Changes in sugar concentration and volume of nectar in Centropogon solanifolius Campanulaceae Jeffrey Benjamin Sanders Department of Biochemistry, University of Wisconsin, Madison ________________________________ ______________________________ Abstract Changes in both the sugar concentration and nectar volume produced by hummingbird pollinated plants may be evolved features to dictate the actions of the pollinators to optimize reproductive success in the flower. This study attempts to determine the changes in sugar concentration and nectar volume produced over the lifespan of Centropogon solanifolius Campanulaceae, a hummingbird pollinated flower. Hummingbird mites Ascidae data was examined to look for correlations between colonization and nectar volume. Nectar volume and sugar concentration were recorded for bagged flowers rose early in the plants life and slowed as the chance of the flower already being pollinated increa sed. Mite colonization proportions mimic nectar volume accumulation patterns. The sugar concentration of nectar remained constant throughout the flowers life and no correlation was found between sugar concentration and nectar volume. A floral nectar produc tion strategy is proposed for C. solanifolius to explain the patterns seen here. Resumen Los cambios en la concentraciÃ³n de azÃºcar y el volumen de nÃ©ctar que estÃ¡n producido por las plantas que estÃ¡n polinizado por los colibrÃs , pueden ser elementos ev olucionados para mandar las acciones de los polinizadores para optimizar el Ã©xito reproductivo de la flor. Este estudio trata de determinar los cambios en la concentraciÃ³n de azÃºcar y el volumen de nÃ©ctar que estÃ¡n producido durante de la vida de Centropog on solanifolius Campanulaceae, una flor que es ta polinizado por los colibrÃes . Los datos de los acÃ¡ridos de los colibrÃs Ascidae estaba examinados para buscar correlaciones entre colonizaciÃ³n y el volumen de nÃ©ctar . El volumen de nÃ©ctar y la concentraciÃ³n de azÃºcar que estaban recordado para las flores que estaban cubiertos, subieron temprano en la vida de las plantas y lentaron cuando la posibilidad de la flor que ya estÃ¡ polinizado subiÃ³ . Las proporciones de la colonizaciÃ³n de los lacÃ¡r idos imitan el molde de la acumulaciÃ³n del vo lumen de nÃ©ctar. La concentraciÃ³n de azÃºcar del nÃ©ctar quedaba constante durante la vida de la flor, y no correlaciÃ³n estaba encontrado entre la concentraciÃ³n del azÃºcar y el volumen de nÃ©ctar. Un estrategia de la producciÃ³n del nÃ©ctar de las flores esta propuesto para C. solanifolius para explicar los moled que estaban observado. Introduction The volume of nectar produced by a hummingbird pollinated flower may vary widely within a single plant population and affect the patterns in which hummingbirds move among them Feinsinger 1983. Plant populations that rely on hummingbirds for out -
crossing, must provide enough of a reward to the pollinator to encourage visitation. Howev er, the reward should not satiate the pollinator or it will not be compelled to visit the other conspecific flowers at other sites Klinkhamer and De Jong 1993. Hummingbird pollinated plants have evolved numerous physiological attributes in order to manip ulate the behavior of their pollinators. Baker 1975 showed that hummingbird pollinated flowers have a low sugar concentration 20% sucrose by weight in their nectar. He noted that nectar with lower sugar concentration is less viscous and thus evolved un der the pressure of hummingbirds to increase their foraging efficiency by shortening the time they require to consume less viscous nectar from the flower. Hummingbird pollinated flowers produce a continuous supply of nectar, but nectar secretion rates vary widely among flowers. Data have shown that older flowers secrete less nectar than younger flowers and that no consistent relation between quantity of nectar and the sugar concentration exists Feinsinger 1978. Nectar volumes found in some hummingbird pol linated flowers show a "bonanza blank" pattern Feinsinger 1978, Bolton & Feinsinger 1978, and Feinsinger 1983. That is, some flowers secrete very little or no nectar "blanks," and other flowers secrete copious amounts "bonanzas." This pattern of nectar production encourages hummingbirds to increase the duration of their foraging events while reducing the caloric expenditure of the plant. Centropogon solanifolius Campanulaceae is a common understory hummingbird pollinated herb with orange protandrous fl owers. Centropogon solanifolius has asynchronous flowering and grows along cloud forest trails Zuchowski 1996. Centropogon solanifolius flowers contain many mites Ascidae that feed on pollen and nectar within the flowers and can significantly affect th e volume of nectar present Colwell 1995. The lifespan of C. solanifolius flowers is 7.0 8.4 days Stratton 1989, Weiss 1996, but there are neither data on the volume of nectar produced over the life of the flower, nor of changes in the sugar concentrati on of the nectar. This study attempts to determine how the sugar concentration of nectar and/or the volume of nectar produced in a hummingbird pollinated flower changes throughout its lifespan. Changes in both the sugar concentration and nectar volume prod uced by the plant may be evolved features to dictate the actions of the pollinators to optimize reproductive success in the flower. A concurrent study by Shelly Gordon this volume looks at mite populations in C. Solanifolius . Together, the studies reveal how the plant and the mites interact. Materials and Methods I conducted this study along the Se ndero Mirador behind the EstaciÃ³ n BiolÃ³gica de Monteverde, Puntarenas, Costa Rica between April 16 th and May 5 th , 2002. The study site was located in lower montane rainforest Holdridge 1967. I labeled un open C, solanifolius flowers with plastic tags tied below the corolla with metal wire so as not to alter the appearance of the flower for the pollinators. I placed mesh nylon bags aroun d and over the flowers to prevent hummingbird visitation and to limit the exposure to hummingbird s mites. Some flowers were also left open without bags to allow for hummingbird consumption of nectar and to act as a control group. Bagged and open flowers were chosen at random as in Gill 1988. I monitored the flowers everyday and noted the day on which each flower opened. The days were numbered consecutively and
correspond to what I call "flower age." Nectar was collected with 100 Â€ l Pipettes. I inserted the pipette into the nectar well at the bottom of the flower and applied suction to draw the nectar into the pipette I calculated the volume Â€ l of nectar in the pipette using r 2 Âh . The radius r of the pipette = 0.7 mm, and the height of the nectar colu mn h was de termined using calipers to to 0 .lmm. To obtain the sucrose concentration equivalence by weight/ total weight of solution of the nectar I placed the nectar on the prism of a % sucrose Hand Refractometer Reichert Co., and noted the value o n the screen. Nectar volume and sugar concentrations were measured in 70 bagged flowers BF 10 flowers per day, days 1 7 and 54 open flowers OF. I compared BF and OF nectar volume and sugar concentration data using a Kruskal Wallis non parametric test. I used a one way ANOVA to test the effects of the flower age on nectar volume present and sugar concentration Zar 1984. A simple regression was used to test for any correlation between sugar concentration and nectar volume fo r bagged flowers. Results The volume of nectar in BF changes significantly over the lifespan of the flower Kruskal Wallis p value < 0.0001, H corrected for ties 38.946. A post hoc analysis was conducted a = 0.05, V = 70, q= 5.863 and showed Day 5 nectar volume was significantly higher from Day 2 nectar volume. Kruskal Wallis tests showed no significance between sugar concentration measurements of bagged flowers vs. day number, volume of nectar in open flowers vs. Da y number, and Sugar concentration measurements of open flowers vs. Day number H corrected for ties: 9.041, 6.252, and 3.993 respectively, p> 0.05 for all. A simple regression showed no correlation between nectar volume and sugar concentration of bagged f lowers p value 0.4968, R 2 = 0.097. Discussion I . Nectar Volume The nectar volume of BF increased from Day 1 to Day 5, then decreased from Day 5 to Day 7. In general, older hummingbird