USF Libraries

Food resource density and territorial behavior of Monteverde Cloud Forest hummingbirds

MISSING IMAGE

Material Information

Title:
Food resource density and territorial behavior of Monteverde Cloud Forest hummingbirds
Translated Title:
Densidad de recursos en alimentos y el comportamiento territorial de los colibríes del Bosque Nuboso de Monteverde ( )
Physical Description:
Book
Language:
English
Creator:
Sawtelle, Sarah
Publication Date:

Subjects

Subjects / Keywords:
Hummingbirds--Food   ( lcsh )
Birds--Behavior   ( lcsh )
Aggressive behavior in animals   ( lcsh )
Monteverde Biological Station (Costa Rica)   ( lcsh )
Colibríes--Alimento
Aves--Comportamiento
Comportamiento agresivo en los animales
Estación Biológica de Monteverde (Costa Rica)
Tropical Ecology Spring 2011
Ecología Tropical Primavera 2011
Genre:
Reports   ( lcsh )
Reports

Notes

Abstract:
The determinants of territorial behavior patterns in aggressive organisms are not well understood. Hummingbirds provide an ideal model for studying territoriality as they exhibit obvious territorial behaviors that are likely to change in response to food resource manipulations. I studied trends in territorial behaviors of hummingbirds in response to varied food density. Observations were made on three different arrangements of hummingbird feeders: small: with twelve feeders in a 1m squared square, medium: with twelve feeders in a 25m squared square, and large: with six feeders in a 50m squared, 5x10m rectangle. There was no significant difference in the number of territorial interactions per unit time between the different feeder arrangements. However, the composition of species and sex of visitors to the feeders did differ, with a lesser proportion of males and a greater proportion of females tending to visit the more distant feeders in most species. Territorial interactions significantly changed between feeder arrangements; the percentage of total aggressive interactions that were intraspecific rose from 12.9% in the small arrangement to 33.3% in the medium and 69.2% in the large, with the rate of interspecific interactions inversely diminishing. The similar rate of territorial interactions between density conditions indicates that territoriality is determined by more than simple energy economics, which would make defending the densest resources most beneficial. Instead, it appears that differences between species and sexes' feeding strategies predict territorial patterns.
Abstract:
Los determinantes de los patrones de comportamiento territorial en los organismos agresivos no son bien entendidos. Los colibríes son un modelo ideal para el estudio de territorialidad al exhibir comportamientos obvios de territorialidad que tienden a cambiar en respuesta a la manipulación de los recursos. Estudie las tendencias en los comportamientos de territorialidad de los colibríes en respuesta a la variación en la densidad de comida. Se hicieron observaciones en tres disposiciones diferentes de comederos: pequeña, con 12 comederos en un área de 1m², media con 12 comederos en un área de 25m², y larga con seis comederos en un rectángulo de 5x10m. No existe diferencia en el número de interacciones territoriales por unidad de tiempo entre las diferentes disposiciones de los comederos. Sin embargo, la composición de las especies y el sexo de los visitantes si difiere, con una menor proporción de machos y una mayor proporción de hembras visitando los comederos a mayor distancia en la mayoría de las especies. Las interacciones territoriales cambian significativamente entre la disposición de los comederos; el porcentaje total de interacciones agresivas intraespecíficas alcanza el 12.9% en la disposición pequeña, 33.3% en la mediana y 69.2% en la larga, con una disminución en la tasa de interacciones interespecíficas. Una tasa similar de interacciones territoriales entre las condiciones de densidad indica que la territorialidad es determinada por más que simple economía de la energía, lo que hace la defensa del recurso denso más beneficioso. En cambio, parece ser que las diferencias entre especies y sexos predice patrones territoriales.
Language:
Text in English.
General Note:
Born Digital

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 - M39-00239
usfldc handle - m39.239
System ID:
SFS0001408:00001


