Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Delayed compensatory responses in a guild of ant-followers Touchton, Janeene M. 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2005-0672.pdf [ 4.48MB ]
JSON: 831-1.0092212.json
JSON-LD: 831-1.0092212-ld.json
RDF/XML (Pretty): 831-1.0092212-rdf.xml
RDF/JSON: 831-1.0092212-rdf.json
Turtle: 831-1.0092212-turtle.txt
N-Triples: 831-1.0092212-rdf-ntriples.txt
Original Record: 831-1.0092212-source.json
Full Text

Full Text

D E L A Y E D C O M P E N S A T O R Y R E S P O N S E S IN A G U I L D O F A N T FOLLOWERS  by  JANEENE M . TOUCHTON B.Sc. The University of Oregon, 1998  A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E REQUIREMENTS FOR T H E DEGREE O F  MASTER O F SCIENCE in T H E F A C U L T Y O F G R A D U A T E STUDIES DEPARTMENT OF ZOOLOGY  T H E U N I V E R S I T Y O F BRITISH C O L U M B I A July 2005  © Janeene M . Touchton 2005  ABSTRACT I studied the influence of competition on the maintenance of a dominance structured multi-species guild of ant-following birds. I explored the numerical and behavioral responses of bicolored (Gymnopithys naevioides)  leucaspis)  and spotted  (Hylophylax  antbirds several generations after the extirpation of the dominant ocellated  (Phaenostictus mcleannani)  antbird on Barro Colorado Island, Panama. I compared the  abundances and behavior of these species to data collected by E.O. Willis and others prior to the decline of ocellated antbirds on Barro Colorado, and to a nearby mainland control in Parque Nacional Soberania, where the complete guild of these ant-followers still exists. Populations of bicolored and spotted antbirds increased in density on Barro Colorado, completely compensating in combined biomass for the loss in overall biomass by ocellated antbirds. Historical records suggest that complete population turnover of these species occurred before density compensation was detectable. At ant swarms on Barro Colorado, the numbers of spotted antbirds doubled from historical records and in comparison to Soberania. The increased proportion of biomass of spotted antbirds at swarms on Barro Colorado compensated for the reduced proportion of biomass of ocellated antbirds. No shifts in microhabitat use by bicolored antbirds was observed after the loss of the dominant ocellated antbird. Bicolored antbirds foraged at similar rates, showed similar aggression towards conspecifics, and equal activity at ant swarms on Barro Colorado and in Soberania. Rates of aggression between bicolored and spotted antbirds on Barro Colorado, however, increased. Ocellated antbirds rarely interacted directly with spotted antbirds in Soberania, consistent with historical observations. Thus,  the limited swarm use by spotted antbirds historically on Barro Colorado and in Soberania likely results from indirect competitive pressure promoted by ocellated antbirds and mediated through direct interactions with bicolored antbirds. My results suggest that interspecific competition actively maintains guild structure in this complex tropical foraging association through direct and indirect interactions. Behavioral adaptations in guilds may occur over several generations, delaying the onset of compensatory responses. Detailed long-term experiments and/or comparative analyses are needed to fully understand the role of competition in the structuring of multi-species guilds in tropical forests.  TABLE OF CONTENTS Abstract  •'  ii  Table of Contents  iv  List of Tables  vi  List of Figures  vii  Acknowledgements  viii  CHAPTER 1 General Introduction: Compensatory Responses in Depauperate Communities  1  Introduction Literature Cited CHAPTER 2 Delayed Numerical and Behavioral Responses in a Guild of AntFollowing Birds , Introduction.; Methods Study areas and species Foraging resources Ant-following birds Statistical analyses and data presentation Results :.. Foraging resources Population responses of focal ant-followers Use of ant swarms Behavioral interactions at swarms. Discussion Effects of competition on the abundance of members of the dinger guild....: : The abundance of guild members at ant swarms... Behavioral responses in relationship to the abundance of guild members. Seasonal impacts on microhabitat use Conclusion Literature Cited Appendix 2.1 Species observed at ant swarms Appendix 2.2 ANOVA Table: Individuals per swarm.... Appendix 2.3 ANCOVA Table: Foraging rates Appendix 2.4 ANCOVA Table: Activity rates  1 6 9 9 13 13 16 18 20 22 22 23 24 25 28 28 31 32 34 35 51 57 59 60 61  iv  CHAPTER 3 General Conclusion Summary of thesis Avenues for future research Literature cited  62 62 63 66  v  LISTS OF TABLES Table 2.1: Estimated mean  E. burchelli  colonies and width  Table 2.2: Mean dry biomass of flushed arthropods Table 2.3: Estimated population densities of ocellated, bicolored, and spotted antbirds  36 37  38  Table 2.4: Maximum numbers of ocellated, bicolored, and spotted antbirds per ant swarm Table 2.5: Mean total number of birds and bird biomass per ant swarm Table 2.6: Prey preference of flushed arthropods by bicolored antbirds Table 2.7: Mean rates of aggression between bicolored antbirds and conspecifics, and other species  39 40 41  42  vi  LISTS OF FIGURES Figure 2.1: Niche partitioning at ant swarms  43  Figure 2.2: Map of Barro Colorado and Soberania, Panama  44  Figure 2.3: Detailed maps of Barro Colorado and Soberania study areas and census plots  45  Figure 2.4: Mean individuals of spotted, bicolored, ocellated, and other antbirds per swarms  46  Figure 2.5: Relative portion of total biomass at swarms of spotted, bicolored, ocellated, and other antbirds  47  Figure 2.6: Percentage of aggressive interactions directed from of to bicolored antbirds at swarms  48  Figure 2.7: Rates of foraging small and large prey per minute at swarms by bicolored antbirds  49  Figure 2.8: Rates of hops and flights per minute by bicolored antbirds at swarms  50  vii  ACKNOWLEDGEMENTS I would first like to extend my gratitude towards my supervisor, James N. M. Smith, for supporting my wish to work with the antbirds and encouraging me and mentoring me throughout this project in many ways. James N. M. Smith provided valuable insight to my research ideas and questions, detailed editing, and was an inspiration by sharing wonderful tales of his field adventures, along with a Scottish jig or two. He also provided the beautiful sketch of niche partitioning at an ant swarm. I am also indebted to my committee members, Peter Arcese and Diane Srivastava for invaluable advice, support, and editing. I have greatly benefited from and enjoyed many wonderful conversations about tropical ecology and antbirds with Egbert Leigh Jr., who encouraged me to continue with tropical studies. Martin Wikelski also provided insightful discussions about antbirds, motivation, assistance, and support throughout this project for which I am very appreciative. I am additionally grateful for the advice provided by Douglas Robinson and his willingness to share numbers of the population abundances of antbirds on Barro Colorado Island. This project would not have been possible without the tireless assistance from David Bradley, Paula Capece, Kyle Elliott, Jessica Knowlton, and Jonathan Reum, who willingly braved chiggers, ticks, thousands of ant stings, and spared no hesitation to brag profusely after sighting an ocelot. My time spent at the Smithsonian Tropical Research Institute was made enjoyable by the well-functioning and supportive staff, particularly Oris Acevedo. I am also grateful for the residents on Barro Colorado Island and in Gamboa for delightful conversations about tropical ecology, singing, and dancing.  viii  Susan Shirley provided much needed statistical advice, general support and wisdom for which I am thankful. Several other hut dwellers also provided support, advice, and made my experience at UBC enjoyable. This project was financially supported by a NSERC operating grant to James N. M. Smith. I am grateful for fellowship support from the University of British Columbia and the Smithsonian Institution. I would like to acknowledge my parents, Jim and Jo Touchton, for amazing amounts of encouragement and for supporting my love of birds and bugs. Finally, I would like to share my appreciation for my partner, Christian Ziegler. Christian helped me in all aspects of this thesis; from emergency ant-tracking, running errands, making coffee and dinner, fixing figures in Photoshop, spending hours upon hours talking about competition or differences between the tropics and temperate zones, and in sharing a love of wildlife and all of its many intricacies.  ix  CHAPTER 1 G E N E R A L I N T R O D U C T I O N : C O M P E N S A T O R Y R E S P O N S E S IN DEPAUPERATE COMMUNITIES  The absence of species in communities may lead to compensatory responses in remaining community members, such as density compensation and niche breadth shifts (defined below) (MacArthur et al. 1972). The absence of putative competitors in species depauperate communities has commonly been suggested as the primary cause for observed density compensation or niche shifts (e.g., Diamond 1978). Evidence of density compensation or niche shifts has thus been utilized to suggest the importance of. competition in structuring communities, particularly when there is simultaneous indication df both patterns (Diamond 1978).  Density Compensation  Some island faunal communities have been observed to contain equal or higher summed population densities of all species than those of mainland communities (e.g., Crowell 1962, MacArthur et al.1972, Case 1975, Diamond 1975, Morse 1977). MacArthur and others (1972) defined this pattern, 'density compensation', as the degree to which the summed species density on islands approaches mainland levels. Density compensation can result with or without niche shifts (a change in niche position towards space that is occupied by other species elsewhere) (MacArthur et al. 1972). Bottom-up or top-down processes can cause density compensation to occur (Faeth 1984). Greater focus has been granted to consumer-driven resource theory (e.g., Colwell and Futuyma  1  1971, Tilman 1982), and thus, bottom-up processes have generally been considered more important for observed density compensation (Faeth 1984).  Community-wide density compensation  Community wide density compensation contradicts conventional niche theory, which predicts that depauperate communities should have lower summed densities than mainland densities in similar habitats (MacArthur et al. 1972). This argument is based on the rationale that the species absent from depauperate communities should be more efficient at utilizing the niche space into which the remaining species expanded (Case and Gilpin 1974). Higher density per species, but lower overall community density would be expected under this scenario. Niche shifts are thought to occur in conjunction with community-wide density compensation, although it is difficult to measure niche dimensions for an entire community. Cox and Ricklefs (1977) observed broader niches in some Caribbean avifaunas compared to their mainland equivalents. These colonists, however, had broadniches and a widespread distribution to begin with. More convincing examples were provided in the Pearl Islands off Panama, where vertical shifts were observed in foraging heights of birds (MacArthur et al. 1972), and off New Guinea, where spatial expansions in elevational ranges were observed (Diamond 1970). Niche shifts, however, are not necessary for density compensation. Species populations may increase in density because resource availability increases. It is often assumed that the absence of interspecific competitors can liberate extra resources (Case et al. 1979). Conversely, top-down processes, such as a reduction of predation or  2  parasitism, could also increase densities of species in a community (Price et al. 1986, Faeth 1984).  Density compensation in species populations  When community-wide density compensation does not occur, some species populations still recognize higher densities on islands. Song sparrows melodia)  (Melospiza  occupy smaller territories on islands than mainlands, for example (Yeaton and  Cody 1974) although some of these populations fluctuate strongly (Arcese et al. 1992). As with community-wide density compensation, increases in density in species populations need not necessarily follow a niche shift. An increase in common snipe (Gallinago gallinago)  and wren  (Troglodytes troglodytes)  density on the Faroe Islands .  