pollinated flowers secrete less nectar than younger ones Feinsinger 1978. This production patte rn is evident in my data Fig. 1 . The mean nectar production for BF on Day 1 was 34.74 Â€ l and production rose slowly until Day 5 when total mean nectar ac cumulation plateaued at 64.07ul . Nectar accumulation in BF began to decl ine after Day 5 and continued declining to Day 7. The reduction in nectar accumulation in the latter days of the flowers' life may have been caused by reabsorbtion of the nectar by the flower before senescence discussed below. Flowers left open showed no trend in nectar volume accumulation and the data were highly variable Table 1, Fig. 1. OF were subject to visits from hummingbirds personal observation and habitation by mites Gordon this volume, both of whom consume significant amounts of nectar. I found very little nectar in OF, results that are similar to those of Colwell 1995. Nectar production in many hummingbird pollinated plants occurs early in the morning Feinsinger 1983, Gill 1988, Colwell 1995 when I observed the most hummingbird activi ty around C. solanifolius. The low nectar volume found in OF may have been caused by hummingbird visitations before data were
collected. The proportion of flowers colonized by mites rises from Day 1 through Day 6 when 100% of the flowers sampled were colonized by mites Fig. 2 from Gordon this volume. The proportion of flowers colonized by mites by day mimics the nectar volume accumulation by day of bagged flowers Fig. 1. The implications of this observation will be discussed later. Nectar volumes among BF were highly variable across day number for the populations ranging from 13.85 Â€ l to 100.34 Â€ l Table 1. Figure 1 shows that the accumulated nectar on Day 2 was lower than on Day 1. The coefficient of varian ce Table 1 shows also that Day 2 BF had the highest variability compared to all other days. The only likely way to explain the lower mean on Day 2 is if some of the Day 2 samples had little or no nectar on Day 1. The large range in accumulated nectar vol ume over the lifespan of the flower and th e high variance in accumulated n ectar volume may be explained by Feinsinger's 1978 "bonanza blank" pattern in which some flowers produce copious amounts of nectar while others produce little or no nectar. The coe fficient of variance Table 1 indicates that accumulated nectar volume is highly variable until Day 5 and declines thereafter. This suggests that if C. solanifolius' nectar production pattern is described by the "bonanza blank" pattern, it is happening before Day 6 and most likely on Days 1 and 2. This nectar production strategy is employed by the plant to attract and sustain hummingbird pollinators, but keep them from becomi ng sedentary. Plants that secrete the minimum amount of nectar to keep pollinators from becoming sedentary have the highest frequency of pollen transfer since pollinators must visit many flowers on different plants in order to satisfy their daily metabolic requirements Klinkhamer and De Jong 1993. II. Sugar Concentration The sugar concentration of nectar remained consistent over the life of C. solanifolius flowers for BF and OF Fig. 3. Percent sucrose measurements averaged 24.90 in BF and 24.70 in OF wh ich is consistent with Bolton & Feinsinger's 1978 data showing an average % sucrose measurement of 26.2 for a Centropogon species. These data show a relatively low sugar concentration compared with the nectar of flowers pollinated by other animals Baker 1975. The low sugar concentration of nectar in hummingbird pollinated flowers could be an adaptation to deter nectar robbery by less effective pollinators Bolton and Feinsinger 1978. Low sugar concentration could also be an adaptation to lower nectar v iscosity, particularly at lower temperatures, and increase hummingbirds foraging efficiency Baker 1975. There was no correlation between the sugar concentration and nectar volume of BF which is consistent with previously reported data Feinsinger 1978, B urquez & Corbet 1991, see Fig. 4. Sugar concentration remained constant around 25 Brix as nectar volume rose from Day 1 to Day 5, and remained constant as nectar volume fell from Day 5 to Day 7. The consistent sugar concentration in the nectar as the volume decreased indicated that evaporation could not have been the cause of the reduced of nectar volume. Loss of nectar volume by evaporation would cause sugar concentrations to rise above the constant % sucrose level of 25.1 hypothesize three mechanisms to explain the loss of accumulated nectar in Day 5 to Day 7 BF: i Centropogon solanifolius flowers are subject to winds and rains which could tip the flower down and cause nectar to simply pour out of
