|USFLDC Home | Tropical Ecology Collection [Monteverde Institute]||| RSS|
This item is only available as the following downloads:
Interaction of color and sugar concentration preference with aggression in Monteverde Cloud Forest hummingbirds (Trochilidae) Laurel Anne Mundy Department of Biology, Western Washington Universit y ABSTRACT In Monteverde, Costa Rica, a number of hummingbird species coexist that are extremely important for pollination. Hummingbirds may prefer flowers of a c ertain color because they associate it with a highe r nectar reward. They also may be territorial and def end especially rich food sources, dominant birds dr iving away all subordinates. The purpose of this study wa s to investigate color and sugar concentration preference in hummingbirds using artificial feeders and to determine whether dominance and aggression has an effect on food choice. Many hummingbird-poll inated flowers are red, so I hypothesized that all species would show a preference, and predicted that all would initially choose the red feeder of highe st sugar concentration, eventually also choosing the t wo other high concentration feeders, realizing thei r equal reward. I also predicted that larger, male hummingb irds would dominate the food source more territoria l, dominating the food source and forcing smaller bird s to choose their second choice feeder. This was do ne by hanging six feeders of red, yellow and purple an d high and low sugar concentration and recording species preference. I found that the Violet Sabrewi ng was the species that most often chose the red, h igh concentration feeder, as predicted (p=0.041669, df= 2, x2= 6.356). No conclusions could be made about the three other species investigated as results were no t significant. Preference was seen in the Violet Sabrewing, as differences in feeder preference decr eased when only the last four days of data were analyzed, suggesting that color preference decrease d. Most aggression was observed from the Violet Sabrewing (p=0.006019, df=9, x2=23.08). This species was clearly dominant likely d ue to its large size. RESUMEN En Monteverde, Costa Rica, un nmero de especies de colibres coexiste y son extremadamente importantes para la polinizacin. Los colibres pu eden preferir flores de ciertos colores debido a qu e lo asocian con un alto contenido de nctar. Ellos tam bin pueden ser territoriales y defender principalm ente recursos alimenticios, aves dominantes ponen lejos a las subordinadas. El propsito de este estudio f ue investigar preferencia por color y concentracin de azcar por colibres utilizando comederos artifici ales, y para determinar si la dominancia y agresin tienen un efecto en la seleccin de alimento. Muchas flor es polinizadas por colibres son rojas, por lo tanto m i hiptesis es que todas las especies presentaran u na preferencia, y que todas presentaran una seleccin inicial por los comederos rojos con altas concentra ciones de azcar, eventualmente seleccionando tambin los otros comederos con altas concentraciones, realizan do que todos tienen la misma recompensa. Tambin pron stico que machos ms grandes dominaran territorialmente el recurso alimenticio, forzando a las especies ms pequeas a seleccionar los comede ros de segunda opcin. El estudio se realiz colgando seis comederos de color rojo, amarillo y morado; y colocando en estos concentraciones altas y bajas de azcar. Encontr que el ala de sable violceo fue la especie que ms frecuentemente seleccion el rojo c on alta concentracin de azcar, como se predij (p = 0.041, gl= 2, x2= 6.356). No se pueden sacar conclusiones acerca d e las otras tres especies estudiadas debido a que los resultados no son significativos. La preferencia observada en el ala de sable violce o, como la diferencia en la preferencia por comederos decrece cuando se analizan nicamente los ltimos cuatro das del estudio, sugiriendo que la preferen cia por color decrece. La mayora de la agresin f ue observada por el ala de sable violceo (p= 0.006, g l= 9, x2=23.08). Est especie fue claramente dominante seguramente debido a su gran tamao.
