placed on the north, east, south, and west sides of the tree. Many of the trees were on a slope, and it was necessary to mark the lower limit of the quadrants by placing a string around the circumference of the bole, level with the ground on the north side. The height of the quadrants was two meters above the lower limit on the north side. The area of each quadrant was measured by multiplying the height (two meters) by the average width of that quadrant. Branches, deep clefs, and the area between trunks whose bole split below the two meter line were not included in the quadrants. In each quadrant, all occurrences of vascular plants were tallied by family, excluding Orchidaceae. Plants with runners were only counted once per quadrant, howe ver, if they crossed over into another quadrant, their occurrence was also tallied there. Plants rooted

PAGE 4

outside of the quadrant, but who were in contact with the tree and shaded the quadrant were also tallied. It was indicated that these plants were from o utside of the quadrant however. Analysis Differences in community composition and diversity were analyzed for each quadrant, across all 20 trees, using a contingency table analysis. Only those families with greater than 20 individuals across all trees we re included. Although Acanthaceae had more than 20 individuals across all trees, it was not included because it is not typically an epiphytic family (Gentry 1993), and those individuals found were non liana herbs growing low on the tree trunks. Additionall y, a second contingency table analysis was created, omitting those plants rooted outside of the quadrants. A post hoc test was also run for each. Diversity parameters, including number of individuals (N), species richness (S), Margalef indices (S marg ), Sh annon tests. Family similarities across the quadrants were also compared using a Sorenson quantitative measure of similarity (C N ). Finally, a one way ANOVA was run to determine whether the quadrants differed in mean area. RESULTS There was a significant difference found in family composition between quadrants by the Chi squared analysis (X 2 = 29.445, df = 12, P = 0.0034). Additionally, the analysis revealed that a higher number of bromeliads were present in the northeast than expected, and a lower number than expected was found in the southeast and southwest (Table 1, Table 2). There were also higher num bers of Pteridophytes found in the southwest than expected, and less than expected found in the northeast (table 1, table 2). The exclusion of those plants rooted outside of the quadrant boundaries did not change this trend (X 2 = 28.550, df = 12, P = 0.004 6 (Appendix 1, Appendix 2). The observed frequencies for all families are seen in Table 3. Individuals rooted outside of the quadrant boundaries have been removed from the family frequencies shown in Table 4. Diversity parameters for each quadrant across all trees are shown in Table 5. Number of individuals and family richness were greatest in the northeast quadrant (N = 208, S = 17) and lowest in the southeast (N = 102, S = 10). Additionally, the Margalef index for the northeast quadrant (S marg = 2.998) w as higher than that of the southwest (S marg = 1.946). The = 263.04), as seen in Table 6. No other Shannon Weiner indices were significantly different. Finally, the northeast and southwest have the least similar family compositions across quadrants (C N = 0.619), while the two southerly quadrants have the most si milar family

PAGE 5

compositions (C N = 0.873) (Table 7). It should also be noted that there were no significant differences in quadrant area across the trees sampled. DISCUSSION These results indicate that there was a significant difference in the community com position and diversity of vascular epiphytes on the different directional faces of tree trunks in a lower montane wet pasture, supporting the hypothesis. More specifically, the greatest differences were between the northeast and southwest facing quadrants, with the northeast having the Vascular epiphytes are most likely to establish on the northeast side of tree trunks, as trade winds bring the majorit y of mist during the otherwise potentially desiccating dry season (Ingram & Nadkarni; 1993, Meisner Bagdahn 2003), and moisture is a limiting factor of epiphyte growth (Veblen 1996). These results are consistent with the Hietz Seifert et al. study (1995), which found that trees with a higher exposure to cool and humid winds from the north had more epiphytes than trees receiving hot and dry winds from the south. Similarly, Levine (1999) suggests that orchids establishing on a northeastern facing site may hav e advantage over the potentially more water stressed epiphytes pointing other compass directions. One conclusion d rawn from this is that differences in community composition and diversity of vascular epiphytes among the intercardinal points are the result of abiotic factors. As seen in Table 1, there was a higher number of bromeliads present in the northeast than expected, and a lower number than expected in the southeast and southwest. Interestingly though, there were a higher numbers of pteridophytes fou nd in the southwest than expected, and less than expected found in the northeast (Table 1). The pattern of bromeliad growth can be explained using the previous argument of the importance of abiotic factors. The higher amount of mist reaching the north side of the tree facilitates germination and growth. Additionally, the higher abundance of bromeliads on the north facing side leads to a higher seed rain there, and more of these seeds germinate than in other quadrants due to the favorable conditions. The pat tern of pteridophytic growth requires an alternate explanation however. One possible explanation may be avoidance of competition and horizontal niche partitioning. Resources such as light and moisture are limiting resources for vascular epiphytes, leading to high competition (Veblen 1996; Hopcus 1999; Haber 2000). This competition may lead to niche partitioning, in which plants have evolved to occupy different microhabitats in order to utilize the available resources and thereby minimize competition (Begon et al. 1990). Most epiphytic plants, including atmospheric bromeliads, can only absorb water in liquid form. Some tropical epiphytic fern species however, are poikilohydric, and can take water from saturated air. These plants can quickly absorb water over the entire surface of their leaves, but can lose it rapidly as well. In a drier atmosphere, the leaves curl