the corolla. This hypothesis is dubious because the nectar in C. solanifolius is stored in a well that is difficult to penetrate with a glass pipette. Simply tipping the flower should not be enough of a force to allow nectar to escape, ii Hummingbird mites reside in C. solanifolius and consume high amount s of nectar. Through continued consumption of nectar combined with the flowers ceased production, nectar levels would drop in the later days of the flowers life before senescence. Looking at mite colonization proportions from Figure 2 for an open flower sy stem, the highest colonization proportion is in the latter days of the flowers life. Perhaps the reduction in nectar volume in our controlled bagged data set can be correlated with the high colonization rates in the later days of the flowers life in an ope n system? iii Nectar accounts for a sizeable proportion of the carbon budget of a flower Southwick 1984. Nectaries may be able to reabsorb nectar Daumann 1930 in Endress 1994, Barquez & Colbert 1991. A flower that reabsorbs nectar not removed by visi ting nectarivores can reclaim at least part of the energy allocated to nectar production. I attribute the loss of accumulated nectar in the later days of the life of C. solanifolius to re absorption of nectar by the flower. Conclusion I found that production in C. solanifolium was highest in younger flowers and slowed as the flower aged. The total nectar accumulation of the flower fell as the flower approached senescence and the mite colonization levels are the highest. I hypothesize th at the decrease in nectar accumulation was caused by the plant reabsorbing the nectar in response to the eminent senescence and high mite colonization. The sugar concentration of the nectar remained constant throughout the life of the flower and did not ch ange as the volume of nectar accumulation rose and fell. Thus, the hummingbird pollinated flower displays evolved features to attract and reward pollinators, conserve its valuable nutrient resources, and maintain an optimal level of pollen distribution. I hypothesize an overall nectar production strategy for C. solanifolius: Day 1 & 2: C. solanifolius produces nectar in a "bonanza blank" pattern Table 1 in order to maximize pollen dispersal and reduce caloric expenditure by the plant. 60% of open flowers are colonized by mites Fig. 2, whose consumption of nectar may contribute to the volume variance in OF. The high variance causes hummingbirds to visit other plants, in search of "bonanzas," thereby spreading pollen to conspecific flowers in a way that pr omotes outbreeding. Day 3: 80% of flowers are now colonized by mites Fig. 2, making it risky for a flower to produce more nectar. If hummingbirds have not yet removed nectar, the plants continue to invest in nectar but at declining rates of accumulation Table 1. This may reflect a trade off between the need to attract pollinators and deter mite population gr owth. Day 6 & 7: All open flowers have mites in them Fig. 2 which likely means that they have all been visited by hummingbirds, and the open flowers lack nectar. The lack of nectar is not simply a consequence of an expanding mite population that consumes all of it, because mite populations do not increase with flower age Gordon this volume. Rather, C. solanifolius flowers are probably reabsorbing nectar. The flower is reabsorbing