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 M39-00239
040
FHM
0 041
eng
049
FHmm
1 100
Sawtelle, Sarah
242
Densidad de recursos en alimentos y el comportamiento territorial de los colibres del Bosque Nuboso de Monteverde
245
Food resource density and territorial behavior of Monteverde Cloud Forest hummingbirds
260
c 2011-05
500
Born Digital
3 520
The determinants of territorial behavior patterns in aggressive organisms are not well understood. Hummingbirds provide an ideal model for studying territoriality as they exhibit obvious territorial behaviors that are likely to change in response to food resource manipulations. I studied trends in territorial behaviors of hummingbirds in
response to varied food density. Observations were made on three different arrangements of hummingbird feeders: small: with twelve feeders in a 1m squared square, medium: with twelve feeders in a 25m squared square, and large: with six feeders in a 50m squared, 5x10m rectangle. There was no significant difference in the number of territorial interactions per unit time between the different feeder arrangements. However, the composition of species and sex of visitors to the feeders did differ, with a lesser proportion of males and a greater proportion of females tending to visit the more distant feeders in most species. Territorial interactions significantly changed between feeder arrangements; the percentage of total aggressive interactions that were intraspecific rose from 12.9% in the small arrangement to 33.3% in the medium and 69.2% in the large, with the rate of interspecific interactions inversely diminishing. The
similar rate of territorial interactions between density conditions indicates that territoriality is determined by more than simple energy economics, which would make defending the densest resources most beneficial. Instead, it appears that differences between species and sexes' feeding strategies predict territorial patterns.
Los determinantes de los patrones de comportamiento territorial en los organismos agresivos no son bien entendidos. Los colibres son un modelo ideal para el estudio de territorialidad al exhibir comportamientos obvios de territorialidad que tienden a cambiar en respuesta a la manipulacin de los recursos. Estudie las tendencias en los comportamientos de territorialidad de los colibres en respuesta a la variacin en la densidad de comida. Se hicieron observaciones en tres disposiciones diferentes de comederos: pequea, con 12 comederos en un rea de 1m, media con 12 comederos en un rea de 25m, y larga con seis comederos en un rectngulo de 5x10m. No existe diferencia en el nmero de interacciones territoriales por unidad de tiempo entre las diferentes disposiciones de los comederos. Sin embargo, la composicin de las especies y el sexo de los visitantes si difiere, con una menor proporcin de machos y una mayor proporcin de hembras visitando los comederos a mayor distancia en la mayora de las especies. Las interacciones territoriales cambian significativamente entre la disposicin de los comederos; el porcentaje total de interacciones agresivas intraespecficas alcanza el 12.9% en la disposicin pequea, 33.3% en la mediana y 69.2% en la larga, con una disminucin en la tasa de interacciones interespecficas. Una tasa similar de interacciones territoriales entre las condiciones de densidad indica que la territorialidad es determinada por ms que simple economa de la energa, lo que hace la defensa del recurso denso ms beneficioso. En cambio, parece ser que las diferencias entre especies y sexos predice patrones territoriales.
546
Text in English.
650
Hummingbirds--Food
Birds--Behavior
Aggressive behavior in animals
Monteverde Biological Station (Costa Rica)
4
Colibres--Alimento
Aves--Comportamiento
Comportamiento agresivo en los animales
Estacin Biolgica de Monteverde (Costa Rica)
653
Tropical Ecology Spring 2011
Ecologa Tropical Primavera 2011
655
Reports
720
CIEE
773
t Monteverde Institute : Tropical Ecology
856
u http://digital.lib.usf.edu/?m39.239



