Preferred sucrose concentration in nectarivorous bats Jeremy C. Sullivan Department of Biology, Northeastern University, Boston, Massachusetts 02115, U.S.A. ABSTRACT Pollinators and plants are involved in a mutualism, where pollination takes place in exchange for some payment, usually in the form of nectar. One might assume that these nectar rewards would closely match a pollinatorÂ€s preference, but flowers may conserve energy by offering an unfavorable, but still suitable nectar reward. Bat pollinated flowers have been found to offer nectar of an 18% sucrose concentration, on average, over a range of 3% 51% Baker 1998. Where in this range of sucrose concentrations do bat preferences lie? To find out, sucrose concentration preferences of a community o f Neotropical bats were studied at the Hummingbird Gallery in Monteverde, Costa Rica. Four hummingbird feeders with solutions of 10%, 20%, 30% and 40% sucrose concentration were observed for 20 nights. Frequency of bat visitations to each feeder was record ed in ten minute intervals. Bats visited the feeders of higher sucrose concentration 30% and 40% more than feeders with lower sucrose concentration 10% and 20%. These findings support the idea that animal pollinated flowers may be providing a minimum r eward to pollinators to ensure increased visitation frequency and cross pollination. RESUMEN Los polinizadores y plantas se envuelven en un mutualismo, donde la polinizacin toma lugar el intercambio por un pago, habitualmente en la forma de nctar Se pue de asumir que esta recompensa de nctar es estrecha con la preferencia del polinizador, pero las flores pueden conservar energa al ofrecer un nctar no favorable pero seguir ofreciendo nctar como recompensa. Las flores que los murcilagos polinizan, tiene una concentracin de nctar que es 18% de sucosa y tiene un rango de 3% 51%. Cual es la concentracin preferida por los murcilagos ? Para responder esta pregunta, se estudio la preferencia en la concentracin de sucosa en la Galera de C olibres en Monteverde, Costa Rica. Cuatro comederos de colibres con soluciones de 10%, 20%, 30% y 40% de sucosa fueron observados por veinte noches. La frecuencia de visitas de murcilagos en cada comedero fue recogida en intervalos de diez minutos. Los murcilagos visitaron los comederos con mayor concentracin de sucosa 30% y 40% mas que los comederos con menor concentracin de sucosa 10% y 20%. Estos resultados apoyan la idea que las flores polinizadas por animales proporcionan una recompensa mnim a a los polinizadores para asegurar que visiten las flores y la polinicen. INTRODUCTION Many tropical Plants depend on bats for pollination Johnson 1986. In Monteverde, Costa Rica, 33 species of plants are bat pollinated. These plants belong primarily to the
Marcgraviaceae, Bombacaceae, Bignoniacea, and Cactaceae families Nadkarni and Wheelright 2000. Floral syndromes of bat pollinated plants include flowers that open at dusk, have large white or yellow petals, have a musky odor and produce large quanti ties of nectar Bawa 1998. Bats and other pollinators are attracted to flowers by a nectar reward, which assures visitation by potential pollinators. Since the number of visits to flowers by an individual pollinator is determined by its need for food, it is essential that the pollinators are not too readily satiated Heinrich 1973. Ideally a flower should provide sufficient reward to attract pollinators, but limit this reward so that animals visit other flowers of the same species. This, not only ensures cross pollination, but also saves energy for the plant. Nectar concentrations of bat and bird pollinated flowers have been researched by Baker et al, in a 1998 study. They found that Neotropical bats pollinate flowers with nectars that exhibit a range of sucrose concentrations from 3% to 51%. They also found that the nectars of plant families pollinated by Neotropical bats have a mean of 18% sucrose concentration. Roces et al., in a 1993 study, show that at least one bat species, Glossophaga soricina, pre fer a higher nectar concentration up to 50%, when given a range of 10% to 80% sugar concentration to choose from. This study was conducted in a flight tunnel in a laboratory and dealt with one bat species. They also measured volumes of nectar removed to de termine preference. The study presented in this manuscript involves an entire guild of bats in their natural habitat and measures frequency of bat visitations to determine preference. Roces 1993 results suggest that bats may prefer a higher sugar concent ration than is present in the nectar of most bat pollinated flowers. Similar results have been found in hummingbirds Roberts 1998, who most frequently visited feeders with the highest available sucrose concentration. These results support the hypothesis that plants are providing a minimum reward to pollinators to ensure a high frequency of visitation. Bats and other pollinators may require a minimum acceptable sugar concentration of the nectar they ingest Roces et al. 1993. This minimum level can be in fluenced by the minimal energetic needs of the pollinator, osmoregulatory constraints and digestive speed and capacity Roces et al. 1993. Upper limits to the acceptable level of sugar concentration in nectar consumed by bats may be due to the exponential increase in viscosity of a solution at high sugar concentration Baker 1975. This increased viscosity reduces fluid ingestion rates and increases the amount of time and energy needed to consume a quantity of nectar Roces et al. 1993. Another factor tha t contributes to this upper limit is osmotic regulation in bats. Bats must maintain a water budget to facilitate food transit through the digestive tract Roces et al. 1993. Roces et al 1993 found that for G. soricina the proportion of non nectar water ingested during a night, increased with the concentration of nectar present. Therefore, it is possible that bats prefer nectar with a higher sugar concentration than most bat pollinated flowers provide. In addition this preferred nectar should not exceed the upper limits of sugar concentration where viscosity and low water to sugar ratio increase the bats energetic cost of foraging. In this study, sucrose concentration