nectar because it is nearing senescence and has likely been visited alread y by a pollinator. Also, because mite colonization levels are the highest Fig. 2 the flower may be reabsorbing the nectar to prevent the mites from consuming it. ACKNOWLEDGMENTS I would like to thank Karen Masters for her guidance and patienc e, Alan Masters for his hugs, ver. and Gatorade , Mauricio Garcia for his help with my Spanish, Andy Rodstrom for his knowledge of everything and for his sarcasm, Will Welder for his wizardry with excel and for answering my stupid questions, the Cruz family for the ir warmth and hospitality, the EstaciÃ³n BiolÃ³gica for their use of the forest, Shelly Gordon for her input on Centropogon, Anna Keller for listening to me whine and helping me with my Spanish, Scotty for teaching me the past participle, Phil Cock for helpi ng me be cool and with excel, cuarto numero uno for all the laughs and Pilsen/Bavaria for not judging me. ________________________________________________________________ __________________________ LITERATURE CITED Baker, H.G. 1975. Sugar concentrations i n nectars from hummingbird flowers. Biotropica 71: 37 41. Bolten, A.B. and Feinsinger, P. 1978. Why do hummingbird flowers secrete dilute NÃ©ctar ? Biotropica 104 307 309. Burquez, A and Corbet, S.A. 1991. Do flowers reabsorb nectar? Functional Ecology. 53: 369 379. Colwell, R. K. 1995. Effects of nectar consumption by the hummingbird flower mite Procolaelaps kirmsei on nectar availability in Hamelia patens. Biotropica 272: 206 217. Daumann, E. 1930. Das Blutennektarium von Nepenthes. Beitr. Bot. Centralbl., Abt. I, 47:1 14 Endress, P.K. 1994. Diversity and evolutionary biology of tropical flowers. Cambridge University Press, New York, Melbourne, pp. 155 162. Feinsinger, P. 1978. Ecological interactions between plants and hummingbirds in a successional tropical community. Ecological Monographs 48:269 287. ------. 1983. Variable nectar secretion in a Heliconia species pollinated by Hermit Hummingbirds . Biotropica 151: 48 52. Gill, F. B. 1988. Effects of nectar removal on nectar accumulation in flowers of Heliconia imbricata Heliconiaceae. Biotropica 202: 169 171. Gordon, S. 2002 spring. Population and community structure of hummingbird mites in Cenropogon solanifolius Campanulaceae. CIEE tropical conservation. Holdridge, L.R. 1967. Lifezone ecology. Tropical Science Center, San Jose, Costa Rica. Klinkhamer, P.G.L and De Jong, T.J. 1993. Attractiveness to pollinators: a plant's dilemma. Oikos 66:180 184. Southwick, E.E. 1984 Photosynthate allocation to floral nectar; a neglected energy Investment . Ecology. 65:1775 1779. Stratton, D.A. 1989. Longevity of individual flowers in a Costa Rican cloud forest: e cological correlates and phylogenetic constraints. Biotropica 214: 30 8 318.
Weiss, M.R. 1996. Pollen feeding fly alters floral phenotypic gender in Centropogon solanifolius Campanulaceae. Biotropica 28 4b: 770 773. Zar, J.H. 1984. Biostatistical Analysis . Prentice Hall. Englewood Cliffs, New Jersey. Zuchowski, W. 1996. Common flowering plants of the Monteverde Cloud Forest Reserve. Tropical Science Center. San JosÃ©, Costa Rica. P. 29.
Table 1 . Data showing changing nectar volume, wide nectar volume range and variance measurements for bagged flowers. Bagged Flower Data Day number N Rate of accumulation % new nectar Mean nectar volume Â€ Â± SD Coefficient of variance s/X Volume range Â€l Mean Brix 1 10 1.00 34.74 +/ 7.66 0.22 26.62 49.86 25.57 +/ 0.76 2 10 0.16 29.29 +/ 10.16 0.35 13.85 43.55 24.75 +/ 2.27 3 10 +0.46 42.89 +/ 10.35 0.24 24.62 54.01 23.97 +/ 2.37 4 10 +0.26 54.19 +/ 14.02 0.26 26.93 81.57 25.62 +/ 1.03 5 10 +0.18 64.07 +/ 15.92 0.25 45.86 100.34 25.37 +/ 1.98 6 10 0.24 48.89 +/ 8.37 0.17 37.86 63.56 24.57 +/ 1.23 7 10 0.17 40.37 +/ 6.47 0.16 30.01 50.17 24.42 +/ 1.32 Means X=44.92 +/ 15.71 24.90 +/ 8.71 Figure 1 . Accumulated nectar volume for bagged flowers grows from day one to day 5. Accumulated nectar volume declines after day 5. There is little accumulated nectar in open flowers.
Figure 2 . Proportion of flowers that are colonized in relation to flower age. N 1 =20, N 2 = 20, N 3 =20, N 4 =21, N 5 = 6, N 6 = 4, N 7 =5 Figure 3 . Mean % sucrose measurements per day remained constant for bagged flowers and open flowers.
Figure 4 . Sugar concentration Brix remains constant as nectar volume ul rises from day 1 to day 5, and falls from day 5 to day 7. Constant sugar concentration during days 5 through day 7 indicates floral absorption of nectar.