was attributed to an increase in suitable habitat (Bengtson and Bloch 1983). Blackthroated sparrow densities, however, were higher on islands in the Sea of Cortez not only from habitat differences, but other unknown factors (George 1987).  Problems underlying island-mainland comparisons  Early studies of island-mainland comparisons illustrating density compensation and niche shifts had several shortcomings. A lack of competitors was assumed to be the mechanism behind observed density compensation or niche shifts. This assumption reflected the close intertwining of competition theory and niche theory in the 1970's (Chase and Leibold 2003). Actual substantive evidence of an absence of competitors, however, was rarely presented and alternative mechanisms were only considered infrequently. For example, changes in predator communities can affect the diversity and  3  abundance of prey through direct and indirect interactions (Holt 1977, Schmitz et al. 2000). Climatic factors can also play a role in abundances of populations and niche characteristics (Bejer and Rudemo 1985), as can habitat characteristics or hidden patchiness that is not detectable by the observer (Abbott 1980, Simberloff and Abele 1982). Additionally, differences in parasitism (Price et al. 1986), colonization and dispersal (MacArthur and Wilson 1967), and disturbance regime (Sousa 1984) may also influence population densities and niche breadth. Many island-mainland comparisons failed to adjust for sampling error discrepancies, such as different census methods or temporal variation in sampling duration (Wiens 1992). There was also a lack of statistical rigor and insufficient use of null models (e.g., Simberloff 1978, Strong et al. 1979). A strong body of theoretical literature relating the overlap of species niches, coexistence, and abundance to competitive interactions thus developed (Chase and Leibold 2003). Empirical studies to test this theory have so far been limited to testing basic hypotheses about the very presence or absence of competition. Controlled competitor removal experiments have primarily documented niche shifts (review in Schoener 1983). Some studies have revealed density compensation; e.g., in coral reef fish guilds (Robertson 1996), surfperches (Schmitt and Holbrook 1990), and herpetofaunal assemblages (Rodda and Dean-Bradley 2002). In addition, communitywide density compensation has been documented in controlled experimental microcosms (McGrady-Steed and Morin 2000). Although the original theory of density compensation was largely based on patterns in avian communities (MacArthur et al. 1972), the mechanisms underlying  4  density compensation in avian guilds remain unclear. My objective in this study was to explore numerical and behavioral responses in a tropical avian guild of ant-following birds after the removal of a dominant guild member.  5  LITERATURE CITED Abbott, I. 1980. Theories dealing with the ecology of landbirds on islands. Advances in Ecological Research 11:329-371. Arcese, P. J., J. N. M. Smith, W. M. Hochachka, C. M. Rogers, and D. Ludwig. 1992. Stability, regulation, and the determination of abundance in an insular song sparrow population. Ecology 73:805-822. Bejer, B., and M. Rudemo. 1985. Fluctuations of Tits (Paridae) in Denmark and their relations to winter food and climate. Ornis Scandinavica 16:29-37. Bengtson, S. A., and D. Bloch. 1983. Island land bird population-densities in relation to island size and habitat quality on the Faroe Islands. Oikos 41:507-522. Case, T. J. 1975. Species numbers, density compensation, and colonizing ability of lizards on islands in Gulf of California. Ecology 56:3-18. Case, T. J., and M. E. Gilpin. 1974. Interference competition and niche theory. Proceedings of the National Academy of Sciences of the United States of America 71:3073-3077. Case, T. J., M. E. Gilpin, and J. M. Diamond. 1979. Overexploitation, interference competition, and excess density compensation in insular faunas. American Naturalist 113:843-854. Chase, J. M. and M. A. Leibold. 2003. Ecological niches: linking classical and contemporary approaches. University of Chicago Press, Chicago. Colwell, R. K., and D. J. Futuyma. 1971. On the measurement of niche breadth and overlap. Ecology 52:567-576. Cox, G. W., and R. E. Ricklefs. 1977. Species-diversity and ecological release in Caribbean land bird faunas. Oikos 28:113-122. Crowell, K. L. 1962. Reduced interspecific competition among the birds of Bermuda. Ecology 43:75-88. Diamond, J. M. 1970. Ecological consequences of island colonization by Southwest Pacific Birds, .1. Types of niche shifts. Proceedings of the National Academy of Sciences of the United States of America 67:529-536. Diamond, J. M. 1975. Assembly of species communities. Pages 342-444 in M. L. Cody and J. M. Diamond, editors. Ecology and evolution of communities. Harvard University Press, Cambridge, Massachusetts. Diamond, J. M. 1978. Niche shifts and the rediscovery of interspecific competition. American Scientist 66:322-331. 6  Faeth, S. H. 1984. Density compensation in vertebrates and invertebrates: a review and an experiment. Pages 491-509 in D. R. J. Strong, D. Simberloff, L. G. Abele, and A. B. Thistle, editors. Ecological communities: conceptual issues and the evidence. Princeton University Press, Princeton, New Jersey. George, T. L. 1987. Greater land bird densities on island vs. mainland - relation to nest predation level. Ecology 68:1393-1400. Holt, R. D. 1977. Predation, apparent competition, and structure of prey communities. Theoretical Population Biology 12:197-229. Martin, P. R., and T. E. Martin. 2001. Ecological and fitness consequences of species coexistence: a removal experiment with wood warblers. Ecology 82:189-206. Macarthur, R., J. M. Diamond, and J. R. Karr. 1972. Density compensation in island faunas. Ecology 53:330-342. Macarthur, R., and E. O. Wilson. 1967. The theory of island biogeography. Princeton University Press, Princeton, New Jersey. Mcgrady-Steed, J., and P. J. Morin. 2000. Biodiversity, density compensation, and the dynamics of populations and functional groups. Ecology 81:361-373. Morse, D. H. 1977. The occupation of small islands by passerine birds. The Condor 79:399-412. Munday, P. L., G. P. Jones, and M. Julian Caley. 2001. Interspecific competition and coexistence in a guild of coral-dwelling fishes. Ecology 82:2177-2189. Price, P. W., M. Westoby, B. Rice, P. R. Atsatt, R. S. Fritz, J. N. Thompson, and K. Mobley. 1986. Parasite mediation in ecological interactions. Annual Review of Ecology and Systematics 17:487-505. Robertson, D. R. 1996. Interspecific competition controls abundance and habitat use of territorial Caribbean damselfishes. Ecology 77:885-899. Rodda, G. H., and K. Dean-Bradley. 2002. Excess density compensation of island herpetofaunal assemblages. Journal of Biogeography 29:623-632. Schmitt, R. J., and S. J. Holbrook. 1990. Population responses of surfperch released from competition. Ecology 71:1653-1665. Schmitz, O. J., P. A. Hamback, and A. P. Beckerman. 2000. Trophic cascades in terrestrial systems: A review of the effects of carnivore removals on plants. American Naturalist 155:141-153. Schoener, T. W. 1983. Field Experiments on interspecific competition. American Naturalist 122:240-285. 7  Simberloff, D. 1978. Using island biogeographic distributions to determine if colonization is stochastic. American Naturalist 112:713-726. Simberloff, D., and L. G. Abele. 1982. Refuge design and island biogeographic theory effects of fragmentation. American Naturalist 120:41-50. Sousa, W. P. 1984. The role of disturbance in natural communities. Annual Review of Ecology and Systematics 15:353-391. Strong, D. R., L. A. Szyska, and D. S. Simberloff. 1979. Tests of community-wide character displacement against null hypotheses. Evolution 33:897-913. Tilman, D. 1982. Resource competition and community structure. Princeton University Press, Princeton, New Jersey. Wiens, J. A. 1992. The ecology of bird communities volume 2: Processes and variations. Harvard University Press, Cambridge. Yeaton, R. I., and M. L. Cody. 1974. Competitive release in island song sparrow populations. Theoretical Population Biology 5:42-58.  8  CHAPTER 2 D E L A Y E D N U M E R I C A L AND B E H A V I O R A L RESPONSES T O C O M P E T I T I V E R E L E A S E IN A G U I L D O F A N T - F O L L O W I N G BIRDS  INTRODUCTION  The coexistence of ecologically similar species has been of interest to ecologists because theory predicts that strong competitors will exclude others when resources are limiting (e.g., MacArthur and Levins 1967, May and MacArthur 1972, Roughgarden 1974). According to competition theory, intraspecific competition must be greater than interspecific competition for species coexistence (Lotka 1932). Resource partitioning by specialization on distinct resources (MacArthur and Levins 1967, Cody and Diamond 1975, Tilman 1982), temporal partitioning of the same resources (Armstrong and McGehee 1976), and spatial partitioning through spatial heterogeneity and variability enabling different dispersal and colonization or storage effects (Tilman 1982, Chesson 2000) all may promote species coexistence. Alternatively, frequency-dependent predation (Oaten and Murdoch 1975) or disturbance (Sousa 1983) may prevent populations from reaching levels where resources become limiting. Neutral models, on the other hand, that assume ecological similarity of species, predict short-term coexistence by chance alone (Hubbell 2001). The extent and circumstances under which competition structures communities is not entirely clear. Removal and addition experiments have sometimes been used to investigate the significance of interspecific competition in structuring guilds of potential competitors (reviews in Schoener 1983, Gurevitch et al. 1992). In such experiments, the magnitude  9  of behavioral or density compensation following competitive release is generally considered to indicate the interaction strength between competing species (Paine 1992). Microhabitat shifts are a common form of behavioral compensation (e.g., Chappell 1978, Griffis and Jaeger 1998) but may be unrelated to demography and may have limited influence on the structure of communities (Schmitt and Holbrook 1990). Density compensation has been demonstrated less often (e.g., Robertson 1996, Martin and Martin 2001, Munday et al. 2001), possibly because experiments ended too soon for effects to be detected (MacNally 1983). Furthermore, indirect interactions between species may obscure density compensation following species loss (Menge 1995, Werner and Peacor 2003), particularly in dominance-structured multi-species guilds (MacNally 1983, Palmer et al. 2003). In the tropics, tighter guild packing has historically been thought to result from competitive forces that are or were more intense than in temperate communities (Dobzhansky 1950, Connell 1978). This view largely stemmed from the notion that tropical climates are more stable than their temperate counterparts (MacArthur 1972). Resource stability is theoretically produced by climatic stability (Leigh 1982), in turn leading to levels of equilibrium in populations of consumers (Connell and Orias 1964). Nevertheless, tropical climates are not as stable as once thought. They recognize distinct seasonal fluctuations in rainfall (Leigh 1999) and additionally are subject to periodic climatic disturbances such as cyclic El Nino Southern Oscillations (Wright et al. 1999). The impact that tropical seasonality has on competitive interactions between species remains equivocal.  