PAGE 1

Food Resource Density and Territorial Behavior of Monteverde Cloud Forest Hummingbirds Sarah Sawtelle Department of Biology and Environmental Studies, Oberlin College ABSTRACT The determinants of territorial behavior patterns in aggressive organisms ar e not well understood. Hummingbirds provide an ideal model for studying territoriality as they exhibit obvious territorial behaviors that are likely to change in response to food resource manipulations. I studied trends in territorial behaviors of hummingb irds in response to varied food density. Observations were made on three different arrangements of hummingbird feeders: small: with twelve feeders in a 1m 2 square, medium: with twelve feeders in a 25m 2 square and large: with six feeders in a 50m 2 5x10m r ectangle There was no significant difference in the number of territorial interactions per unit time between the different feeder arrangements. However, the composition of species and sex of visitors to the feeders did differ, with a lesser proportion of males and a greater proportion of females tending to visit the more distant feeders in most species. Territorial interactions significantly changed between feeder arrangements; the percentage of total aggressive interactions that were intraspecific rose fr om 12.9% in the small arrangement to 33.3% in the medium and 69.2% in the large, with the rate of interspecific interactions inversely diminishing. The similar rate of territorial interactions between density conditions indicates that territoriality is det ermined by more than simple energy economics, which would make defending the densest resources most beneficial. Instead, it appears that differences between species and sexes' feeding strategies predict territorial patterns. RESUMEN Los determinantes de los patrones de comportamiento territorial en organismos agresivos no son bien entendidos. Los colibres son un modelo ideal para el estudio de territorialidad al exhibircomportamientos obvios de territorialidad que tienden a cambiar en respuesta a la man ipulacin de recursos. Estudie las tendencias en los comportamientos de territorialidad de colibres en respuesta a la variacin en la densidad de comida. Se hicieron observaciones en tres disposiciones diferentes de comederos: pequea, con 12 comederos en un rea de 1m2, media con 12 comederos en un rea de 25m2, y larga con seis comederos en un rectngulo de 5x10m. No existe diferencia en el nmero de interacciones territoriales por unidad de tiempo entre las diferentes disposiciones de los comederos. Sin embargo, la composicin de especies y sexo de los visitantes si difiere, con una menor proporcin de machos y una mayor proporcin de hembras visitando los comederos a mayor distancia en la mayora de las especies. Las interacciones territoriales cambian significativamente entre la disposicin de los comederos; el porcentaje total de interacciones agresivas intraespecficas alcanza el 12.9% en la disposicin pequea, 33.3% en la mediana y 69.2% en la larga, con una disminucin en la tasa de interacciones i nterespecficas. Una tasa similar de interacciones territoriales entre las condiciones de densidad indica que la territorialidad es determinada por ms que simple economa de la energa, lo que hace la defensa del recurso denso ms beneficioso. En cambio, parace ser que las diferencias entre especies y sexos predice patrones de territorialidad.

PAGE 2

INTRODUCTION Territoriality is a complex behavior that can affect the population dynamics of aggressive organisms (Eldridge 2001). Animals exhibit territorial behaviors to exclude competitors from an area and ensure access to vital resources, such as food (Maher & Lott, 2000). Territoriality is largely dictated by energetics: it is adaptive when the benefits gained from an area outweigh the energetic costs of de fense (Powers & McKee 1994). However, there is mounting evidence that territorial dynamics are more complex than can be explained by energetics alone (Powers & McKee 1994). Nonetheless, the value of a territory can be experimentally manipulated by changing the density of food resources (Stiles & Wolf 1970). Examining how resource density influences territoriality can help elucidate the factors that drive this behavior (Armstrong 1987). Many studies of the energy economics of territory defense have been conducted with hummingbirds because they provide an ideal model for manipulation (Camfield 2006). They are extremely sensitive to food stress due to their high energetic demands, so their energy budgets are balanced intricately with narrow margins for wast e (Dearborn 1998). Most importantly, hummingbirds are frequently pugnacious and engage in visible conflicts over nectar resources (Stiles & Skutch 1989). Hummingbirds employ a variety of feeding strategies and degrees of territoriality, likely due to n iche partitioning (Temeles et al. 2005; Fogden & Fogden 2005). Traplining species, including Campylopterus hemileucurus and Elvira cupreiceps feed on a series of widely dispersed flowers and tend to be non territorial (Stiles & Skutch 1989). Resident spec ies, such as Eupherusa eximia and Heliodoxa jacula, control territories around concentrated food resources from which they violently eject all intruders (Brown & Brown 1978, Stiles & Skutch 1989). Sex is another factor that influences territoriality. Hummi ngbirds also have strong sexual dimorphism for aggressiveness: males tend to be more territorial, while females more often trapline (Stiles and Skutch 1989). Male territoriality may drive female hummingbirds to utilize larger, less resource rich territorie s than males (Wolf et al. 1976). It is hypothesized that both degree and type of territoriality dependent upon a myriad of factors that may change seasonally, daily or even hourly for a particular hummingbird (Dearborn 1998; Fogden & Fogden 2005; Camfield 2006). Ecological circumstances such as food abundance and quality, competitor pressure, and predation risk are thought to influence territorial behavior, but the relative importance of each of these factors is unknown (Dearborn 1998). Intraspecific terri toriality should be more developed because individuals of the same species compete for the same resources (Lyon 1976; Powers & McKee 1994). There are a number of theories for the determinants of interspecific aggression: it may be shaped by body size or t he degree of threat of resource removal posed by competitors (Dearborn 1998). The interplay of these factors may result in the exclusion of individuals or the establishment of a dynamic equilibrium in which resources are partitioned (Stiles & Wolf 1970). A lthough there has been extensive speculation on the factors that shape territoriality in hummingbirds, it is not well understood which are most important or how they interplay between different species and sexes. Many studies of foraging and competitive be haviors in variable situations are needed to demystify the intricacies of hummingbird territoriality (Altshuler et al 2004). In this study, artificial hummingbird feeders were set up in three different configurations: with feeders concentrated close toget her, at a medium distance from each other, and then far apart. Decreasing the density of the food resources increases the amount of energy required for defense, and thus should alter territorial dynamics (Tiebout 1991). This study examined the how the diff erent resource densities affected the quantity of territoriality, feeder visitation rates by