INTRODUCTION Hummingbirds are important rainforest pollinators, often observed in pollination mutualisms between specific plants and hummingbirds as seen in Monteverde, Costa Rica. Most hummingbird-pollinated flowers are a sha de of red (which makes them infrequently visited by insects, that cannot distin guish red hues from their background), but may be other shades of bright colors to contras t with green foliage, including pink or purple (Fogden & Fogden 2005). Hummingbirds are mo st sensitive to colors with the longest wavelengths which are the red hues (Long 19 97). The nectar concentration of hummingbird-pollinated flowers is generally 20% suc rose, which balances caloric intake and viscosity (Baker 1975), while conserving energy (Blem & Blem 2000). Studies support the theory that hummingbirds may pr efer certain colors of flowers and sugar concentrations in nectar. Striped-tailed Hummingbirds were shown to exhibit a preference to red artificial feeders (Pogue 1992). Various hummingbirds were also shown to prefer red feeders, at least initially, which co uld suggest an innate preference for red (McGee 1998). When learning and flower color prefer ence were studied in conjunction, however, preference for red flowers was observed in itially, but if the nectar reward was higher in other flower colors, visitation to those colors increased over time (MelendezAckerman et al 1997). Thus it seems that in nature, red flowers generally may contain a higher reward, which is why hummingbirds may appear to prefer red feeders. Some species of red hummingbird-pollinated flowers are p roven to have higher nectar concentrations than their yellow or purple counterp arts (Fenster et al. 1006). Color and concentration may differ, however, for flowers visi ted by different species, as certain birds have bills shaped appropriately for only spec ific flower types (Busby 2000). Therefore, niche partitioning could be significant when investigating potential color or sugar preferences. Melendez-Ackerman et al (1997) concluded that once trained, hummingbirds would prefer the most rewarding choice, regardless of color, suggesting they are capable of learning. Learning and memory are important to h ummingbirds, as they may help for avoidance of previously visited flowers (which woul d no longer contain a nectar reward). It was also shown that hummingbirds could recall th e best food source locations despite identical visual cues, with the use of spatial memo ry (Gonzalez-Gomez and Vasquez 2006). Additionally, birds in past studies have lea rned to associate certain colors with a rich or poor reward, and without these color signal s they were observed to forage randomly (Sandlin 2000). Hummingbirds are often noted for their aggression a nd eagerness to quarrel, energetically defending feeding territories, and ch asing away others regardless of sex or species. Hermits and females, however, are usually considered less pugnacious (Stiles & Skutch 1989). Aggressive behavior of the most terri torial, dominant birds excludes subordinate birds from the food source, and the les s territorial birds are unlikely to pursue defended flowers if undefended food sources are ava ilable (Tiebout 1993), which suggests that aggression may ultimately determine f lower preference. Defense of territories also depends on whether it is energetic ally worthwhile (Beletsky 2007). Dominance is often simply determined by bird size, so large birds may exploit the resource and chase away all smaller birds.
The purpose of this study is to gain insight into h ummingbird aggression and its affect on food choice. From reviewing previous resu lts, I hypothesize that species of Monteverde hummingbirds will exhibit color preferen ce, initially. I predict that all species will pursue red feeders with the highest su crose concentration first, and will eventually visit feeders of other colors with the s ame sucrose concentration. I also expect that aggression will be exhibited more by the large r hummingbird species. METHODS Study Organism Hummingbirds (Trochilidae) are plentiful in Costa R ica and serve as pollinators for many plants, and as there are a large variety of plants, there are a large variety of hummingbird species. During data collection seven different spe cies of hummingbird were seen. These included the Violet Sabrewing, the Purple-throated Mountain-gem, the Stripe-tailed Hummingbird, the Green Violet-ear, the Magenta-thro ated Woodstar, the Ruby-throated Hummingbird, and the Green-crowned Brilliant. Some species were far more common than others. As pollinators, hummingbirds are very important members of the community in a tropical forest, especially in cooler highland areas similar to Monteverde Cloud Forest Reserve where insect and bat activity are li mited for pollination (Stiles and Skutch 1989). Hummingbirds have the highest metabolism of all birds and for their small body size need a high amount of energy to maintain an ap propriate body temperature; thus, they may eat up to their own body weight in food ea ch day (Long 1997). Although nectar is a