preferences in a community of nectarivorous bats were studied by presenting bats with s olutions of different sucrose concentrations, and observing the frequency of bat visitations to each solution. METHODS Study Site Nectarivorous bats were studied for 20 nights from October 28 to November 19, 2000 at the Hummingbird Gallery in Monteverde, Costa Rica, where bats forage on hummingbird feeders at night. The Hummingbird Gallery is located in the Monteverde Cloud Forest Pr eserve at an altitude of 1535m Premontane Wet Forest Lifezone. Mist Netting To determine which species of nectarivorous bats were likely to visit the feeders, mist nets were used to catch and identify bats on the nights of October 21 and October 24. One ten meter mist net was extended along the bats flight patch to the feeders. Bats were caught and identified to species. Concentration Preference Sucrose concentration preference by bats was measured by counting the number of bat visitations to each of four feeders with different sucrose concentrations. These feeders had four openings and perches were present extending from each opening. A bat visitation consisted of a bat foraging on the feeder. Bats were presented with four identical hummingbird feeder s containing 10%, 20%, 30%, and 40% sucrose solutions. Feeders were mounted one meter apart and suspended two meters off the ground on a nylon cord. Feeders were positioned along the bats flight path to ensure immediate discovery. Each feeder was observed in ten minute intervals and bat visitations were counted. The position of feeders was rearranged each night to eliminate a possible positional bias a greater frequency of bat visitation to feeders placed on ends. No attempt was made to identify bats as t hey visited the feeders. Bats were observed for 120 minutes between 6:30pm and 8:30pm each night for 20 nights. RESULTS Mist Netting A total of fifteen bats were captured in mist nets and identified to species Timm and Laval 1998. All bats were nectarivorous and belonged to the Glossophaginae family Table 1. One Anoura geoffroyi five Glossophaga commissarisi and nine Hylonycteris underwoodi were captured.
Concentration Preference A significant difference in visitation frequency was found for the four sucrose concentrations, indicating that bat preference is effected by sucrose concentration Friedman test for 4 variables, P Value = .0231. Mean visitation frequency and total visitations were higher for sucrose concentrations of 30% and 40% Fig 1 and Fig 2. The mean visitations rates to each feeder and their standard deviations were as follows: 10%: mean = 14.80, S.D = 13.12, 20%: mean = 14.30, S.D. = 15.68, 30%: mean = 26.40, S. D. = 17.49, 40%: mean = 27.50, S.D. = 18.07. The sum ranks for visitations to the 30% and 40% solutions were higher than for 10% and 20% solutions Friedman rank test, 10% = 42, 20% = 40.5, 30% = 61, 40% = 56.5. A multiple comparison test did not reveal any significant pair wise differences in visitation frequency 10% vs. 20%, q= 0.259; 10% vs 30%, q = 3.29; 10% vs 40%, q = 2.51; 20% vs 30%, q = 3.55; 20% vs 40%, q = 2.77; 30% vs 40%, q = 0.779, cri tical value for q= 3.633. Nightly visitation rates were found to vary widely for all concentrations, but values for the 30% and 40% were higher than for the 10% and 20% on a majority of the nights Fig 3.. DISCUSSION The guild of nectarivorous bats containing A. geoffroyi G. commissarisi and H. underwoodi tended to visit feeders containing 30% and 40% sucrose solutions more frequently. Therefore, it is clear that nectarivorous bats in the Monteverde community prefer more sucrose rich nectars than commonly found by Baker et al. 1998 in bat pollinated flowers mean of 18%. Around 20% sucrose concentration in nectar is the lower limit, physiologically, for bats Helversen 1993. Bats have an extremely high energy req uirement, which is probably at the upper limit for mammals Helverson & Reyer 1984. Nectars with a lower sugar concentration may not provide enough energy to the bat to cover the energy expenditure of foraging. So, bat pollinated plants, on average, seem to produce nectar around this minimum, and invest the least amount of energy possible to ensure repeated bat visitation. Still, there are plants that offer nectar well below and above the 18% average reported by Baker 1998. Flowers with a lower nectar c oncentration may be deceiving bats through mimicry of some sort, or could be visited when other flowers with higher concentrations of nectar are not available. Flowers with a higher sucrose concentration in their nectar should be visited preferentially by bats. A well developed ability for pattern recognition Heinrich 1975 and an excellent spatial memory allow bats to find better nectar providers quickly. Therefore, in environments with high competition for bat visitations, plants may be forced to provide nectars with a higher sucrose concentration than the minimum of around 20%. Still, because the mean sucrose concentration 18% is so close to the minimum, this situation of competition for bat visits among plants may not arise often.