PAGE 3

species and sex, and interspecific versus intraspecific territoriality. These observations indicate which factors dictating hummingbird territoriality and communit y structure are influenced by food density. MATERIALS AND METHODS Study Organisms Six species of hummingbird were observed at the feeders, four of which were considered in this study. The natural history, known feeding territorial strategies, and addit ional observations of these birds are as follows. Campylopterus hemileucurus, or the Violet Sabrewing, is the largest of the hummingbirds regularly spotted in this study, with a length of 15cm (Garrigues & Dean 2007). Only the brilliant purple male was reg ularly observed at the feeders. C. hemileucurus lives in low parts of mature wet forests, as well as in openings and forest edges (Stiles & Skutch 1989). This species is known to use a traplining feeding strategy (Fogden & Fogden 2005). C. hemileucurus is not considered particularly aggressive as its large size might suggest (Stiles & Skutch 1989). Elvira cupreiceps or the Coppery Headed Emerald, is 8 cm long, the smallest of the hummingbirds studied (Garrigues & Dean 2007). This species is found at forest edges and at all levels of mature wet montane forest (Garrigues & Dean 2007). Females tend to live in forest understories, and males in canopies, which could explain why only females were regularly observed at the feeders (Stiles & Skutch 1989). E. cuprei ceps is known to be a low reward trapliner, meaning it is not aggressive and favors low quality and widely spaced food sources (Fogden & Fogden 2005). Eupherusa eximia, or the Stripe Tailed Hummingbird, is 10cm long and identifiable by the distinct black a nd white pattern on its tail (Garrigues & Dean 2007). Like E. cupreiceps males forage more in the canopy and females in the understory, but both sexes will descend to shrubs at forest edges and in clearings (Fogden & Fogden 2005). Male E. eximia are known to be particularly aggressive and territorial (Stiles & Skutch 1989). Heliodoxa jacula or the Green Crowned Brilliant, is a shiny emerald hummingbird measuring 13cm (Garrigues & Dean 2007). This species is known to favor gaps, forest edges, and secondary growth. (Stiles & Skutch 1989). Males intermittently defend territories around large plants, while females prefer to visit understory flowers (Stiles & Skutch 1989). Both sexes of Lampornis calolaemus (Purple Throated Mountain Gem) and Phaethornis guy (Gr een Hermit), as well as female C. hemileucurus and male E. cupreiceps were seen with such infrequency that they were not considered in the analysis of this study. Study Site This study was conducted in Monteverde, Costa Rica at the Estacion Biologica, lo cated at an altitude of 1550 meters in the Holdridge life zone: lower montane rain forest. The study site was a mowed grassy area adjacent to forest edge to the south, and a garden to the north. The presence of a nearby Stachytarpheta phetofranzii ( Verbana ceae) bush and Callistemon spp. ( Myrtaceae) tree were additional food source for hummingbirds, and appeared to be favored particularly by E. eximia and E. cupreiceps. In contrast, H. jacula and C. hemileucurus were often spotted perching and interacting in the trees of the forest edge when not at the experimental feeders.