hummingbirds main food source, providing them with necessary carbohydrates, hummingbirds are not restricted to feeding on necta r and supplement their diets with small insects for protein. Study Site & Preparation To test my hypothesis, several species of hummingbi rds were observed feeding at artificial feeders in the garden of the Estacin Bi ologca de Monteverde, on the edge of the Monteverde Cloud Forest. A spot was chosen to h ang artificial feeders on the border of the garden and the forest, so several species po tentially belonging to both habitats could be observed. A rope was tied between two tree s for feeders to be hung. Six feeders were used, of three colors and two different sugar concentrations. The base of each feeder was painted, along with the flowers surrounding the feeding holes. Two feeders were painted red, two purple and two yellow. Besides col or feeders were identical. Ten percent and 20% sugar water solution were used for the nect ar, and one feeder of each color were filled with the 10% solution and one feeder of each color with the 20% solution. Ten percent sugar concentration was considered low co ncentration and 20% was considered high concentration. Feeders were filled as needed Data Collection Data collection lasted for 11 days, for at least th ree hours every morning between 7:30 10:30 a.m. Feeders were hung around 7:00 a.m. every morning and removed late in the afternoon. Feeder location was rotated each day, bu t same concentration feeders always sat next to each other. Feeders were observed from approximately four meters. When a hummingbird was spotted, The Birds of Costa Rica by Richard Garrigues and Robert
Dean (2007) and binoculars were used to identify th e species (and sex if possible). Feeder choice was recored each time the hummingbird return ed for a drink. Aggressive behavior that caused a hummingbird to be chased away and be deprived of food was noted. This behavior was deemed chasing. Territoriality was d efined as one species causing another to not be able to feed. Any attempt at chas ing without successfully forcing another individual away from the feeders was not co unted in my tally of aggressiveness. A total of seven species were observed, but only fo ur had enough visits to use in my results. After tallying totals for feeder preferenc e, a Chi Squared Contingency Analysis was used for each of the four species to determine if results were significant. The same test was also used once for aggression between all species. RESULTS I observed 652 frequencies of feedings from four di fferent hummingbird species. Over 11 days, the results were different for the Violet Sab rewing (p=0.041669, df=2, x2= 6.356), which preferred the high concentration red feeder ( Fig. 1, Appendix), choosing it 35.7% of the time. All three high concentration feeders w ere chosen 77.1% of the time. Overall, red was chosen 42.2% of the time. Frequency of visi ts for the three low concentration feeders were all nearly equal, and only totaled to 22.9% of all visits. Results were not different for the three other species of hummingbir ds observed. Trends suggest that the Purple-throated Mountain-gem also preferred the hig h concentration red feeder (Fig. 2), as it was chosen 33.1% of the time, both red feeder s being chosen 55.8%, but results were not significant (p=0.168649, df=2, x2=3.560). Trends in the Stripe-tailed Hummingbird suggest that they also preferred the high concentra tion red feeder (Fig. 3), as they chose it 43.6% of the time, but these also were not signific ant (p=0.642385, df=2, x2=0.8851). Finally, trends in data on the Green Violet-ear sug gest that it too preferred the high concentration red feeder (Fig. 4), as it chose it 4 0.8% of the time, but these were not significant (p=0.533274, df=2, x2=1.257). All three of these final species showed tr ends that they had a high aversion to yellow feeders, as the Purple-throated Mountain-gem, Stripe-tailed Hummingbird, and Green Violet-ear onl y chose them 11.7, 3.8, and 3.1% of the time, respectively. Results of aggressive behavior (Fig. 5) were signi ficantly different between species (p=0.006019, df=9, x2=23.08). Aggressiveness was strongly exhibited by t he Violet Sabrewing, who out of 132 displays of aggres sion was the aggressor 73.5% of the time. Aggression was witnessed from one dominant ma le of this species. The three other species only occasionally exhibited aggressive beha vior, but rarely against the Violet Sabrewing. They would attempt a chase and fail, whi ch was not counted toward results of aggressiveness. It was evident that all other speci es present at the feeders were subordinate to the Violet Sabrewing. Efforts to com bat his dominance always failed, but sometimes other birds were very persistent. Aggress ion toward the Violet Sabrewing was only 6.8% of the total aggression observed (includi ng that observed by one Violet Sabrewing individual towards another). The Purple-t hroated Mountain-gem was the next most likely to exhibit aggression, most often towar ds the Stripe-tailed Hummingbird or the Green Violet-ear.