10  Here, I examine the role of interspecific competition in maintaining a dominancestructured Neotropical avian guild. Competition in multi-species avian guilds has rarely been demonstrated (Williams and Batzli 1979, Wiens 1992a,b), despite the historical importance of avian competition in community ecology (e.g., MacArthur 1958, Cody 1974, Wiens 1992a,b). In particular, I explore the role of interspecific competition several generations after the loss of a dominant competitor from a specialized guild of ant-following birds. Ant-following birds are unique to the Neotropics. They follow army ant swarms to forage on leaf-litter arthropods that flee from the ants. Nearly fifty species of birds (primarily of the families Thamnophilidae and Dendrocolaptidae) regularly follow army ant swarms (Willis and Oniki 1978). The spatial arrangement of an ant swarm permits different mechanisms of niche partitioning by ant-followers. For example, species use different perch types and can be segregated at ant swarms by their dominance rank (Figure 2.1; Willis and Oniki 1978). Dominant birds at ant swarms include large species of the "dinger" guild that occupy the two-dimensional foraging space at the advancing swarm front near the forest floor (Willis 1967, 1972, 1973). Other species use tree trunks or walk along the ground, and generally interact less with the dinger species than the latter do with each other (Figure 2.1, Willis and Oniki 1978). In Central America, the dinger guild consists of three species of regular antfollowers (listed in decreasing dominance): the ocellated bicolored (Gymnopithys  leucaspis),  and spotted  (Phaenostictus mcleannani),  (Hylophylax naevioides)  antbirds. The  ocellated antbird was extirpated on Barro Colorado Island, Panama, by 1978 (Willis and Eisenmann 1979), but still persists on the adjacent mainland in Soberania National Park  11  (Robinson et al. 2000). The dinger guild was studied extensively on Barro Colorado prior to the disappearance of ocellated antbirds (Willis 1967, 1972, 1973). This guild of ant-following birds was thus suitable for exploring numerical and behavioral responses several generations after the loss of a dominant competitor. In this study, I examined the population density, biomass, and behavioral use of ant swarms of professional ant-followers in the dinger guild on Barro Colorado to test for compensatory responses since the 1960s. I then compared the dinger guild on Barro Colorado to the dinger guild in Soberania as a reference to test the hypothesis that competition maintains avian guild structure (i.e., the relative abundances of guild members) through interactions in dominance hierarchies, affecting only immediate subordinates. Alternatively, I hypothesized that competition maintains avian guild structure affecting all guild members, both through direct interactions and indirect interactions mediated through guild members. To test these hypotheses, I addressed the following questions: 1) Do dominant guild members limit the population abundance and biomass of subordinates through competitive interactions? 2) Do dominant species limit resource use (the use of ant swarms), affecting behavioral interactions, and is this coupled to the overall abundance of guild members? 3) Are there strong seasonal influences on resource use that might influence interaction strengths between guild members? Under my general hypothesis that competition actively maintains guild structure through interactions with immediate subordinates, I predicted only bicolored antbirds would experience compensatory responses on Barro Colorado. I expected current population densities and biomass of bicolored antbirds to be higher on Barro Colorado than both in the 1960s and in Soberania, resulting from an increase in microhabitat use at .  12  ant swarms. Specifically, I expected bicolored antbirds on Barro Colorado to exploit larger prey items, experience lower rates of interspecific aggression, and suffer higher rates of intraspecific aggression than in Soberania. I expected bicolored antbirds to experience higher rates of activity (movement) at swarms on Barro Colorado than in Soberania either by avoiding or pursuing intraspecific attacks. If all guild members interact directly and indirectly, as under my alternative hypothesis, I predicted that other guild members, in addition to bicolored antbirds, would exhibit compensatory responses on Barro Colorado since the 1960s and as compared to Soberania. Under both hypotheses, I expected stronger interactions in the dry season between guild members because fewer arthropods should occur on Barro Colorado at that time (Levings and Windsor 1982).  METHODS Study Areas and Species Study Areas  This study was conducted on Barro Colorado Island (9°09'N, 79°51'W) and in Parque Nacional Soberania (9°09'N, 79°44'W) in the Republic of Panama (Figure 2.2). I worked in a 500 ha area on Barro Colorado with an extensive trail system (Figure 2.3a). Barro Colorado (1500 ha) was formed in 1914 by the flooding of Gatun Lake for the construction of the Panama Canal. The island has a long history of ornithological study (e.g., Chapman 1929, Eisenmann 1952, Willis and Eisenmann 1979). In Soberania, I worked in a 200 ha area of study encompassing "Limbo Plot," a 100 ha study plot, also with an extensive trail system (Figure 2.3b; Robinson et al. 2000). The Limbo basin lies  13  within 22,000 ha of continuous forest, approximately 8 km east of Barro Colorado (Robinson et al. 2000). I collected data during the wet season from May 2003 to August 2003, and the dry season from January 2004 to April 2004. The vegetation of Barro Colorado and Soberania is tropical moist forest (Holdridge 1967) with an average annual rainfall of 2600 mm, 90% of which falls during the wet season (May - December). The average daily temperature is 27°C (Leigh 1999).  Army Ants: Background  ,  Two species of Neotropical army ants forage in swarms, forming a dense carpet of ants along the forest floor and sub-canopy: Eciton  burchelli  and  Labidus praedator  (Formicidae: subfamily Ecitoninae, tribe Ecitonini). While traversing the forest floor in an advancing fan-like raid, these army ants flush leaf-litter arthropods (Rettenmeyer 1963). Escaping arthropods provide an accessible and abundant foraging resource for birds (Willis 1974). E. burchelli  follows a ca. 35-day rhythm governed by a regular endogenous cycle.  Two distinct life cycle phases, nomadic and statary, are synchronized to brood development (Schneirla 1933). During the nomadic phase (11-14 days), the entire colony moves to a new temporary nest (bivouac) site nearly every night after foraging each day. In the statary phase, the colony first nests in a protected site for about three weeks once larvae from the previous cycle have pupated. Halfway through the statary phase, the queen lays a new cohort of worker eggs (Schneirla 1971). During the statary phase, the ants have less brood to feed and only raid three days out of five (Willis 1967). Once the new workers eclose from their pupal cases and the eggs hatch, the colony begins another  14  nomadic cycle (Schneirla 1971). When a colony grows to a critical size, the queen lays a reproductive brood and the colony divides into two (Schneirla 1971) or occasionally three parts (J. Touchton, personal  observation).  About one of every three colonies also divides  at the end of the dry season (Franks 1982). If colonies fall below a threshold size where they can no longer raid efficiently enough to meet their food requirements, they shrink slowly and may die. Three independent estimates have been made of the number of E.  burchelli  colonies on Barro Colorado, and all approximate 50 colonies on the island, (i.e., ca. 3 per 100 ha) (Schneirla 1949, Willis 1967, Franks 1982). Less is known about L. praedator, as this species generally only forages above ground when water fills subterranean soil cavities in the wet season. It rarely appears above ground in the dry season (Willis 1967). E. burchelli  swarms are therefore a much more predictable foraging resource for ant-  following birds.  E. burchelli  Ant-following Birds:  is the focal ant species in my study.  Background  The 45-55g ocellated antbird dominates regular ant-followers in Central America (Willis 1974). During the 1960s, Willis (1973) studied the natural history and rapid decline of ocellated antbirds on Barro Colorado (Table 2.3a). Ocellated antbirds were extirpated on Barro Colorado by 1978 (Willis and Eisenmann 1979). The critical event causing this loss may have been a severe dry spell in 1968 (Willis 1974, Willis and Eisenmann 1979). No concurrent population studies of ocellated antbirds were performed on the mainland from 1960-1970. In the 1990s, density estimates of ocellated antbirds on the mainland were higher than original estimates on Barro Colorado in the  15  early 1960s (Table 2.3a). The 28-33g bicolored antbird is subordinate to the ocellated antbird; it has also been studied extensively on Barro Colorado (Willis 1967). It too declined in the 1960s but did not suffer extirpation (Table 2.3a). The 16-20g spotted antbird is considered a facultative ant-follower and is subordinate to both bicolored and ocellated antbirds (Willis 1972). Density estimates of spotted antbirds remained relatively constant on Barro Colorado during Willis' study on Barro Colorado (Willis 1972), and in Soberania since the late 1960s (Table 2.3a).  Foraging Resources Army Ant Colony Abundance E. burchelli  colonies were located by walking both on and off trails in both study  sites. Each study area was monitored continuously and regularly each field season to locate new colonies entering the area. Once a colony was located, it was monitored daily and its movements tracked nightly to new bivouac sites during the nomadic phase. Colonies were tracked until they were lost or no longer needed for study. In all,13 colonies were monitored on Barro Colorado and 13 in Soberania. E. burchelli  colony density was estimated by a weekly census of a 100 ha plot in  each study site (Figure 2.3). During the wet season, the census lasted eight weeks on Barro Colorado and six weeks in Soberania. In the dry season, the census lasted five weeks on Barro Colorado and seven weeks in Soberania. Each plot consisted of 17 lines marked with flagging tape 100 meters apart north to south, and 200 meters apart east to west. Flags were placed every 10 meters on each line. The lines in the plot were walked  16  from 1400h to 1700h to increase chances of seeing a column of ants (by the afternoon, they had traveled further from their overnight bivouac).  Flushed Arthropod Abundance  I estimated flushed arthropod abundance at ant swarms in both sites during the wet and dry seasons. To sample arthropods, 0.62m by 0.40m sampling quadrats were placed on the leaf litter three meters in front of the center of approaching ant swarms. The observer sat on stools 0.38m in height next to the quadrat. When ants came within one meter of the front edge of the quadrat, the observer scored all arthropods either flushed into or out of the quadrat for the next ten minutes. Arthropods were identified to taxonomic order or family. Ten minutes allowed the entire depth (1-2 meters) of the swarm front to pass through the quadrat, as the ant colony progresses at a rate of ca. 0.25m per minute (Willis 1967). This procedure was replicated three to five times per day per nomadic swarm between 0900h and 1 lOOh. The quadrats were constructed of 0.025m in diameter white PVC tubing connected by corner joints. The corner joints held the tubing off the ground, allowing the raiding army ants to pass underneath. These dimensions produced a fairly large quadrat that allowed us to score all movements of arthropods consistently. All observers were trained to score arthropods within a 90% precision fate. We observed no avoidance behavior towards the quadrats by the army ants or arthropods.  17  Ant-following Birds Ant-following birds were captured with two to three 6 m mist-nets placed along the entire front of moving ant swarms (Swartz 1995) ant then marked over a period of two days at each newly discovered colony. All mist-netting occurred between 0700h and 1400h. Captured birds were fitted with a unique combination of a single aluminum and two to three colored celluloid bands. Additional banding was performed as needed when the ants entered a new territory with un-banded ant-followers.  Population densities of focal species  Adult populations of bicolored and ocellated antbirds were estimated in the study areas during both the wet (2003) and dry (2004) season from swarm attendance. As bicolored and ocellated antbirds are obligate ant-followers, I assumed that they were always at swarms. Thus, my estimates are conservative. If some individuals foraged elsewhere, I underestimated these densities. Because spotted antbirds are thought to follow army ant swarms only facultatively (less than 50% of the time; Willis and Oniki 1978), population density estimates could not be generated using the same method as for bicolored and ocellated antbirds. During this study, D. Robinson estimated population density of spotted antbirds on Barro Colorado by territory mapping in a 50 ha area. He generously provided these estimates (Table 2.3a). For a mainland density estimate for spotted antbirds, I used Strysky's (2003) numbers from 1998 - 2000 in the 100 ha Limbo plot. Again, this estimate was generated by mapping the movements of known individual (Table 2.3a). As is customary  18  for territory mapping, no error was assumed and any unmarked birds in the mapped area were assumed to be single individuals.  