PAGE 6

up, and expand when moist conditions return (Richards 1996). As it may be possible for some pteridophytes to live in more xeric conditions than oth er epiphytes, those who do so will avoid competition for other abiotic factors like space and light. As greater amounts of tropical land are being converted into pastures, research needs to be directed toward understanding and accurately describing how ep iphytes partition resources among themselves. Epiphytes form important centers of biodiversity in pastures and can act as useful climatic indicators, as they retain atmospheric nutrients and pollutants. In order to utilize them as climatic indicators and s ites of forest reestablishment in conservation efforts though, a better understanding of how they are affected by and respond to both abiotic and biotic factors is needed. ACKNOWLEDGEMENTS the previous frustrations. It could not have been done without her plant identifications. A big thanks also to Alan Masters for running such an amazing program. C heers to Andrew Rodstrom for critiquing my paper, spending the time to teach me to climb, and for caring so much about the whereabouts of my ladder. Matt and Carmen, your statistical help was a big gica Monteverde, for allowing me to the nicest person in the world, Catherine Ross for reviewing my paper, Chelsea Beebe for letting me use her map when mine blew away, and Alyson for moving like a Jellyfish. A special final thanks to the Solis family, who graciously allowed me to live with them and gave me the amount of support I would expect from my own family. __________________________________________________________________________ LITERATURE CITED Begon, M., J.L. Harper, and C.R. Townsend. 1990. Ecology: Individuals, Populations and Communities Blackwell Scientific Publications, Boston. Catling, P.M. and L.P. Lefkovitch. 1989. Association of vascular epiphytes in a Guatemalan cloud forest. Biotropica 21: 35 40. Gentry, A.H. 1993. A Field Guide to the Families and Genera of Woody Plants of Northwest South America (Colombia, Ecuador, Peru) The University o f Chicago Press, Chicago; pg. 205 212.

PAGE 7

Monteverde: ecology and conservation of a tropical cloud forest Oxford University Press, New York; pg. 55 Hietz Seifert, U., P. Hietz, and S. Guevara. 1996. Epiphyte vegetation and diversity on remnant trees after forest clearance in southern Veracruz, Mexico. Biological Conservation 75: 103 111. Oncidium bract eatum and Oncidium bryolophotum Ingram, S.W. and N.M. Nadkarni. 1993. Composition and distribution of epiphytic organic matter in a neotropical cloud forest, Costa Rica. Biotropica 25: 370 383. Lesica, P. an d P.K. Antibus. 1990. The occurrence of mycorrhizae in vascular epiphytes of two Costa Rican rain forests. Biotropica 22: 250 258. Levine, M.T. 1999. Niche partitioning of congeners in two orchid genera, Scaphyglottis and Epidendrum CIEE Spring Tropical Ecology and Conservation. Meisner Bagdahn, Hanna. 2003. Climate change and microclimate variation: Effects on fitness and abundance of cloud forest orchids. American Colleges of the Midwest Spring Program. Richards, P.W. 1996. The Tropical Rain Forest Cambridge University Press, Cambridge; pg. 142 143. Ter Steege, H. and J.H.C Cornelissen. 1989. Distribution and ecology of vascular epiphytes in lowland rain forest of Guyana. Biotropica 21: 331 339. Veblen, K. 1996. Comparison of epiphytes on stumps i n a pasture and in the forest. CIEE Summer Tropical Ecology and Conservation. Zimmerman, J.K. and I.C. Olmsted. 1992. Host tree utilization by vascular epiphytes in a seasonally inundated forest (Tintal) in Mexico. Biotropica 24: 402 407.

PAGE 8

Table 1 The number of observed and expected vascular epiphytes by family, on 20 trees, across the four directional quadrants. N= 504 vascular epiphytes. Family NE NW SE SW Observed Expected Observed Expected Observed Expected Observed Expected Araceae 10 11.13 6 8.91 10 7.07 7 5.89 Bromeliaceae 54 34.74 27 27.79 15 22.07 7 18.39 Ericaceae 21 20.91 15 16.73 13 13.29 13 11.07 Piperaceae 31 33.73 27 26.98 24 21.42 18 17.86 Pteridophyta 54 69.48 61 55.59 46 44.14 45 36.79 Table 2. Post Hoc test results of contingency table analysis. N = 504 vascular epiphytes. Family NE NW SE SW Araceae 0.431 1.178 1.285 0.521 Bromeliaceae 4.50 0.198 1.904 3.286 Ericaceae 0.025 0.529 0.094 0.683 Piperaceae 0.645 0.004 0.700 0.042 Pteridophyta 2.968 1.105 0.410 1.943