PAGE 4

Study Design Twelve miniature hummingbird feeders were constructed with a red base and yellow markings around the straw through which nectar was accessed. The feeders had a capacity of 60 mL, and were filled with between 40 50mL of sugar solution daily. The solution consisted of twenty percent sugar in water, equivalent to the sucrose concentration of nectar in most hummingbird flowers (Bolten & Feinsinger, 1978). The feeders were hung at t he beginning of each study period, and removed at the end. The feeders were hung in one of three arrangements (Fig 1). The feeders were hung from long lines of rope tied between trees or held up by bamboo poles, at a height of approximately 1.5 m. (a) (b) (c) F IGURE 1. Small, medium and large feeder arrangements, drawn to scale. Black dots indicate feeder placement. The arrangement was located in a grassy area adjacent to the lower lab between a forest edge a nd a garden at the Estacion Biologica de Monteverde. The small feeder grid (a), with 12 feeders in a 1m 2 area, measures 1 meter on each side. The medium feeder arrangement (b), with 12 feeders in a 25m 2 area, measures 5 meters on each side. The large feed er arrangement (c), with 6 feeders in a 50m 2 area, measures 5 by 10 meters. I sat approximately 3 meters from the feeder arrangement while recording data. All hummingbirds that visited the feeders were identified and the number of each species and sex th at visited the feeders was recorded. A visit was only recorded if there were at least 30s (approximate) between visits; a bird visiting multiple feeders or rapidly flitting in and out of the feeder area was only recorded once. Many of the data points were the same bird visiting multiple times over the course of each day and even the entire study, but individuals were not tracked. For territorial interactions, the species and sex of the aggressor and the victim were recorded. Territorial interactions were ea chase was defined as one bird chasing another away from the feeders, with both at least temporarily leaving the feeder site to carry out the dispute. These conflicts often involved vocalizations or physical contact. Displacement interactions were when a dominant bird quickly approached a feeder at which a submissive bird was already eating, and the submissive bird flew away immediately, making way for the other bird to eat without a ny dispute. Data was collected between 6:30 and 9:30 am on days with weather suitable for seeing many hummingbirds: some sunlight and no mist or rain. A total of 27 hours worth of data was

PAGE 5

collected: 12 hours over 4 days at the small feeder arrangement, 1 2 hours over 5 days at the medium feeder arrangement, and 3 hours in one day at the large feeder arrangement. This yielded observations of 80 territorial interactions and 740 feeder visitations. RESULTS There was no difference in the quantity of territor ial interactions observed at the different feeder arrangements (Mann Whitney U Test, x 2 =0.0154, df=1, p>0.05; Fig. 2). Only the small and medium arrangements were compared, as there was not sufficient data for comparison of the large arrangement. The small feeder arrangement had an average of 2.680.81 territorial incidents per hour, while the medium feeders had slightly more, with 2.831.26 per hour. The large feeder arrangement had 4.33 territorial incidents per hour. At the small feeder, the minimum numb er of territorial interactions seen in a single observation period was 5, and the maximum was 10. The medium arrangement had a greater range with a minimum of 2 and a maximum of 10. F IGURE 2. Number of territorial interactions between hummingbirds ob served per observation period at each of the feeder arrangements. Small feeder arrangement (n=4), medium feeder arrangement (n=5), large feeder arrangement (n=1), n equals number of days. Error bars represent 1 standard deviation.

PAGE 6

The proportion of inte rspecific versus intraspecific territoriality changed between feeder arrangements (X 2 =14, df=2, p=0.9013; Table 1). The proportion of intraspecific aggression increased as the feeders were moved further apart. The most intraspecific aggressive incidents to ok place between E. eximia (62.5% of all intraspecific aggression), which was also the most territorial species overall, being the aggressor in 35% of all territorial disputes. C. hemileucurus was particularly aggressive in the small feeder arrangement, ac counting for nearly half (45.2%) of territorial interactions in that setup, but it became far less so when the feeders were moved further apart. E. cupreiceps, was rarely territorial but was the victim in 36.3% of all disputes. T ABLE 1. All territorial interactions observed between six species of hummingbirds under small, medium and large feeder arrangements at Estacion Biologica de Monteverde between April 20th and May 3rd, 2011 Interactions are sorted by aggressor and victim. Shaded cells indicate in traspecific interactions (n=24) and white cells contain interspecific interactions (n=56). Victim Treatment Aggressor E. cupreiceps H. jacula P. guy L. calolaemus E. eximia C.hemileucurus Total V isits Total Hours Small E. cupreiceps 1 1 Total 31 251 12 H. jacula 5 1 # Intra 4 E. eximia 3 3 2 1 % Intra 12.9 C. hemileucurus 4 3 1 5 1 Medium E. cupreiceps 4 Total 33 352 12 H. jacula 6 3 1 3 1 # Intr a 11 L. calolaemus 1 1 % Intra 33.3 E. eximia 3 2 4 1 C. hemileucurus 2 1 Large H. jacula 3 Total 13 103 3 E. eximia 9 # Intra 9 C. hemileucurus 1 % Intra 69.2 There was a signific ant difference in the visitation by different species and sexes of hummingbirds of the three different feeder arrangements (X 2 =68, df=10, p<.00001; Fig. 3). Male mal e E. eximia visitation increased as the feeders were spaced further apart. In contrast, male C. hemileucurus visitation started with 21.5% at the small feeder arrangement, but fell to 11.1% visitation at the medium arrangements. Male H. jacula visitation also fell from 17.9% to 13. 1%.