Microhabitat use of ant swarms E. burchelli  colonies were visited daily for 1 - 4h between 0700h and 1200h, the  period of greatest bird activity (Willis 1967). If the colony was swarming, one or two observers stood at least five meters to the side of and behind the swarm front. Individuals of all species of birds present were then identified by their band combinations. If band combinations could not be read or bands were absent, the number of individuals per species was estimated using sightings of species and/or counts of species-specific calls. At the end of the observation period, the ant swarm width was estimated by a tape measure to estimate the available foraging space for the birds. If a colony was not swarming upon arrival, it was checked for activity until 1400h.  Behavior at swarms  Marked bicolored antbirds were observed at ant swarms between 0700h and 1200h at both sites and seasons. Continuous watches were performed on focal individuals for 5-25 minutes divided into five-minute segments. Individuals were observed sequentially at a swarm until all possible marked bicolored antbirds had been watched. Individuals were observed only once per season. Observations ceased during heavy rain as all bird and ant activity stopped. Activity levels, foraging rates, and numbers of aggressive interactions by bicolored antbirds were scored during observations. Activity rates were counts of the  19  numbers of hops and flights 3 to 10 meters long per minute. Flights > 10 m were not included in analyses here because they usually entailed the focal bird flying out of view of the observer and leaving the swarm. Foraging rates were estimated from the number of prey captured during an observation. Prey length and prey order or family were scored when possible. For analysis, prey were scored as small (0.5 - 1 x bill length) and large (>1 x bill length). Although prey smaller than half the length of the bird's bill were also consumed, they were not analyzed due to their small contribution to the overall biomass consumed by the birds. Aggression took the form of supplanting attacks where a dominant bird physically hit a subordinate, driving it off its perch, and passive displacements where the subordinate moved away from an approaching dominant. Supplantings and displacements were combined for analysis. Statistical analyses and data presentation Foraging  Resources  I used two-way analysis of variance (ANOVA) to test for the effects of the independent factors site and season on the mean number of army ant colonies per 100 ha and the mean swarm width of the colonies. Army ant colony density means were generated from weekly census means, and the mean swarm width of each colony was estimated from daily measurements per colony. Numbers of flushed arthropods were converted to dry biomass following Gruner (2003). I compared the dry biomass of the flushed arthropod communities at army ant swarms between sites and seasons with a two-way crossed analysis of similarity (ANOSIM) with site and season as factors, using the PRIMER software package (Clarke and Gorley 2001). ANOSIM is a test built on a nonparametric permutation procedure 20  combined with a general randomization approach (Clarke and Warwick 1994). The ANOSIM test uses the ratio of the mean within group to between group distances to calculate the test parameter Global R, which falls between 0 and 1. Values approaching 1 indicate similarity where values approaching 0 indicate dissimilarity. To determine site and seasonal differences for each arthropod group, I performed a two-way MANOVA, followed by two-way ANOVA with Fischer's least significant difference (LSD) post-hoc tests.  Ant-following bird population abundances  I estimated adult population densities of bicolored and ocellated antbirds in the study areas during both the wet (2003) and dry (2004) season using Bowden's Model Estimation (Bowden 1993) in Program NOREMARK (White 1996). The Bowden Model Estimation is a mark-recapture analysis that uses the number of banded birds, the number of banded but unidentified bird sightings, and the number of unhanded bird sightings to estimate the total bird density using the sampled area and its 95% confidence intervals. Heterogeneity of sighting probabilities is assumed. I added a 400-meter border to the study areas to account for the large home ranges of these species (Willis 1967, 1973) and then calculated the density per 100 ha from the Bowden Estimate.  Ant-following bird use of ant swarms and behavioral interactions  I examined the mean number of all ant-following birds present at ant swarms and their total mean biomass per swarm with two-way ANOVAs. Sites and seasons were the two factors. Biomass data were square-root transformed to normalize the data. I then  21  used a two-way MANOVA followed by two-way ANOVA with LSD post-hoc tests (Appendix 2.2) to explore differences in attendance of the four species groups of birds at ant swarms between sites and seasons. The composition of attendant birds at ant swarms varied daily. Every new ant swarm was considered an independent unit for analyses. Bicolored antbird foraging rates, activity rates, and aggressive interactions at ant swarms were analyzed with two-way MANCOVA. Observation length was included as a covariate. These data are presented as mean rates (activities per minute). Data on foraging rates and aggressive interactions were rank transformed prior to running the analyses to meet the assumptions of MANCOVA (Zar 1991). After each MANCOVA, I used two-way ANCOVAs with LSD post-hoc tests to help ascertain which variables were responsible for significant main effects. ANOVA tables are in Appendices 2.3 and 2.4. Preferences for flushed arthropod prey was estimated by calculating selection ratios for each prey item (proportion of occurrence in diet/relative biomass in flushed arthropod community (Manly et al. 2000). A standardized selection ratio over 0.14 here indicates the respective prey item is preferred, while a selection ratio less than 0.14 indicates avoidance (Manly et al. 2000). To test for differences in prey preference at swarms between sites and seasons, selection ratios for each prey item were compared using chi-squared tests (after Manly et al. 2002).  RESULTS Foraging resources E. burchelli  colony density varied strongly across seasons. On average, colony  density in the dry season was nearly double the density in the wet (Table 2.1). The mean  22  swarm width of E.  burchelli  colonies, however, did not differ reliably from season to  season, although averaged 26% wider in Soberania than on Barro Colorado (Table 2.1). Four taxa dominated the flushed arthropods at ant swarms. Tettigoniidae made up over 40% of the biomass, followed by Gryllidae (ca. 32%), large Araneida (ca. 18%), and Blattodea (ca. 7%). The composition of flushed arthropods at swarms differed between seasons (ANOSIM, Global R = 0.36, p < 0.01), but not sites (ANOSIM, Global R = 0.0, p > 0.05). Specifically, samples averaged 44% grams of dry weight more arthropods in the wet season due to higher numbers of Tettigoniidea, Araneida, and Blattodea (Table 2.2). Gryllidae and Isopodea, however, were 2.3 and 10 times less dense in the wet season than the dry, respectively (Table 2.2). I examine preference of these arthropods by the birds later.  j  Population responses of focal ant-followers  Overall densities for bicolored and spotted antbirds have changed since the 1960s on Barro Colorado. I estimated the bicolored antbird density on Barro Colorado to be 12.6 individuals per 100 ha, 66% higher than Willis' 1961 estimate, an increase of ca. 150 g/100 ha (Table 2.3b). This difference could be statistically significant if Willis estimated densities of bicolored antbirds with similar accuracy and precision to my study. After the extinction of ocellated antbirds in the late 1970s, spotted antbird density on Barro Colorado appeared to remain constant at 45 - 50 individuals per 100 ha until the mid-1990s, when numbers began to rise (Table 2.3a). According to recent estimates, spotted antbird density apparently has risen by 44% on Barro Colorado since the mid1990s, an increase of ca. 350 g/lOOha (Table 2.3a, D. Robinson personal  comm.).  The  23  combined increase in biomass of bicolored and spotted antbirds per 100 ha more than doubles the loss in biomass (ca. 200 g/100 ha) by ocellated antbirds. In Soberania, densities of ocellated, bicolored, and spotted antbirds have remained more constant over time, but they differ from Barro Colorado. In Soberania, I estimated numbers of ocellated antbirds to be 80% higher than on Barro Colorado before their decline there (Table 2.3b). Again, assuming that previous estimates contained a similar amount of error as my estimates, this difference would be significant. My density estimate for bicolored antbirds in Soberania is nearly double that on Barro Colorado , (Table 2.3b). In contrast, however, recent estimates of spotted antbirds in Soberania (Strysky 2003) are 16% less than on Barro Colorado, or ca. 180 g/lOOha less (Table 2.3a, D. Robinson personal  comm.).  Use of ant swarms Although the total number of individual birds and biomass at ant swarms on Barro Colorado could not be compared to the 1960s, differences in swarm attendance by bicolored and spotted antbirds were apparent. The number of bicolored antbirds at swarms on Barro Colorado was 30% higher than in the 1960s (Table 2.4). Spotted antbirds attended swarms on Barro Colorado more than twice as often as they did in the 1960s (Table 2.4). The combined number of individual birds at ant swarms did not differ between Barro Colorado and Soberania, but the combined biomass did. Biomass was ca. 30% less per swarm on Barro Colorado than in Soberania (Table 2.5). This biomass difference reflects site differences in species composition of the ant-following bird community  24  (Figure 2.4). On Barro Colorado there were, on average, 27% fewer of the large bicolored antbird per swarm then in Soberania, but 140% more of the small spotted antbird per swarm (Figure 2.4). Relative proportion biomass per species group differed at swarms on Barro Colorado compared to Soberania. The small spotted antbirds accounted for 20-25 percent of the total biomass at ant swarms on Barro Colorado but only 7-10 percent in Soberania (Figure 2.5; Kruskal Wallis, x = 84.69, df = 1, p < 0.001, n = 196). The large 2  ocellated antbird accounted for 10-15 percent of the total biomass at ant swarms in Soberania, equal to the difference in biomass of spotted antbirds between Barro Colorado and Soberania (Figure 2.5). Thus, on Barro Colorado, spotted antbirds have increased in number at swarms, filling the "gap" left by the absence of ocellated antbirds. I will suggest below that dominance relations between the species help to explain these patterns. Ant-following bird communities were similar between the wet and dry seasons at ant swarms (Table 2.5). Despite this overall similarity, there were some differences within the species groups. Nearly 40% more spotted antbirds attended swarms in the dry season than in the wet (Figure 2.4). The obligate ant followers, however, were more abundant at swarms in the wet season with 16% more bicolored antbirds, and 8% more ocellated antbirds (Figure 2.4).  Behavioral interactions at swarms Foraging  25  The overall foraging rates of bicolored antbirds did not differ between sites and seasons (Figure 2.6). Slightly more (44%) small prey, however, were captured in Soberania as compared to Barro Colorado (Figure 2.6a). Prey preferences of bicolored antbirds were similar between sites (Table 2.6a). The standardized selection ratios indicate that bicolored antbirds typically selected similar prey (i.e., Tettigoniidae, Blattodea, and large Araneida) in both sites. Bicolored antbirds in Soberania also selected weakly for Coleoptera (Table 2.6a). Despite preference for similar prey between sites, the strength of the selection ratios varied. For example, bicolored antbirds on Barro Colorado selected the three most preferred prey items (Tettigoniidae, Blattodea, and large Araneida) more strongly than in Soberania (Table 2.6a). Slightly more prey items were selected for in the dry season than the wet season. For example, Blattodea were not selected in the dry season (Table 2.6b). Of the prey items selected, selection ratios were significantly higher in the dry season than the wet (Table 2.6b). In the dry season, large prey (Tettigoniidae, Blattodea, and large Araneida) were less abundant than in the wet (Table 2.2). (We did not detect any small Blattodea in our sampling.) Although they were consistently abundant, small Gryllidea and small Araneida were not preferred by bicolored antbirds (Table 2.6b).  