DISCUSSION I hypothesized that red feeders containing the 20% sugar solution would be preferred by all species, and that with increasing visitation in dividuals would discover that color did not necessarily indicate level of reward. This tren d was shown in my results, but significant differences were only found for the Vio let Sabrewing. Finally, I expected larger hummingbirds to be dominant, causing smaller species to choose their second choice food source more frequently over time, and t hat territoriality would be exhibited to an extent depending on species because of the natur e of all hummingbirds. Aggression was far more common in the largest species, the Vio let Sabrewing, and uncommon to rare for the other three observed species. The dominant male Violet Sabrewing quickly learned that reward was not colordependent. Quick learning could be attributed to hi s high dominance of all six feeders, especially that which occurred in the first few day s. He chose high concentration feeders much more frequently when he learned where they wer e spatially located at the beginning of each day. Stronger color preference wa s therefore observed in the first days of data collection because this individual soon lea rned to associate all three feeder colors with an equally high reward. My prediction was supp orted for this species, in that the high concentration red feeder was initially preferr ed, and the trend decreased as time progressed. The same Violet Sabrewing could be seen in his sen tinel perch in a nearby tree or shrub, often chirping to assert his presence. He wa s clearly territorial, returning to the same spot to guard against other hummers even when he wasnt feeding. His strong dominance of all feeders caused most other attempts at feeding by other birds impossible. Thus no color or sugar concentration was observed d uring this initial stage, due to low success of feeding by other hummingbirds. Preferenc es were also not easily observed because it was clear that each time another bird ca me to feed, it would fly to the nearest and most easily accessible feeder, to feed briefly before the Violet Sabrewing was aware of its presence. Aggression was most evident from the Violet Sabrew ing, and it was also the most successful. His dominance of the area is probably d ue to his large size in comparison to all of the other species that frequented the feeder s. Attempts at aggression by other individuals were largely unsuccessful, and always s o when displayed toward the Violet Sabrewing. This may explain why the Violet Sabrewin g may have seemed to be the only territorial bird, as unsuccessful attempts of aggre ssion were not counted in my data. The other species may have been territorial but simply unsuccessful at actually guarding the food source. It was still clear, however, that desp ite their efforts, all species were subordinate to the Violet Sabrewing, due to their l ack of feeding. The dominance of the Violet Sabrewing supports my prediction that size w ould affect level of aggression, as it was the largest bird observed during data collectio n. This does not explain, however, the aggression dis played by several females Purple-throated Mountain-gem individuals, towards s everal other species. This goes against my prediction that larger birds, usually ma les, would be aggressive, because the Purple-throated Mountain-gem is the smallest specie s that I frequently observed at the feeders. Although they usually were not successful at driving other birds away, divebombing and attempted chasing were fairly common. H owever, these actions were not