PAGE 9

Table 3 The abundance of vascular epiphytes and shading vascular plants by family, across the four directional quadrants on 20 trees. Italicized values were those used in the contingency table analysis. N = 597 vascular plants. Family NE NW SE SW Totals Araceae 10 6 10 7 33 Bromeliaceae 54 27 15 7 103 Ericaceae 21 15 13 13 62 Piperaceae 31 27 24 18 100 Pteridophyta 54 61 46 45 206 Gesneriaceae 8 4 1 3 16 Clusiaceae 4 1 0 0 5 Asteraceae 1 1 1 0 3 Mimosaceae 1 0 0 0 1 Acanthaceae 10 6 4 2 22 Moraceae 1 0 0 0 1 Rubiaceae 3 2 1 1 7 Poaceae 3 5 2 0 10 Rosaceae 0 1 1 1 3 Fagaceae 0 0 1 0 1 Smilacaceae 0 0 5 5 10 Solanaceae 0 0 1 0 1 Passifloraceae 2 0 1 0 3 Araliaceae 0 1 0 0 1 Lauraceae 2 2 0 0 4 Melastomataceae 2 0 1 0 3 Anacardiaceae 1 0 0 0 1 Cunoniaceae 0 1 0 0 1 Total 208 160 127 102 597(504)

PAGE 10

Table 4 The abundance of vascular epiphytes, omitting individuals rooted outside of quadrant boundaries, across the four directional quadrants on 20 trees. Italicized values were those used in the contingency table analysis. S = species richness. N = 578 vascular plants. Family NE NW SE SW Totals Araceae 10 6 9 7 32 Bromeliaceae 54 27 15 7 103 Ericaceae 21 15 12 13 61 Piperaceae 31 27 23 18 99 Pteridophyta 54 61 46 45 206 Gesneriaceae 8 4 1 3 16 Clusiaceae 4 1 0 0 5 Asteraceae 1 1 1 0 3 Mimosaceae 0 0 0 0 0 Acanthaceae 10 6 3 2 21 Moraceae 1 0 0 0 1 Rubiaceae 2 2 0 0 4 Poaceae 3 5 0 0 8 Rosaceae 0 1 1 1 3 Fagaceae 0 0 0 0 0 Smilacaceae 0 0 5 5 10 Solanaceae 0 0 1 0 1 Passifloraceae 2 0 1 0 3 Araliaceae 0 1 0 0 1 Lauraceae 0 0 0 0 0 Melastomataceae 1 0 0 0 1 Anacardiaceae 0 0 0 0 0 Cunoniaceae 0 0 0 0 0 Total 202 157 118 101 578(501) S 14 13 12 9 18 Table 5. Diversity parameters for each quadrant, with all individuals present. N = 597 vascular plants. Parameter NE NW SE SW N 208 160 127 102 S 17 15 16 10 Smarg 3 .00 2.76 3.1 0 1.95 Shannon Weiner 2.07 1.91 1.98 1.82 Evenness 0.73 0.7 0 0.71 0.79

PAGE 11

Table 6 Results of t test for differences in diversity of Shannon the northeast, 0.0068 for the northwest, 0.0083 for the southeast, and 0.0053 for the southwest. N= 597 vascular plants. Quadrants Compared t test Value DF Significant NE: SW 2.50 263.04 Yes NW: SW 0.80 259.39 No SE: SW 1.35 226.08 No Table 7 : Values for Sorenson quantitative measure of similarity (CN). Higher values indicate that those quadrants are more similar in family composition. N = 597 vascular plants Quadrants Compared C N NE: NW 0.804 NE: SE 710 NE: SW 0.619 NW: SE 0.794 NW: SW 0.733 SE: SW 0.873

PAGE 12

Appendix 1 The number of observed and expected vascular epiphytes by family, omitting individuals rooted outside of quadrant boundaries, across the four directional quadrants on 20 trees. N = 501 vascular epiphytes. Family NE NW SE SW Observe d Expecte d Observe d Expecte d Observe d Expecte d Observe d Expecte d Araceae 10 10.86 6 8.69 9 6.71 7 5.75 Bromeliaceae 54 34.95 27 27.96 15 21.59 7 18.5 Ericaceae 21 20.7 15 16.56 13 12.78 12 10.96 Piperaceae 31 33.59 27 26.87 23 20.75 18 17.78 Pteridophyta 54 69.9 61 55.92 46 43.17 45 37.01 Appendix 2 Post Hoc test results of contingency table analysis omitting individuals rooted outside of quadrant boundaries. N = 501 vascular epiphytes. Family NE NW SE SW Araceae 0.331 1.104 1.03 0.596 Bromeliaceae 4.448 0.239 1.789 3.313 Ericaceae 0.087 0.479 0.263 0.727 Piperaceae 0.614 0.032 0.621 0.063 Pteridophyta 3.049 1.037 0.631 1.891