Aggressive interactions  Overall rates of aggression involving bicolored antbirds differed between Barro Colorado and Soberania (MANCOVA, F  6147  = 4.08, p = 0.001). There were 43% more  interspecific aggressive interactions on Barro Colorado (Table 2.7 a). Bicolored antbirds  26  also supplanted spotted antbirds three times as often on Barro Colorado as in Soberania (Table 2.7 a). The directions of arrows in Figure 2.8 illustrate the direction of aggression either from or to bicolored antbirds on Barro Colorado and in Soberania 2003-2004, and on Barro Colorado 1960-1961 (Willis 1967). The proportion and targets of aggressive interactions involving bicolored antbirds on Barro Colorado in the 1960s strongly resembles those in Soberania 2003 - 2004. In contrast, fewer current aggressive interactions occur with other bicolored antbirds on Barro Colorado, where a higher proportion of aggression involves spotted antbirds (Figure 2.7). I expected both interspecific and intraspecific aggressive interactions to increase during the dry season due to lower abundances of preferred arthropods. Levels of aggression involving bicolored antbirds did vary seasonally (MANCOVA, F  6147  = 5.83, p  < 0.001). Intraspecific aggression, however, did not change (Table 2.7 b). Aggression towards spotted antbirds by bicolored antbirds was higher in the dry season, but aggression involving all other species was higher in the wet season (Table 2.7 b).  Activity  Bicolored antbirds were slightly more active in Soberania than on Barro Colorado (MANCOVA, F  2151  = 4.58, p = 0.012), and they were more active in the dry season  (MANCOVA, F  2151  = 4.59, p = 0.012) (Figure 2.8). Bicolored antbirds hopped 16%  more often in Soberania than on Barro Colorado, and 39% more in the dry season than in the wet season (Figure 2.8a). Bicolored antbirds also flew 27% more often at swarms in Soberania and 50% more during the dry season (Figure 2.8b).  27  DISCUSSION Below, I discuss the numerical and behavioral responses of the ant-following dinger guild on Barro Colorado Island, Panama, several generations after the loss of their dominant competitor, the ocellated antbird. I elaborate on possible mechanisms behind density and biomass compensation at the population level, and at the microhabitat level by abundances at ant swarms. I further discuss direct and indirect behavioral interactions as they may relate to population abundances, and how seasonal variation may influence these interactions.  Effects of competition on the abundance of members of the "dinger" guild  Population estimates on Barro Colorado from 2003-2004 suggest that the loss of ocellated antbirds has led to compensatory gains in the absolute abundance of bicolored and spotted antbirds. Density estimates for these two guild members exceed 1961 estimates by about 40-60%. This amounts to a gain in combined biomass of nearly double that from the loss of biomass incurred through the extirpation of ocellated antbirds from Barro Colorado (Table 2.3). If, however, biomass is scaled to account for differences in food consumption rates of differing body sizes (Walsburg 1983), bicolored and spotted antbird combined biomass compensates for about 100% of the missing biomass by ocellated antbirds. When compensatory responses at the population level have been measured in competitor removal experiments, complete or over-density and biomass compensation has rarely been documented (Rodda and Dean-Bradley 2002). Most studies have revealed under-compensation (Hairston 1981, Holbrook and Schmitt 1995, Robertson 1996, Munday et al. 2001).  28  The over-compensation in density at the population level observed here could be an artifact of the timing of the initial density estimates on Barro Colorado. Willis (1974) documented a decline of ocellated antbirds from 1961 to 1971. Bicolored antbird density also declined in the 1960s, but appeared to stabilize in the early 1970s (Willis 1974). Spotted antbirds, on the other hand, maintained a similar density throughout the 1960s (Willis 1974). It is possible that the initial population density estimates generated by Willis (1974) for these guild members occurred after ocellated and bicolored antbirds already had begun to decline on Barro Colorado. Willis (1974) states that, in the 1960s, numbers of bicolored and ocellated antbirds were as high or higher on the mainland than on Barro Colorado. Estimates from the mid 1990s and this study document higher numbers of ocellated and bicolored antbirds in Soberania than were on Barro Colorado during the 1960s. It remains possible that ocellated and bicolored antbirds have always been less numerous on Barro Colorado than in Soberania. Several competitor removal experiments have failed to demonstrate complete density compensation (Schoener 1983, Gurevitch et al. 1992). Often, long periods are needed for density compensation to occur, particularly for long-lived organisms. Surfperch under-compensated in density after one generation of competitive release (Schmidt and Holbrook 1990). Two generations have been suggested as a suitable time scale to measure such population responses (Yodzis 1988). A maximum compensatory response in damselfish took four generations, however, and was only apparent after complete population turnover (Robertson 1996). Complete population turnover has also been proposed as a minimum time frame to assess population stability (Connell and Sousa 1983).  29  Bicolored and spotted antbirds on Barro Colorado took over 20 years to increase in abundance following the extirpation of ocellated antbirds. These species of antbirds are known to reach ages of at least 12 and 15 years (Willis 1983). By 1994, these populations may have undergone a complete turnover with only slight increases in abundances (Robinson 1999). There are several possible explanations for the delay in population increases by bicolored and spotted antbirds on Barro Colorado, including fluctuations in foraging resources, recruitment limitation, and slow behavioral shifts. If such fluctuations were extreme, they could exacerbate the time required for population abundances to increase. Fluctuations in foraging resources may be responsible for not only the initial declines of ocellated and bicolored antbirds, but also for slow increases in bicolored and spotted antbird density on Barro Colorado. The abundance of ant-following birds may be closely coupled to army ant colony density. In Cocha Cashu, Peru, forest flooding influenced the availability of army ants, which could have affected adult survival of antfollowing birds (Willson 2004).  E. burchelli  colony density was similar between Barro  Colorado and Soberania in this study (Tables 2.1), but was lower than previous estimates for Barro Colorado (Schneirla 1949, Willis 1967, Franks 1982). The population of E. burchelli  has been proposed to be one of the most stable on Barro Colorado (Franks  1982). However, my E.  burchelli  density estimates indicate that numbers may fluctuate  more than suggested by Franks (1982). Climatic fluctuations may play a role here. Repeated food shortages following unusual seasonal extremes have caused famine for some species on Barro Colorado (Leigh 1999). S. Powell  {personal comm.)  noted fewer  large colonies and evidence of colony fission on Barro Colorado in 2002 following the  30  harsh dry season of 2001. Willis (1974) also found it difficult to locate E.  burchelli  colonies on Barro Colorado in 1968 following an extremely dry year. High nest predator densities on Barro Colorado may hinder recruitment in bicolored and spotted antbirds and thus slow increases in population abundance. According to Glanz (1982, 1990) and Sieving (1992), common mammalian nest predators are more abundant on Barro Colorado than in Soberania. Experiments using artificial nests suggested that ant-following birds, including spotted and bicolored antbirds, experienced higher levels of nest predation on Barro Colorado than in Soberania (Sieving 1992). The validity of artificial nest studies, however, has been questioned (review in Moore and Robinson 2004). Nevertheless, low juvenile recruitment in obligate ant-following birds in Peru contributed to large population fluctuations there (Willson 2004). Nest predation may ultimately be one reason why bicolored antbirds have not been able to increase to greater densities on Barro Colorado.  The abundance of guild members at ant swarms  My results suggest that ocellated antbirds reduce the spotted antbird swarm use by forcing them to the periphery (zone C; Figure 2.1). More than twice as many spotted antbirds foraged at ant swarms on Barro Colorado as they did on Barro Colorado in the 1960s and in Soberania during this study. I did not expect spotted antbirds to display density or behavioral compensation at swarms. Spotted antbird swarm attendance was expected to be limited, as they were previously only known to forage at ant swarms when they passed through their territories. Usually, there were only two or three individual spotted antbirds per swarm (Willis 1972). Given there was a similar composition of  31  attendant ant-followers on Barro Colorado in the 1960s as I observed in Soberania, spotted antbirds will have compensated for 100% of the proportion of biomass missing at swarms on Barro Colorado after the extirpation of ocellated antbirds (Figure 2.5). Bicolored antbirds may also limit ant swarm use by spotted antbirds. The number of bicolored antbirds per swarm was higher in Soberania than on Barro Colorado (Table 2.4). Despite this fact, mean numbers of bicolored antbirds at swarms on Barro Colorado in the 1960s were lower than in this study while numbers of spotted antbirds per swarm were similar to those in Soberania today. Thus, ocellated antbirds may have more influence on swarm use by spotted antbirds than bicolored antbirds.  Behavioral responses in relationship to abundance of guild members  Ocellated antbirds could limit spotted antbird swarm use through either direct or indirect interactions. Willis (1972) noted that spotted antbirds maintain a sufficient distance from ocellated antbirds that they are rarely supplanted or displaced by them. Bicolored antbirds, however, actively supplant spotted antbirds away from prime foraging spaces at ant swarms (represented by zone B in Figure 2.1; Willis 1972). My observations from Soberania confirm these observations (Figure 2.7). It thus seems plausible that ocellated antbirds limit spotted antbird swarm use indirectly, possibly through interactions with bicolored antbirds in zones A and B (Figure 2.1). Such competitor induced behavioral plasticity has also been shown to occur with larval green frogs (Relyea 2000). Bicolored antbirds behaved differently at swarms on Barro Colorado and in Soberania, suggesting that other changes have occurred on Barro Colorado since the loss  32  of ocellated antbirds. Bicolored antbirds still preferred and captured large prey items on both Barro Colorado and in Soberania, yet they foraged on smaller prey more frequently in Soberania than on Barro Colorado. This result suggests that some bicolored antbirds in Soberania are foraging in poorer locations (zones B or C; Figure 2.1) at swarms than on Barro Colorado where bicolored antbirds may continuously forage in zone A (figure 2.1). Rates of intraspecific aggression were similar between sites, but the number of flights by bicolored antbirds at swarms was higher in Soberania than on Barro Colorado. Bicolored antbirds maintain areas at swarms by displacing or supplanting subordinate birds that come within 2m (Willis 1967, J. Touchton personal  observation).  