well reported in my results, which only considered successful aggressiveness. Most successful aggressiveness was only observed near to the end of my study. Territoriality exhibited by these birds can only be explained by i ncreased confidence as the dominant bird, the Violet Sabrewing, was more frequently abs ent, as they were potentially trying to fill his place at the territory-holder. It is uncle ar to me whether this species is always very territorial. Future research could potentially look specifically at this species and its interactions with others, as well as its level of t erritoriality on a regular basis. This study could be potentially improved by includ ing data on aggression from birds they may not have been successful at driving away others, despite their efforts. This might give a more accurate depiction of the level o f territoriality within each species. Also, the results may improve if hummingbirds were observed over a longer time period, increasing the hours of observation each day. More learning may be observed if the birds with less regular access to the feeders were able t o use them for more than 15 days. ACKNOWLEDGEMENTS I would like to thank all of those that played a pa rt in my project. Especially Tania, who led the way with statistics; Karen, who helped me get started and wa s always so positive; Pablo & Moncho, who were always there to answer my silly questions (and help me translate into Spanish); La Estacin Biolgica de Monteverde and its beautiful garden; and Ben, who s at with me under that tree and kept me awake on my rainiest of days. LITERATURE CITED Baker, H.G. 1975. Sugar Concentrations in Nectars f rom Hummingbird Flowers. Biotropica 7(1): 37-41. Beletsky, L. 2007. Travellers Wildlife Guides: Costa Rica Interlink Books. Northampton, MA. Blem, C.R. and L.B Blem. 2000. Rufous Hummingbird S ucrose Preference: Precision of Selection Varies with Concentration. The Condor 102(1): 235-238. Busby, W.H. 2000. Hummingbird Pollination of Epiphy tic Ericaeae in the Cloud Forest Canopy. Monteverde: Ecology and Conservation of a Tropical Cloud Forest New York. Fenster, C.B., G. Cheely., M.R. Dudash, and R.J Rey nolds. 2006. Nectar reward and advertisement in hummingbird-pollinated Silene virginica (Caryophyllaceae). American Journal of Botany 93: 1800-1807 Fogden, M. and P. Fogden. 2005. Hummingbirds of Costa Rica Distribuidores Zona Tropical, S.A. Miami, FL. Gonzlez-Gmez, P.L. and R.A. Vsquez. 2006. A Fiel d Study of Spatial Memory in Green-Backed Firecrown Hummingbirds ( Sephanoides sephaniodes ). Ethology 112(8): 790-795. Long, K. 1997. Hummingbirds: A Wildlife Handbook Johnson Books. Boulder, CO. McGee, A. 1998. Interaction of Color and Nectar Rew ard on Hummingbird Visitation. Tropical Ecology and Conservation CIEE. Monteverde, Costa Rica.
Meledez-Ackerman, E., D.R. Campbell, and N.M. Waser 1997. Hummingbird Behavior and Mechanism of Selection on Flower Color in Ipomopsis Ecology 78(8): 2532-2541. Pogue, T. 1992. Feeder Frenzy: Hummingbird Attracti on to Variation in Color and Sugar Concentration. Tropical Ecology and Conservation CIEE. Monteverde, Costa Rica. Sandlin, E.A. 2000. Foraging information affects th e nature of competitive interactions. Oikos 91: 18-28. Stiles, G.F. and A.F. Skutch. 1989. A Guide to the Birds of Costa Rica Cornell University. Ithaca, NY. Tiebout, H.M. 1993. Mechanisms of Competition in Tr opical Hummingbirds: Metabolic Costs for Losers and Winners. Ecology 74(2): 405-418.
Violet Sabrewing Feeder Visit Frequencies Over 11 D ays92 17 23 19 24.93 17.15 56 51 16.92 57.85 57.08 84.07 0 10 20 30 40 50 60 70 80 90 100 red, highpurple, highyellow, highred, lowpurple, lo wyellow, low Feeder (color, sugar concentration)Visit Frequency observed expected Figure 1. The frequency of visits made by Violet Sa brewing individuals at all six feeders over 11 days, observed (blue) and expected (purple) Results were statistically significant (p=0.041669, df=2, x2= 6.356). This species preferred high concentration red feeders first, then high concentration purple and yellow fe eders about equally. Purple-throated Mountain-gem Feeder Visit Frequenci es Over 11 Days54 23 11 37 30 8 28.61 10.26 41.87 8.74 24.39 49.13 0 10 20 30 40 50 60 red, highpurple, highyellow, highred, lowpurple, lo wyellow, low Feeder (color, sugar concentration)Visit Frequency observed expected Figure 2. The frequency of visits made by Purple-th roated Mountain-gem individuals at all six feeders over 11 days, observed (blue) and e xpected (purple). Trends showed that this species preferred the red feeders over the oth er colors, but no conclusions can be drawn as results were not statistically significant (p=0.168649, df=2, x2=3.560).