The fact that very few bat pollinated plants exceed 50% sucrose concentration Baker 1998, which is the physiological maximum for bats, is also interesting. At this high concentration great amounts of non nectar water might be required to balance the bats daily water budget and a high viscosity might impede fluid intake. Also the production of nectar with a high sucrose concentration might cost the plant large amounts of energy. Finally, if hummingbird feeders usually offer a solution with a 20% sucrose concentration to bir ds, bats foraging on those feeders can get an abundant supply of adequate nutrition that also maintains a proper water balance. This may increase bat populations to levels that disrupt the local ecosystems or decrease bat visitations to flowers with a 20% sucrose nectar or less, and thereby decrease cross pollination. ACKNOWLEDGEMENTS I wish to thank Alan Masters, Karen Masters, and Mauricio Garcia for their support, guidance and constructive criticism. Also IÂ€m greatly appreciative of all the help Tim Kuh mun and Andrew Rodstrom offered to make this study possible. In addition, the advice and support of Richard K. Laval has been invaluable in the production of this manuscript. I am indebted to the owners and staff of the Hummingbird Gallery for allowing me to conduct research there. Christina Keppel, I thank you for being such good company on those rainy nights at the Hummingbird Gallery. Lastly I would like to thank the students and faculty of CIEE 2000 for making this the most worthwhile experience of my a cademic career. LITERATURE CITED Baker, H.G., Baker, I., and Hodges S.A. 1998. Sugar Composition of Nectars and Fruits Consumed by Birds and Bats in the Tropics and Subtropics. Biotropica 304: 559 586. -------, 1975. Sugar Concentrations in Nectars from Hummingbird Flowers. Biotropics 71: 37 41. Bawa, K.S. 1990. Plant Pollinator Interactions in Tropical Rain Forests. Annual Review of Ecological Systems 21: 399 422. Heinreich, B. 1973. The Role of Energetics in Bumblebee Flowers Interelationships. In L.E. Gilbert and P.H. Raven Eds, pp. 141 158 Coevolution of Animals and Plants. University of Texan Press, Austin, U.S.A. Helversen, O.V. 1993. Adaptations of Flowers to the Pollination by Glossophagine Bats. In W. Barthlott et al. Eds, pp. 159 165. Plant Animal Interactions in Tropical Environments. Museum Alexander Koenig, Bonn. Nadkarni, N.M., and Wheelwright, N.T., 2000. Monteverde, Oxford University Press, New York, U.S.A. Proctor, M., Yeo, P., and Lack, A. 1996. T he Natural History of Pollination. Timber Press, Portland, U.S.A.
Raven, P.H. and Johnson, G.B. 1986. Biology. Times Mirror/Mosby College Publishing, St. Louis, Missouri, U.S.A. Roberts, M.R. 1996. Hummingbirds Nectars Concentration Preferences at Low Vol ue: The Importance of Time Scale. Animal Behavios 52: 361 370. Roces, F., Winter, Y., Helverson, O.V. 1993. Nectar Concentration Preference and Water Balance in a Flower Visiting Bat: Glossophaga soricina antillarum. In W Barthlott et al. Eds, pp. 1 59 165. Plant Animal Interactions in Tropical Environments. Museum Alexander Koenig, Bonn. Timm, R.M., LaVal, R.K. 1998. A Field Key to the Bats of Costa Rica. The Center of Latin American Studies No. 22: 1 30.
TABLE 1. Monteverde nectarivorous bat s and the number captured at the Humming Gallery in Moneverde Reserve Scientific Name Common Name Abundance Number captured Anoura cultrate HandleyÂ€s Tailless bat Uncommon 0 Anoura geoffroyi GeoffroyÂ€s Tailess Bat Common 1 Choeroniscus godmani GodmanÂ€s Long nosed bat Uncommon 0 Glossophaga commissarisi CommissarisÂ€ Long tongued bat Common 5 Glossophaga soricina PallasÂ€ Long tongued Bat Common 0 Hylonycteris underwoodi UnderwoodÂ€s Long tongued Bat Common 9 FIGURE 1. Average visitation frequency and S.D.Â€s to hummingbird feeders with different sucrose concentrations over a 20 day period. Data was gathered at the Hummingbird Gallery in the Monteverde Reserve Premontane Wet Forest Lifezone. Mean visitations were ca lculated for ten minute intervals at each feeder. This demonstrates that bats, on average, prefer a higher sucrose concentration than is offered by most bat pollinated plants which have a mean sucrose concentration of 18%.
FIGURE 2. Total visitations for different sucrose concentrations over the entire 20 day study period to each hummingbird feeder. 30% and 40% solutions were visited more often than 10% and 20% solutions, indicating that a higher sucrose concentration is preferred by a ajority of bats in the area Premontane Wet Forest Lifezone.