More often,  however, subordinate birds simply avoid dominants; thus supplanting or displacement need not occur. The frequent flights by bicolored antbirds at swarms in Soberania suggests that birds have to change their foraging locations more often than on Barro Colorado. These changes in location may be in response to ocellated antbirds coming and going from swarms and taking over the most profitable foraging areas when they do so. These behavioral differences in bicolored antbirds suggest that profitable foraging locations for spotted antbirds in zone B, for example (Figure 2.1), are available at swarms on Barro Colorado but not in Soberania. Shifts in microhabitat use following competitive release are generally immediate (review in Schoener 1983, Holbrook and Schmidt 1995, Griffis and Jaeger 1998). For example, information gathering (e.g., learning, particularly of reliable information) affected the nature of competitive interactions between three coexisting hummingbird species (Sandlin 2000). Williams and Batzli (1979) demonstrated that red-bellied woodpeckers utilize forest they are normally excluded from by red-headed woodpeckers,  33  following removals of the latter species. During less productive years, red-headed woodpeckers also vacate the forested area where the two species competed. Perhaps the information red-bellied woodpeckers obtained over the years supported such rapid shifts in behavior when their dominant competitor was absent. The short duration of many studies may not have revealed delays in behavioral adaptations allowing for shifts in microhabitat use. Bicolored and spotted antbirds took at least ten years to display recognizable behavioral shifts on Barro Colorado. Willis did not report significant changes in microhabitat use by bicolored or spotted antbirds during his work in the 1960s (Willis 1972, 1974). The decline of bicolored antbirds on Barro Colorado may have contributed to an increase in use of army ant swarms by spotted antbirds. It is possible that spotted antbirds took several years to learn that space at ant swarms was sufficiently reliable to leave their territories to look for these swarms.  Seasonal impacts on microhabitat use Seasonal limitation in foraging resources is known to affect species interactions within guilds (Terborgh 1983). Seasonal variation in the number of army ant colonies and leaf litter arthropods (Levings and Windsor 1982, Table 2.2) may therefore affect seasonal differences in behavior of ant-following birds. The greater number of ant colonies and longer foraging periods in the dry season may compensate for the fact that there are fewer flushed arthropods during this time. Afternoon rains generally interrupt army ant foraging behavior in the wet season (J. Touchton personal  observation),  forcing  ant-following birds to "fill-up" in the morning. Greater numbers of ocellated and bicolored antbirds were present at swarms during the wet season in the morning. More  34  intensive foraging by obligate ant-followers at swarms in the wet season could explain the smaller numbers of spotted antbirds at swarms during this time. Increased competition for swarm space in the wet season may not be the only reason why spotted antbirds attended swarms more in the dry season. During the dry season, spotted antbirds are not restricted to a nest area, allowing them to travel further to ant swarms. Although flushed leaf-litter arthropods are less abundant at ant swarms in the dry season, they are also less abundant away from swarms (Levings and Windsor 1982).  CONCLUSION This study suggests that interspecific competition actively maintains structure in this tropical guild by limiting the use of shared resources and setting overall population density. Following the prolonged absence of a dominant guild member, density compensation may have occurred through delayed direct and indirect interactions between remaining guild members. Surprisingly, complete population turnover occurred before evidence of density compensation was detectable. Delayed behavioral adaptations of guild members could have contributed to the delay in density compensation. Other factors, such as recruitment limitation could have also prevented more immediate density compensation. Moreover, seasonal variation in resource levels and variability in life history parameters of guild members probably influenced the strength of competitive interactions. Long-term experiments and comparative studies measuring population abundances and behavioral interactions are thus important to fully understand the role of competition in the maintenance and re-structuring of multi-species tropical guilds.  35  Table 2.1: Mean number of E. burchelli colonies per 100 ha, mean width of swarms per colony (meters), and two-way ANOVA results. Factors: a) site; Barro Colorado Island (BCI) and Soberania National Park (SNP), and b) season; Wet and Dry.  E. burchelli  Mean ± SE  colonies per 100 ha F  df  0.88  1,22 0.214  P  Swarm width of colonies (m) Mean ± SE  F  df  7.52  1,16 0.014  0.98  1,16 0.337  P  a) SITE BCI  2.08 ±0.18  SNP  2.62 ± 0.35  f BCI (1961)  3.17  9.60 ±0.51 12.21 ±0.85 6.00  b) SEASON WET DRY  1.71 ±0.13 3.08 ± 0.29  c) SITE X SEASON  10.8  1,22 <0.001  2.45 1,22 0.131  10.56 ± 1.27 11.13 ±0.48  0.50 1, 16 0.490  tWillis 1967  36  Table 2.2: Mean dry biomass (mg) of flushed arthropods per quadrat between a) sites (Barro Colorado: BCI, and Soberania: SNP) and b) seasons (wet and dry). Two-way ANOVA results on the ranked data are presented.  a) SITE Tettigoniidae  BCI  SNP  n = 134  n = 170  Two-way ANOVA  52.03 ± 12.61 41.01 ± 10.63  Blattodea Araneida > 0.5 cm  Araneida < 0.5 cm Gryllidae Coleoptera Lepidoptera Isopodea TOTAL  0.09  0.771  14.43 ± 2.32  18.88 ±2.86  2.92  0.088  4.71 ±0.42  3.66 ± 0.28  7.30  0.007  29.52± 3.47  35.32 ±3.85  0.19  0.667  . 2.27 ± 0.42  1.56 ±0.32  2.47  0.117  0.09 ± 0.06  0.43 ±0.12  7.30  0.007  2.61 ±0.50  2.48 ± 0.47  0.00  0.984  n= 100  Tettigoniidae Blattodea Araneida > 0.5 cm  113.54 ± 12.2 DRY n = 204  Coleoptera Lepidoptera Isopodea TOTAL  P  F  30.15 ±7.50  5.33  0.022  11.33 ± 1.37  5.55 ±1.11  29.33  <0.001  30.86 ± 4.62  10.08 ±1.48  21.31  <0.001  6.83 ± 0.48  2.79 ± 0.23  91.60  <0.001  12.81 ± 1.91  42.55 ± 3.63  25.68  <0.001  1.80 ±0.42  1.91 ±0.33  0.00  0.965  0.28 ± 0.09  0.28 ±0.12  0.03  0.857  0.36 ±0.18  3.59 ± 0.48  27.69  <0.001  145.10 ±20.48  100.78 ± 8.86  b  Gryllidae  Two-way ANOVA  77.92 ± 19.06 a  Araneida < 0.5 cm  0.641  8.19 ± 1.35  117.67± 13.52 WET  b) SEASON  0.22  6.51 ± 1.03 a  b  P  F  Site X Season, p<0.05 Site X Season, p<0.001  a  b  37  Table 2.3: Population density (mean number and grams per 100 ha) for ocellated, bicolored, and spotted antbirds on Barro Colorado (BCI) and in Soberania (SNP) from a) other studies and b) this study. C.I.'s were generated from Bowden's Model Estimator. Date  Ocellated antbird  a)  BCI  SNP  Species  Surveyor  Bicolored antbird  Ind./lOO ha  g/100 ha Ind./lOO ha  g/100 ha Ind./lOO ha  Willis (1974)  3.8  193.8  7.6  228.0  46.4  788.8  1970  Willis (1974)  0.3  15.3  3.2  96.0  45.2  768.4  1994  Robinson (1999)  0.0  0.0  4.0  120.0  48.0  816.0  2003  D. Robinson (pers.comm.)  66.7  1133.9  1968  Karr(1971)  1994  Robinson et al. (2000)  50.0  850.0  56.0  952.0  408.0  8.0  Ocellated antbird  b)  SNP  g/100 ha  1961  20.0  600.0  24.0  720.0  1998 - 2000 Strysky (2003)  BCI  Spotted antbird  Bicolored antbird  Ind./lOO ha  g/100 ha Ind./lOO ha  g/100 ha  2003  This study  0.0  0.0  12.6(11.1-14.3)  378.0  2004  This study  0.0  0.0  13.0(11.1-15.1)  390.0  2003  This study  6.8 (5.9 -7.4) 346.8  21.9 (20.1-23.8)  657.0  2004  This study  7.0 (4.8 - 8.7) 357.0  24.4 (22.0 - 27.1) 732.0  38  Mean number and maximum number of ocellated, bicolored, and spotted antbirds per ant swarm on Barro Colorado (BCI) and in Soberania (SNP). Historical data for BCI from 1960-1964 are from Willis (1967, 1972, 1973). Table  2.4:  Site BCI  Date  Mean ± SE Max  N  Mean ± SE Max  12 448 4.1  1960-64  2.9  this study  0  0  1.4 ±0.12  5 108 6.8 ± 0.29  SNP this study  Spotted  Bicolored  Ocellated  85 5.3 ± 0.24  N  Mean ± SE Max N  15 1116 1.8 10  8 428  85 5.5 ±0.32  14 85  16 108 2.3 ±0.16  6 108  39  Table 2.5: Mean numbers of all birds counted per ant swarm and mean total biomass of all birds per ant swarm for a) sites (Barro Colorado: BCI, and Soberania: SNP) and b) seasons (WET and DRY). Individuals per swarm Mean±SE  F  df  Total biomass per swarm (g) P  Mean ± SE  F  df  a) SITE BCI  14.87 ±0.60  SNP  14.31 ±0.51  0.03 1, 189  0.870  416.53 ±18.80 541.87±24.07  13.03 1, 189 <0.001  b) SEASON WET  14.04 ±0.73  DRY  14.74 ± 0.46  c) SITE X SEASON  0.59 1, 189  0.440  3.61 1,189 0.060  455.02 ±30.35 497.44 ±19.38  0.93 1, 189 0.760  1.37 1,189 0.244  40  Table 2.6: Estimates of the selection ratios (relative occurrence of arthropods in the diet/relative abundance of flushed arthropods available) of bicolored antbirds a) between sites (BCI and SNP) and b) between seasons (WET and DRY). A standardized selection index > 0.14 indicates preference and is depicted in bold. Results from % test of differences between selection ratios between a) sites and b) seasons are reported, x values are Bonferroni adjusted. Statistically significant different differences are in bold (P < 0.05). 2  2  Proportion a) SITE  in Diet BCI  Proportion in Sample  SNP  BCI  SNP  Selection Ratio BCI  Standardized  2  X  Selection Index  SNP  BCI  SNP  8.40  1.13  2.46  8.65  3.41  0.46  0.31  12.28  24.97 23.04  6.65  10.85  3.76  2.12  0.20  0.20  17.55  Araneida >0.5 cm 21.81 23.70  6.65  11.29  3.28  2.10  0.18  0.19  10.01  18.82  48.80 35.31  0.31  0.53  0.02  0.05  12.32  18.29 16.77  19.66 30.54  0.93  0.55  0.05  0.05  10.30  Tettigoniidae Blattodea  9.79  Araneida <0.5 cm 15.11 Gryllidae Coleoptera  6.47  7.17  4.53  4.05  1.43  1.77  0.08  0.16  0.69  Other  3.56  2.10  12.59  5.50  0.28  0.38  0.02  0.04  0.49  N b) SEASON  350  488  748  751  Proportion  Proportion  Selection  Standardized  in Diet  in Sample  Ratio  Selection Index  WET  DRY  WET  DRY  WET  DRY  WET  DRY  10.31  1.93  1.70  3.11  6.06  0.35  0.37  7.94  12.92 27.31  11.80  6.92  1.09  3.94  0.12  0.24  78.12  Araneida >0.5 cm 31.85  19.34  14.51  5.68  2.20  3.41  0.25  0.21  11.81  Araneida <0.5 cm 32.68  12.23  51.84 36.44  0.63  0.34  0.07  0.02  20.53  Gryllidae  9.92 20.45  8.70 34.62  1.14  0.59  0.13  0.04  7.19  Coleoptera  0.00  8.56  3.68  4.65  0.00  1.84  0.00  0.11  50.36  Other  4.81  1.80  7.54  9.99  0.64  0.18  0.07  0.01  6.92  Tettigoniidae Blattodea  N  6.01  300  538  526  973  41  Table 2.7: Mean rates of aggression ± SE (per minute) directed TO species groups by bicolored antbirds or received by bicolored antbirds FROM species groups. Mean rates reported are between a) sites (BCI and SNP) and between b) seasons (WET and DRY). Results from two-way ANCOVA are presented with site and season as factors and observation length as a covariate. N = number of observations. BCI  SNP  N = 69  N = 88  Bicolored  0.05 ± 0.01  0.09 ± 0.02  Spotted  0.06 ± 0.01  Other  a) SITE  Two-way ANCOVA df  P  1.15  1, 152  0.284  0.02 ± 0.01  9.17  1, 152  0.003  0.02 ± 0.01  0.01 ±0.01  0.85  1, 152  0.357  Bicolored  0.05 ± 0.01  0.09 ± 0.02  1.15  1, 152  0.284  Spotted  0.01 ±0.01  0.00  0.72  1, 152  0.398  0.00  0.02 ± 0.01  9.14  1, 152  0.003  0.01 ±0.01  0.02 ± 0.01  2.847  1, 152  0.094  F  To:  3  From:  Ocellated Other  Wet  Dry  N = 48  N = 109  Bicolored  0.09 ± 0.02  0.07 ± 0.01  Spotted  0.01 ± .01  Other  b) SEASON  Two-way ANCOVA df  P  2.16  1, 152  0.143  0.05 ± 0.01  9.95  1, 152  0.002  0.03 ± 0.02  0.01 ± 0.004  10.94  1, 152  0.001  0.09 ± 0.02  0.07 ± 0.01  2.16  1, 152  0.143  0.00  0.01 ± 0.004  2.97  1, 152  0.087  Ocellated  0.01 ± 0.003  0.01 ±0.01  0.92  1, 152  0.339  Other  0.02 ± 0.01  0.01 ±0.01  6.65  1, 152  0.011  F  To:  3  From: Bicolored Spotted  Site X Season; p < 0.05  a  42  Figure 2.1: Spatial partitioning at an ant swarm. The greatest abundance of arthropods occurs in zone A, where the swarm front drives arthropods from the leaf litter, here, occupied by the dominant ocellated antbird. Zone B, still an area of high arthropod activity, is occupied by the subordinate bicolored antbird. Surrounding zone B is zone C, occupied by the small subordinate spotted antbird, considered a facultative ant follower. Vertical partitioning of resources at ant swarms also occurs. In zone D, a plain-brown woodcreeper forages on invertebrates flushed upwards by the ants. E represents the bivouac, the army ant nest location, while the arrow depicts the direction of forward movement by the ant swarm. Illustration by James N. M. Smith 43  Figure 2.2: Study area in the Republic of Panama. Focal sites were Barro Colorado Island, and surrounding the Limbo plot in Soberania National Park.  •  44  Figure 2.3: Study areas in a) Barro Colorado, and b) Soberania. Light gray areas depict study area, while the 100 ha census plots are designated by dark gray.  45  a)  Spotted  Bicolored  Ocellated  Others  Species Group - BCI  b)  10.0-  •  DRY  •  WET  7.5-  a  in at  5.0-  11  •o  •5  2.5-  c  o.o-  —\—•—I—•—I—•—r Spotted  Bicolored  Ocellated  -  Others  Species Group - SNP  Figure 2.4: Mean numbers of individual spotted, bicolored, and ocellated antbirds, and all other species attending ant swarms during the dry (black) and wet (gray) season, a) Barro Colorado (BCI), b) Soberania (SNP). N = 189 swarms. Error bars represent 95% CPs of the mean. Two-way ANOVA results are given in Appendix 2.2; * Site: p<0.05; t Season: p<0.05; ^Interaction (Site X Season): p<0.05.  46  Figure 2.5: Percent of total biomass at ant swarms of spotted, bicolored, and ocellated antbirds, and all others species for both Barro Colorado (BCI; black hatched) and Soberania (SNP; gray hatched) over both seasons.  47  a)  BCI  •  Dry  •  Wet  SNP  Site  b)  BCI  SNP  Site  Figure 2.6: Foraging rates per minute by bicolored antbirds during both season on Barro Colorado (BCI) and in Soberania (SNP) during the dry (black) and wet (gray) seasons; a) small prey (0.5-1.0 x bill length), b) large prey (>1.0 x bill length). N = 152 observations. Two-way ANOVA results given in Appendix 2.3; * Site: p<0.05; t Season: p<0.05; :j:Interaction (Site X Season): p<0.05. 48  BCI 2003-04  A  SPAB  BCAB  OTHER  B  SNP 2003-04 BCAB 8%  36%fr  SPAB  36%  BCAB  OCAB  OTHER  OCAB  OTHER  BCI 1960-61 BCAB 7°/c  36%fr  SPAB  3  BCAB  6  %  Figure 2.7: Percentage of aggressive interactions (supplanting or displacing) made by or received by bicolored antbirds on a) Barro Colorado (BCI) 2003-2004, b) Soberania (SNP) 2003-2004, and c) BCI 1960-1961 (Willis 1967). The direction and width of the arrows indicate the direction of and strength of aggression, respectively. Bicolored (BCAB); Spotted (SPAB); Ocellated (OCAB); Other (OTHER).  49  a) •  Dry  b)  BCI  SNP  Site  Figure 2.8: Rate of activity by bicolored antbirds per minute during both seasons on Barro Colorado and in Soberania; a) hops per minute, b) flights of 3 to 10 meters. N 152 observations. Conventions as in Figure 2.6. Two-way ANOVA results given in Appendix 2.4.  LITERATURE CITED Armstrong;, R. A., and R. McGehee. 1976. Coexistence of 2 competitors on one resource. Journal of Theoretical Biology 56:499-502. Chapman, F. M. 1929. My tropical air castle. Appleton-Century, New York. Chappell, M. A. 1978. Behavioral factors in altitudinal zonation of chipmunks (Eutamias). Ecology 59:565-579. Chesson, P. 2000b. Mechanisms of maintenance of species diversity. Annual Review of Ecology and Systematics 31:343-366. Clarke, K. R., and R. N. Gorley. 2001. PRIMER v5: User manual/tutorial. PRIMER-E, Plymouth, UK. Clarke, K. R., and R. M. Warwick. 1994. Similarity-based testing for community pattern - the 2-way layout with no replication. Marine Biology 118:167-176. Cody, M. L. 1974. Competition and the structure of bird communities. Monographs in Population Biology 7. Princeton University Press, Princeton, New Jersey. Cody, M. L., and J. M. Diamond. 1975. Ecology and the evolution of communities. Harvard University Press, Cambridge, Massachusetts. Connell, J. H. 1978. Diversity in tropical rainforests and coral reefs. Science 199:13021310. Connell, J. H. 1980. Diversity and the coevolution of competitors, or the ghost of competition past. Oikos 35:131-138. Connell, J. H. And Orias, E. 1964. The ecological regulation of species diversity. The American Naturalist 98:399-414. Connell, J. H. And W. P. Sousa. 1983. On the evidence needed to judge ecological stability or persistence. American Naturalist 121:789-824. Dobzhansky, T. Evolution in the tropics. American Scientist 38: 209-221. Eisenmann, E. 1952. Annotated list of birds of Barro Colorado Island, Panama Canal Zone. Smithsonian Institution, Washington, D.C. Franks, N. R. 1982. Ecology and population regulation in the army ant Eciton burchelli. In E. G. Leigh, Jr., A. S. Rand, and D. M. Windsor, editors. The ecology of a tropical forest: seasonal rhythms and long-term changes. Smithsonian Institution Press, Washington, D.C.  51  Glanz, W. E. 1982. The terrestrial mammal fauna of Barro Colorado Island: censuses and long-term changes. Pages 455-468 in E. G. Leigh, Jr., A. S. Rand, and D. M. Windsor, editors. The ecology of a tropical forest: seasonal rhythms and longterm changes. Smithsonian Institution Press, Washington, D.C. Glanz, W. E. 1990. Neotropical mammal densities: how unusual is the community on Barro Colorado Island, Panama? Pages 287-313 in A. Gentry, editor. Four Neotropical rain forests. Yale University Press, New Haven, Connecticut.  Griffis, M. R., and R. G. Jaeger. 1998. Competition leads to an extinction-prone species of salamander: Interspecific territoriality in a metapopulation. Ecology 79:24942502. Gruner, D. S. 2003. Regressions of length and width to predict arthropod biomass in the Hawaiian Islands. Pacific Science 57:325-336. Gurevitch, J., L. L. MorroW, A. Wallace, and J. S. Walsh. 1992. A metaanalysis of competition in field experiments. American Naturalist 140:539-572. Hairston, N. G. 1981. An experimental test of a guild - salamander competition. Ecology 62:65-72. Holbrook, S. J., and R. J. Schmitt. 1995. Compensation in resource use by foragers released from interspecific competition. Journal of Experimental Marine Biology and Ecology 185:219-233. Holdridge, L. R. 1967. Life zone ecology, San Jose, Costa Rica. Hubbell, S. P. 2001. The unified neutral theory of species abundance and diversity. Princeton University Press, Princeton, New Jersey. Karr, J. R. 1971. Ecological, behavioral, and distributional notes on some central Panama birds. Condor 73:107-111. Leigh, E. G., Jr. 1999. Tropical forest ecology. Oxford University Press, New York, New York. Leigh, E. G., Jr. 1982. Introduction: the significance of population fluctuations. Pages 435-440 in E. G. Leigh, Jr., A. S. Rand, and D. M. Windsor, editors. The ecology of a tropical forest: seasonal rhythms and long-term changes. Smithsonian Institution Press, Washington, D.C. Levings, S. C , and D. M. Windsor. 1982. Seasonal and annual variation in litter arthropod populations. In E. G. Leigh, Jr. And D. M. Windsor, editors. The Ecology of a Tropical Forest: Seasonal Rhythms and Long-term Changes. Smithsonian Inst. Press, Washington, D. C. 52  Lotka, A. J. 1932. The growth of mixed populations: two species competing for a common food supply. Journal of the Washington Academy of Sciences 22:461469. MacArthur, R. H. And R. Levins. 1967. Limiting similarity convergence and divergence of coexisting species. American Naturalist 101:377-385. MacArthur R. H. 1958. Population ecology of some warblers in northeastern coniferous forests. Ecology 39:599-619. MacNally, R. C. 1983. On assessing the significance of interspecific competition to guild structure. Ecology 64:1646-1652. Manly, B. F. J., L. L. McDonald, D. L. Thomas, T. L. McDonald, and W. P. Erickson. 2002. Resource selection by animals: statistical design and analysis for field studies. Kluwer Academic Publishers, Dordrecht, The Netherlands. Martin, P. R. And T. E. Martin. 2001. Ecological and fitness consequences of species coexistence: a removal experiment with wood warblers. Ecology 82:189-206. May, R. M., and R. H. Macarthur. 1972. Niche overlap as a function of environmental variability. Proceedings of the National Academy of Science (USA) 69:11091113. Menge, B. A. 1995. Indirect effects in marine rocky intertidal interaction webs - patterns and importance. Ecological Monographs 65:21-74. Moore, R. P., and W. D. Robinson. 2004. Artificial bird nests, external validity, and bias in ecological field studies. Ecology 85:1562-1567. Munday, P. L., G. P. Jones, and M. J. Caley. 2001. Interspecific competition and coexistence in a guild of coral-dwelling fishes. Ecology 82:2177-2189. Oaten, A., and W. W. Murdoch. 1975. Switching, functional response, and stability in predator-prey systems. American Naturalist 109:299-318. Paine, R. T. 1992. Food-web analysis through field measurement of per-capita interaction strength. Nature 355:73-75. Palmer, T. M., M. L. Stanton, and T. P. Young. 2003. Competition and coexistence: Exploring mechanisms that restrict and maintain diversity within mutualist guilds. American Naturalist 162:S63-S79. Rastetter, E. B., and G. I. Agren. 2002. Changes in individual allometry can lead to species coexistence without niche separation. Ecosystems 5:789-801. Relyea, R. A. 2000. Trait-mediated indirect effects in larval anurans: Reversing competition with the threat of predation. Ecology 81:2278-2289. 53  Relyea, R. A., and K. L. Yurewicz. 2002. Predicting community outcomes from pairwise interactions: integrating density- and trait-mediated effects. Oecologia 131:569579. Rettenmeyer, C. W. 1963. Behavioral studies of army-ants. University of Kansas Science Bulletin 44:281-465. Robertson, D. R. 1996. Interspecific competition controls abundance and habitat use of territorial Caribbean damselfishes. Ecology 77:885-899. Robinson, W. D. 1999. Long-term changes in the avifauna of Barro Colorado Island, Panama, a tropical forest isolate. Conservation Biology 13:85-97. Robinson, W. D. 2001. Changes in abundance of birds in a Neotropical forest fragment over 25 years: a review. Animal Biodiversity and Conservation 24:51-65. Robinson, W. D., J. D. Brawn, and S. K. Robinson. 2000. Forest bird community structure in central Panama: Influence of spatial scale and biogeography. Ecological Monographs 70:209-235. Rodda, G. H., and K. Dean-Bradley. 2002. Excess density compensation of island herpetofaunal assemblages. Journal of Biogeography 29:623-632. Roughgarden, J. 1974. Species packing and the competition function with illustrations from coral reef fish. Theoretical Population Biology 5:163-186. Sandlin, E. A. 2000. Foraging information affects the nature of competitive interactions. Oikos 91:18-28. Schmitt, R. J., and S. J. Holbrook. 1990. Population responses of surfperch released from competition. Ecology 71:1653-1665. Schneirla, T. C. 1933. Studies on army ants in Panama. Journal of Comparative Psychology 15:267-299. Schneirla, T. C. 1949. Army-ant life and behavior under dry-season conditions. Bulletin of the American Museum of Natural History 94:7-81. Schneirla, T. C. 1971. Army ants: a study in social organization. W. H. Freeman, San Francisco, California. Schoener, T. W. 1983. Field experiments on interspecific competition. American Naturalist 122:240-285. Sieving, K. E. 1992. Nest predation and differential insular extinction among selected forest birds of Central Panama. Ecology 73:2310-2328. SPSS. 2003. SPSS statistical software. Eleventh edition. 54  Styrsky, J. N. 2003. Life-history evolution and population dynamics of a Neotropical forest bird (Hylophylax naevioides). Ph.D. Dissertation. University of Illinois, Urbana, IL. Swartz, M. B. 1997. Behavioral and population ecology of the army and Eciton burchelli and ant-following birds. Ph.D. Dissertation. University of Texas, Austin. Terborgh, J. 1983. Five new world primates: a study in comparative ecology. Princeton University Press, Princeton, New Jersey. Tilman, D. 1982. Resource competition and community structure. Princeton University Press, Princeton, New Jersey. Walsberg, G. E. 1983. Avian ecological energetics. Pages 161-220 in D. S. Farner, J. R. King, and K. C. Parkes, editors. Avian Biology. Academic Press, New York. Werner, E. E., and S. D. Peacor. 2003. A review of trait-mediated indirect interactions in ecological communities. Ecology 84:1083-1100. White, G. C. 1996. NOREMARK: population estimation from mark-resighting surveys. Wildlife Society Bulletin 24:50-52. Wiens, J. A. 1992a. The ecology of bird communities volume 1: Foundations and patterns. Harvard University Press, Cambridge. Wiens, J. A. 1992b. The ecology of bird communities volume 2: Processes and variations. Harvard University Press, Cambridge. Williams, J. B., and G. O. Batzli. 1979. Competition among bark-foraging birds in Central Illinois - Experimental-Evidence. Condor 81:122-132. Willis, E. O. 1967. The behavior of bicolored antbirds. University of California Publications in Zoology 79. Willis, E. O. 1972. The behavior of spotted antbirds. Ornithological monographs 10. The American Ornithologists' Union, Washington, D.C. Willis, E. O. 1973. The behavior of ocellated antbirds. Smithsonian Contributions to Zoology 144. Willis, E. O. 1974. Populations and local extinctions of birds on Barro Colorado Island, Panama. Ecological Monographs 44:153-169. Willis, E. O. 1983. Longevities of some Panamanian forest birds, with note of low survivorship in old spotted antbirds (Hylophylax-Naevioides). Journal of Field Ornithology 54:413-414.  55  Willis, E. O., and E. Eisenmann. 