PAGE 7

F IGURE 3. Proportion of feeder visitation by category of hummingbirds at the Estacion Biologica de Monteverde between April 20th and May 3rd, 2011. Compares visitors to the small feeder arrangement (n=251),the medium feed er arrangement (n=352), and the large feeder arrangement (n=103). DISCUSSION Energy economics dictates that defending a larger territory requires more energy; if resources become too dispersed guarding them becomes inefficient (Lyon 1976). The fact that increasing distance between hummingbird feeders did not alter the number of territorial interactions observed indicates that energy economics alone cannot explain the observed trends in hummingbird territoriality. It is possible that the differences in fe eder distances were not large between more distant feeders. However, the fact that feeder distance significantly altered visitation and proportions of interspecif ic versus intraspecific territoriality indicates that changes in feeder distance were great enough to otherwise alter hummingbird community structure around the feeders. Variation in visitation patterns between feeder arrangements could be a subtle indi cator of shifting territorial dynamics. Active territorial disputes are energetically expensive and increase interactions altogether (Dearborn 1998). They may do so b y forfeiting food resources already controlled by a dominant hummingbird, or by using different resources via niche partitioning. In this study, the changes in visitation proportions to different distributions of food resources may reflect how feeding stra tegies of different sexes of hummingbirds change with respect to resource density (Wolf et al. 1976). Male H. jacula and C. hemileucurus visited the distant feeders less

PAGE 8

often, supporting evidence that males prefer more densely concentrated resources. Male hummingbird morphology reflects this, as males have higher wing disc loading (WDL), or a lower wing length to body ratio, making flight more energetically expensive (Altshuler et al. 2004). Despite the fact that each feeder contained plentiful nectar, vis itors to the setup would frequently move between feeders, so distance between feeders probably impacted the energy expenditures for all visitors. Female E. cupreiceps and H. jacula visited the medium distance feeder arrangements more than the small setup, reflecting their tendency to specialize on more distant, low quality food resources not favored by males (Wolf et al. 1976). These dynamics are indicative of niche partitioning of nectar resources by different sexes of hummingbirds. However, female visita tion to the most distant feeders was at a lesser proportion than the medium distance. This could have been related to the high amount of aggression observed at the large feeder setup, which may have deterred females from the area. The single day during whi ch data was recorded for the large feeder arrangement, an individual male E. eximia was particularly pugnacious, and expelled many conspecifics and also interspecific competitors from the feeding area. Males are known to be dominant over females in most hu mmingbird species (Lyon 1976). If the territoriality of a male hummingbird was the determinant of female visitation, then perhaps that competitive hierarchy shaped feeder visitation. Further research is needed to understanding the relative importance of hi erarchy versus niche partitioning in determining which species and sexes access which food resources in hummingbird communities. Niche partitioning may also be a factor in shaping the differences in interspecific and intraspecific territoriality with resp ect to feeder density. Decreasing feeder density correlated with a decreasing proportion of interspecific territoriality, but inversely correlated with the proportion of intraspecific territoriality. Interspecific territoriality is thought to depend on com petitor features, such as size, potential for resource removal, and threat of territorial takeover, but these factors are unaffected by resource density ( Powers & McKee 1994) However, interspecific competition also drives differentiation in foraging habi tats in heterogeneous areas (Stiles & Wolf 1970). The feeders in the small arrangement were a homogenous resource: all were virtually equidistant from the adjacent forest and garden. However, as the feeders were spaced further apart, they became more heter ogeneous in their locations, and it appears that facultative interspecific niche partitioning developed. The feeders on the north side of the grid, closest to the garden, were favored by E. cupreiceps as they were also feeding on adjacent garden plants, fe eders on the south side were favored by H. jacula and C. hemileucurus which was often seen perching in the adjacent forest. This partitioning reduced interspecific competition, but because conspecifics were favoring the same few feeders, they encountered each other more and intraspecific competition intensified. This exemplifies how conspecifics share the same ecological requirements, which drives intraspecific territoriality to be more developed than interspecific territoriality (Lyon 1976). This study p rovides evidence that competition, in the form of territoriality, may be a driver of niche differentiation between sexes and species within hummingbird communities. While territoriality can be advantageous as it permits exclusive resource access, it is sti ll an energetically expensive behavior (Powers & McKee 1994). Niche partitioning allows similar exclusive resource access without the energetic costs of territoriality, so it is evolutionarily more favorable. What is most intriguing is that much of the evi dence of niche partitioning appears to be associated with food resource density, indicating that resource distribution may play a large role in shaping competition dynamics and subsequent specialization in hummingbird communities.