Stripe-tailed Hummingbird Feeder Visit Frequencies Over 11 Days 58 28 3 25 17 2 55.54 30.11 3.35 27.46 14.89 1.65 0 10 20 30 40 50 60 70 red, highpurple, highyellow, highred, lowpurple, lo wyellow, low Feeder (color, sugar concentrationVisit Frequency observed expected Figure 3. The frequency of visits made by Stripe-ta iled Hummingbirds at all six feeders over 11 days, observed (blue) and expected (purple) Trends showed that this species preferred the high concentration red feeder and mos tly avoided the yellow feeders, but results were not statistically significant (p=0.642 385, df=2, x2=0.8851). Green Violet-ear Feeder Visit Frequencies Over 11 D ays40 26 2 14 15 1 37.47 28.45 2.08 16.53 12.55 0.92 0 5 10 15 20 25 30 35 40 45 red, highpurple, highyellow, highred, lowpurple, lo wyellow, low Feeder (color, sugar concentration)Visit Frequency observed expected Figure 4. The frequency of visits made by Green Vio let-ear individuals at all six feeders over 11 days, observed (blue) and expected (purple) Trends showed that this species especially preferred the high concentration red as well as purple feeders, avoiding the yellow feeders, but results were not statistically significant (p=0.533274, df=2, x2=1.257).
Observed Frequency of Aggressiveness between All Fo ur Species of Monteverde Hummingbird2 4 1 2 33 1 33 22 7 22 40 8 2 0 0 5 10 15 20 25 30 35 40 45 Violet SabrewingPurple-throated Mountain-gem Stripe-tailed Hummingbird Green Violet-ear SpeciesFrequency of Aggressiveness Violet Sabrewing Purple-throated Mountain-gem Stripe-tailed Hummingbird Green Violet-ear Figure 6. Observed frequency of aggressiveness betw een Violet Sabrewing (blue), Purple-throated Mountain-gem (pink), Stripe-tailed Hummingbird (orange), and Green Violet-ear (green). Statistically significant resul ts (p=0.006019, df=9, x2=23.08) showed differences in frequency of aggressive behavior, wi th much more aggression exhibited by the Violet Sabrewing, especially towards the Green Violet-ear and Purple-throated Mountain-gem.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00052
Mundy, Laurel Anne
Agresin en la interaccin del color y la preferencia de la concentracin de azcar en los colibres (Trochilidae) del bosque nuboso de Monteverde
Interaction of color and sugar concentration preference with aggression in Monteverde Cloud Forest hummingbirds (Trochilidae)
In Monteverde, Costa Rica, a number of hummingbird species coexist that are extremely important for pollination. Hummingbirds may prefer flowers of a certain color because they associate it with a higher nectar reward. They also may be territorial and defend especially rich food sources, dominant birds driving away all subordinates. The purpose of this study was to investigate color and sugar concentration preference in hummingbirds using artificial feeders, and to determine whether dominance and aggression has an effect on food choice. Many hummingbird-pollinated flowers are red, so I hypothesized that all species would show a preference, and predicted that all would initially choose the red feeder of highest sugar concentration, eventually also choosing the two other high concentration feeders, realizing their equal reward. I also predicted that larger, male hummingbirds would dominate the food source more territorial, dominating the food source and forcing smaller birds to choose their second choice feeder. This was done by hanging six feeders of red, yellow and purple and high and low sugar concentration and recording species preference. I found that the Violet Sabrewing was the species that most often chose the red, high concentration feeder, as predicted (p=0.041669, df=2, x2= 6.356). No conclusions could be made about the three other species investigated as results were not significant. Preference was seen in the Violet Sabrewing, as differences in feeder preference decreased when only the last four days of data were analyzed, suggesting that color preference decreased. Most aggression was observed from the Violet Sabrewing (p=0.006019, df=9, x2=23.08). This species was clearly dominant likely due to its large size.
El propsito de este estudio fue investigar el color y la preferencia de la concentracin de azcar en colibres usando comederos artificiales, y determinar si la dominancia y la agresin tienen un efecto en la eleccin de alimentos.
Text in English.
Hummingbirds--Feeds and feeding--Costa Rica--Puntarenas--Monteverde Zone
Colibres--Alimentos y alimentacin--Costa Rica--Puntarenas--Zona de Monteverde
Tropical Ecology 2008
Ecologa Tropical 2008
t Monteverde Institute : Tropical Ecology