1979. A revised list of birds of Barro Colorado Island, Panama. Contributions to Zoology 291. Willis, E. O., and Y. Oniki. 1978. Birds and army ants. Annual Review of Ecology and Systematics 9:243-263. Willson, S. K. 2004. Obligate army-ant-following birds: a study of ecology, spatial movement patterns, and behavior in Amazonian Peru. Ornithological Monographs 53. The American Ornithologists' Union, Washington, D.C. Wright, S. J., C. Carrasco, O. Calderon, and S. Paton. 1999. The El Nino Southern Oscillation, variable fruit productions, and famine in a tropical forest. Ecology 80:1632-1647. Yodzis, P. 1988. The indeterminacy of ecological interactions as perceived through perturbation experiments. Ecology 69:508-515.  56  Appendix 2.1: Frequency of species observed at or near ant swarms on Barro Colorado and in Soberania during both wet and dry seasons. Occurrence at E. burchelli swarms per season § Mass(g) t Barro Colorado Island Soberania National Park Wet Dry Wet Dj2_  Species t Tinamidae Tinamus major  VR  1160  VR  Accipitridae Leucopternis  semiplumbea  Leucopternis albicollis Harpagus bidentatus fasciatus Accipiter super ciliosus fontanieri  278 736 185 100  VR VR VR  179 650  VR  VR  VR  Falconidae Micrastur ruficollis interstes Micrastur semitorquatus naso  VR  VR  U  U  Cuculidae Neomorphus geoffroyi salvini  340  Momotidae Momotus momota Baryphthengus martii Electron platyrhynchum  105 162 62  R R  R R  VR R R  VR R R  44  R  R  R  R  VR  VR  Bucconidae Malacoptila  pandmensis  Furnariidae Philydor  fuscipennis  Xenops minutus Sclerurus guatemalensis  11 34  VR VR  VR  VR  VR  41 41 24 15 68 47 51  C  C  C  C VR VR VR  Dendrocolaptidae Dendrocincla fuliginosa Dendrocincla homochroa Deconychura longicauda Clyphorhynchus spirurus Dendrocolaptes certhia Xiphyorhynchus guttatus Xiphorhynchus lachrymosus  VR C R  VR C R  U  C R  U c  R 57  Appendix 2.1: confined  Species t  Occurrence at E. burchelli swarms per season § Mass(g) % Barro Colorado Island Soberania National Park Wet Dry Wet , Dry  Thamnophilidae Cymbilaimus lineatus Thamnophilus punctatus Myrmotherula fulviventris Myrmotherula axillaris Microrhopias quixensis Myrmeciza exsul Hylophylax naevioides Gymnopithys leucaspis Phaenostictus mcleannani Formicarius analis Hylopezus perspicillatus  37 22 10 8 8 27 17 30 51 57 42  Pipridae Manacus vitellinus Pipra mentalis  17 • R 15  Troglodytidae • Henicorhina leucosticta Cyphorhinus phaeocephalus  17 . 25  Sylviinae Microbates cinereiventris Polioptila plumbea  12 7  R VR  R VR  Thraupinae Eucometis penicillata  30  C  C  C  u u u c A A  C U U  u c A A  R  VR C R R R VR C A C VR VR  VR C R R R  R R  R R  R U  R U  u  u  C A C VR VR  f Species names and sequence follow Ridgely and Gwynne (1989) X Mass (g) follow Robinson et al. (2000) § VR: very rare, only a few sightings; R: rare, sightings < 25% of swarms; U : uncommon, sightings 20-50% of swarms; C: common, > 75% sightings in small numbers; A : abundant, always present in large numbers  58  Appendix 2.2: Results of two-way M A N O V A for overall effects of site and season on the individuals per swarm at ant swarms of four main groups: spotted, bicolored, ocellated, and all other antbirds and between subject effects from two-way A N O V A . df MANOVA Site  56.42  4, 186  <0.001  Season  10.94  4, 186  <0.001  6.94  4, 186  <0.001  Site  66.10  1, 189  <0.001  Season  19.53  1, 189  <0.001  5.60  1, 189  0.019  16.56  1, 189  <0.001  Season  7.00  1, 189  0.009  Site X Season  3.20  1, 189  0.075  134.44  1, 189  <0.001  Season  7.01  1, 189  0.009  Site X Season  7.01  1,189  0.009  Site  3.95  1, 189  0.048  Season  2.32  1, 189  0.130  Site X Season  3.79  1, 189  0.053  Site X Season  ANOVA Spotted antbirds  Site X Season  Bicolored antbirds Site  Ocellated antbirds Site  Other  Appendix 2.3: Results of two-way MANCOVA for foraging rates of bicolored antbirds at ant swarms and two-way ANCOVA results of the response variables, small and large prey. Factors: site and season; Covariate: observation length. F  df  P  Site  2.18  2, 15.1  0.117  Season  1.38  2, 151  0.256  Site X Season  1.32  2, 151  0.271  Site  3.96  1, 152  0.048  Season  2.24  1, 152  0.136  Site X Season  0.10  1, 152  0.753  Site  0.05  1, 152  0.832  Season  1.05  1, 152  0.306  Site X Season  2.24  1, 152  0.137  MANCOVA  ANCOVA Small Prey  Large Prey  60  Appendix 2.4: Results of two-way MANCOVA for main activity rates of bicolored antbird, and the effects of the response variables hops and flights, at ant swarms by twoway ANCOVA. Factors: site and season; Covariate: observation length.  F  df  P  MANCOVA Site  4.58  2, 151  0.012  Season  4.59  2, 151  0.012  Site X Season  0.32  2, 151  0.725  Site  3.01  1, 152  0.085  Season  6.36  1, 152  0.013  Site X Season  0.46  1, 152  0.499  Site  8.66  1, 152  0.004  Season  6.13  1, 152  0.014  Site X Season  0.42  1, 152  0.520  ANCOVA Hops  Flights  61  CHAPTER 3: GENERAL CONCLUSIONS  Summary of thesis My primary objective in this study was to test if interspecific competition maintains the structure of a tropical avian guild, either through direct or indirect interactions. I explored both numerical and behavioral compensatory responses, several generations following the loss of a dominant competitor from a specialized guild of antfollowing birds. I document three new results suggesting that competition influences the structure of this avian guild. First, following the sustained absence of a dominant competitor, the ocellated antbird, the remaining guild members completely compensated in biomass density. Although data on population abundances of this guild were only collected intermittently before my study, evidence of density compensation only became apparent following complete population turnover. Second, compensatory use of microhabitat (ant swarms) was also observed. Unexpectedly, the subordinate spotted antbird exhibited the largest compensatory response. Spotted antbirds doubled in number at ant swarms on Barro Colorado Island, compensating for the relative proportion of missing biomass from ocellated antbirds. Bicolored antbirds, which are dominant to spotted antbirds, did not differ in rates of foraging, aggression with conspecifics, and activity levels at ant swarms. Rates of aggression between bicolored and spotted antbirds, however, increased. Ocellated antbirds have rarely been observed to directly supplant spotted antbirds. Thus, limited swarm use by spotted antbirds likely resulted from avoidance of ocellated antbirds by spotted antbirds, and indirect competitive pressure by ocellated antbirds,  62  mediated through their direct interactions with bicolored antbirds. Third, life-history characteristics such as limited mobility during the breeding season, in addition to seasonally varying levels of competition for resources, influenced the interaction strength between guild members and the use of resources.  Avenues for future research Spatial dynamics  Often, population responses are coupled to shifts in territory sizes in birds (Newton 1998) but less is known about how shifts in abundance affect home range sizes. Indeed, my conclusion that spotted antbirds increased in abundance on Barro Colorado since the 1990s was based on territory size (D. Robinson, personal  communication).  However, I observed pairs of spotted antbirds following army ant swarms for several consecutive days, far outside of what would be considered territorial boundaries by Willis' (1972) criteria. For birds or species that hold loose territories, such as obligate ant-followers, little is known about how density or behavioral compensation may influence space use. Home range sizes of obligate ant-followers in Cocha Cashu, Peru, also increased during an early rainy season. In that case, flooding of the forest floor reduced the area that was suitable for foraging (Willson 2004). An expansion in home range by spotted antbirds on Barro Colorado as a result of behavioral compensation leads to another interesting question. Increased foraging at ant swarms by spotted antbirds on Barro Colorado, and its corollary increased distances traveled to follow ant swarms, challenges the notion that these birds are only facultative ant-followers (i.e., they forage at ant swarms only when passing through their territory).  63  If spotted antbirds on Barro Colorado begin to more obligately follow ant swarms, will their territory structure break down as the need to hold loose territories increases in order to decrease agonistic aggression at ant swarms with conspecifics? A shift in territoriality could affect population abundances. Greenberg and Gradwohl (1986) hypothesized that year-round territoriality is one proximate cause of population stability in tropical birds. Support for this hypothesis was provided by Willson (2004), who documented population declines in three non-territorial ant-following birds, while population of strictly territorial species remained stable.  Role of competition in species guilds  The idea of competition as a driving force behind community structure currently receives little focus because of the past criticism that many studies testing the idea lacked alternative hypotheses (Wiens 1992). More recent experimentally rigorous studies nevertheless have supported the notion of the importance of competition to guild structure through observed density compensation following species removals: particularly in damselfish (Robertson 1996), surfperches (Schmitt and Holbrook 1990), herpetofaunal assemblages (Rodda and Dean-Bradley 2002). These studies are further supported by this study on a tropical avian guild of ant-followers. In a study of gobies  (Gobiodon),  however, a variety of mechanisms in addition to competition were proposed to affect guild structure (Munday et al 2001). Competition is also thought to impact local community structure in less isolated systems. Species compensation, rather than density compensation, was observed in a species rich guild of desert rodents following local extinction of guild members (Goheen et al. 2005).  64  We thus have evidence of competition in a variety of taxa and environmental settings. Despite this experimental evidence for the importance of competition in these particular cases, we lack predictive power for when and what guilds might experience competition as a structuring factor, and to what degree. For example, are other tropical guilds influenced by competitive interactions similarly to the ant-followers and can we predict that the same pattern may not be revealed in temperate communities? Does the degree of specialization found within communities influence the nature of competitive forces? How does climatic stability and fluctuations in predators or parasites influence interactions among species? These questions will require a systematic body of work, including long-term observations and experimental approaches in a variety of environments and species guilds. We may then begin to predict the compensatory responses and lasting impacts in communities following periodic disturbances or species loss.  65  LITERATURE CITED Goheen, J. R., E. R White, S. K. M. Ernest, and J. H. Brown. 2005. Intra-guild compensation regulates species richness in desert rodents. Ecology 86:3 567-573. Greenberg, R. and J. Gradwohl. 1986. Constant density and stable territoriality in some tropical insectivorous birds. Oecologia 69:618-625. Munday, P. L., G. P. Jones, and M. J. Caley. 2001. Interspecific competition and coexistence in a guild of coral-dwelling fishes. Ecology 82:2177-2189. Newton, I. 1998. Population limitation in birds. Academic Press. New York, New York. Robertson, D. R. 1996. Interspecific competition controls abundance and habitat use of territorial Caribbean damselfishes. Ecology 77:885-899. Rodda, G. H., and K. Dean-Bradley. 2002. Excess density compensation of island herpetofaunal assemblages. Journal of Biogeography 29:623-632. Schmitt, R. J., and S. J. Holbrook. 1990. Population responses of surfperch released from competition. Ecology 71:1653-1665. Wiens, J. A. 1992. The ecology of bird communities volume 2: Processes and variations. Harvard University Press, Cambridge. Willis, E. O. 1972. The behavior of spotted antbirds. Ornithological monographs 10. The American Ornithologists' Union, Washington, D.C. Willson, S. K. 2004. Obligate army-ant-following birds: a study of ecology, spatial movement patterns, and behavior in Amazonian Peru. Ornithological Monographs 53. The American Ornithologists' Union, Washington, D.C.  66  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items