PAGE 9

ACKNOWLEDGEMENTS Thank you to my adviser, Pablo Allen, for much advice, extensive help with statistical analysis, and machete skills for fashioning bamboo poles. Thank you to Alan Masters for the idea for this study and for advice on methodology Thank you Gisella Fernndez Estr ada for helping me construct my hummingbird feeders, and to Jos Carlos Caldern Ulloa for always being available to answer questions and help me find all of the supplies I needed. LITERATURE CITED A RMSTRONG D.P. 1987. Economics of breeding rerritorial ity in male Calliope hummingbirds. The Auk 104: 242 253. A LTSHULER D.L., G. S TILES AND R. D UDLEY 2004. Of Hummingbirds and Helicopters: hovering costs, competitive ability, and foraging strategies. The American Naturalist 163: 16 25. B OLTEN A.B. AND F E INSINGER P. 1978. Why do hummingbird flowers secrete dilute nectar? Biotropica 10: 307 309. C AMFIELD A.F. 2006. Resource value affects territorial defense by Broad Tailed and Rufous hummingbirds. Journal of Field Ornithology 77: 120 125. C ARPENTER F.L. 1987. Food abundance and territoriality: to defend or not to defend? American Zoologist 27: 387 389. D EARBORN D.C. 1998. Interspecific Territoriality by a Rufours Tailed Hummingbird ( Amazilia tzacatl ): Effects of Intruder Size and Resource Value. Biot ropica 30: 306 313. E LDRIDGE S.A. 2001. Approaches to the study of territory size and shape. Annual Review of Ecology and Systematics 32: 277 303. F OGDEN M. AND P. F OGDEN 2005. The Hummingbirds of Costa Rica. Distribuidores Zona Tropical. Miami, FL. G ARRIGUES R. AND R. D EAN 2007. The Birds of Costa Rica. Cornell University. Ithaca, NY. L YON D.L. 1976. A montane hummingbird territorial system in Oaxaca, Mexico. The Wilson Bulletin 88: 280 299. M AHER C.R., AND D.F. L OTT 2000. A review of ecologica l determinants of territoriality within vertebrate species. American Midland Naturalist 143: 1 29. P OWERS D.P., AND T. M C K EE 1994. The effect of food availability on time and energy expenditures of territorial and non territorial hummingbirds. The Cond or 96: 1064 1075. S TILES G.F. AND A.F. S KUTCH 1989. A Guide to the Birds of Costa Rica. Cornell University. Ithaca, NY. T EMELES E.J., R.S., G OLDMAN AND A.U. K UDLA 2005. Foraging and Territory Economics of Sexually Dimorphic Purple Throated Caribs ( Eulampis jugularis ) on Three Heliconia Morphs. The Auk 122: 187 204. T IEBOUT H.M 1991. Daytime energy management by tropical hummingbirds: responses to foraging constraint. Ecology 72: 839 851. W OLF L.L., F.G. S TILES AND F.R. H AINSWORTH 1976. Ecolog ical Oragnization of a Tropical Highland Hummingbird Community. Journal of Animal Ecology 45: 349 379.