USF Libraries

Orchid fragrance complexity as a mechanism for Euglossine bee pollinator specialization

MISSING IMAGE

Material Information

Title:
Orchid fragrance complexity as a mechanism for Euglossine bee pollinator specialization
Translated Title:
Complejidad en la fragancia de la orquídea como un mecanismo de especialización polinizadora de la abeja Euglosina ( )
Physical Description:
Book
Language:
English
Creator:
Moore, Luci
Publication Date:

Subjects

Subjects / Keywords:
Orchids--Ecology   ( lcsh )
Bees--Behavior   ( lcsh )
Costa Rica--Puntarenas--Monteverde Zone--San Luis   ( lcsh )
Orquídeas—Ecología
Abejas--Comportamiento
Costa Rica--Puntarenas--Zona de Monteverde--San Luis
Tropical Ecology Fall 2009
Ecología Tropical Otoño 2009
Genre:
Reports   ( lcsh )
Reports

Notes

Abstract:
Neotropical orchids of subtribes Stanhopeinae and Catasetinae are exclusively pollinated by euglossine bees (Euglossini). This study determines whether orchids present specific combinations and proportions of chemical odors in order to attract a limited subset of euglossine bee species, thereby avoiding hybridization that may result from the attraction of too many pollinators. Chemical baits of cineole, benzyl acetate, methyl salicylate, cineole/benzyl acetate (a Stanhopea tricornis fragrance analogue), and cineole/methyl salicylate/methyl benzoate (a Catasetum gnomus fragrance analogue) were used to attract fifteen species of euglossine bees in the San Luis Valley, Monteverde, Costa Rica. The diversity of bee species attracted to the cineole/benzyl acetate solution was significantly lower than either cineole or benzyl acetate (Shannon-Weiner index; for cineole/benzyl acetate: H’ = 0.89; for cineole: H’ = 1.91; for benzyl acetate: H’ = 1.82; for cineole/benzyl acetate and cineole: F = 4.89, df = 68.00; for cineole/benzyl acetate and benzyl acetate: F = 5.66, df = 64.29, p < 0.05). The diversity of species attracted to the cineole/methyl salicylate/methyl benzoate solution was significantly lower than cineole (ShannonWeiner index; for cineole/methyl salicylate/methyl benzoate: H’ = 0.82; for cineole: H’ = 1.91; F = 6.16, df = 61.25, p < 0.05). The results provide evidence that orchid fragrance complexes are designed to attract specific euglossine species that are both effective pollinators and specialized enough to reduce risk of interspecific outcorossing.
Abstract:
Las orquídeas neotropicales de las subtribus Stanhopeinae y Catasetinae son exclusivamente polinizadas por las abejas euglosinas (Euglosini). Este estudio determinó si las orquídeas presentan combinaciones y proporciones de olores químicos específicos con el fin de atraer un grupo limitado de abejas euglosinas, para evitar la hibridación que puede resultar de la atracción de muchos polinizadores. Sebos de cineol, bencyl acetato, metil salisilato/metil benzoato (un análogo de la fragancia de Stanohopea tricornis), y cineol/metil salicilato/metil benzoato (un análogo de la fragancia de Catasetum gnomus) fueron usadas para atraer quince abejas euglosinas en el Valle de San Luis, Monteverde, Costa Rica. La diversidad de las especies de abejas atraídas al cineol/benzyl acetato fue significativamente menor que el cineole o benzyl acetato (Indice Shannon-Weiner index; cineol/benzyl acetato H’ =0.89; for cineol H’= 1.91; for benzyl acetato H’=1.82; for cineol/ benzyl acetato and cineol F=4.89, df=68; for cineol/benzyl acetato and benzyl acetate F=5.66, df=64,29; p<0.05). La diversidad de especies atraídas al cineol/metil salicilato/metil benzoato, fue significativamente menor que el cineol (Índice de Shannon-Weiner para cineol metil salicilato/metil benzoato H’ =0.82; for cineol H’=1.91; F=6.16 df=61.25, p<0.05). Los resultados presentan evidencia que los complejos de las fragancias de las orquídeas están diseñados para atraer las especies específicas de abejas euglosinas que sean polinizadores efectivos y suficientemente especializados para reducir el riesgo de ex cruzamiento interespecífico.
Language:
Text in English.
General Note:
Born Digital

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - M39-00328
usfldc handle - m39.328
System ID:
SFS0001497:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00328
040
FHM
0 041
eng
049
FHmm
1 100
Moore, Luci
242
Complejidad en la fragancia de la orqudea como un mecanismo de especializacin polinizadora de la abeja Euglosina
245
Orchid fragrance complexity as a mechanism for Euglossine bee pollinator specialization
260
c 2009-11
500
Born Digital
3 520
Neotropical orchids of subtribes Stanhopeinae and Catasetinae are exclusively pollinated by euglossine bees (Euglossini). This study determines whether orchids present specific combinations and proportions of chemical
odors in order to attract a limited subset of euglossine bee species, thereby avoiding hybridization that may result from the attraction of too many pollinators. Chemical baits of cineole, benzyl acetate, methyl salicylate, cineole/benzyl acetate (a Stanhopea tricornis fragrance analogue), and cineole/methyl salicylate/methyl benzoate (a
Catasetum gnomus fragrance analogue) were used to attract fifteen species of euglossine bees in the San Luis Valley, Monteverde, Costa Rica. The diversity of bee species attracted to the cineole/benzyl acetate solution was significantly lower than either cineole or benzyl acetate (Shannon-Weiner index; for cineole/benzyl acetate: H =
0.89; for cineole: H = 1.91; for benzyl acetate: H = 1.82; for cineole/benzyl acetate and cineole: F = 4.89, df =
68.00; for cineole/benzyl acetate and benzyl acetate: F = 5.66, df = 64.29, p < 0.05). The diversity of species attracted to the cineole/methyl salicylate/methyl benzoate solution was significantly lower than cineole (ShannonWeiner index; for cineole/methyl salicylate/methyl benzoate: H = 0.82; for cineole: H = 1.91; F = 6.16, df = 61.25,
p < 0.05). The results provide evidence that orchid fragrance complexes are designed to attract specific euglossine species that are both effective pollinators and specialized enough to reduce risk of interspecific outcorossing.
Las orqudeas neotropicales de las subtribus Stanhopeinae y Catasetinae son exclusivamente polinizadas por las abejas euglosinas (Euglosini). Este estudio determin si las orqudeas presentan combinaciones y proporciones de olores qumicos especficos con el fin de atraer un grupo limitado de abejas euglosinas, para evitar la hibridacin que puede resultar de la atraccin de muchos polinizadores. Sebos de cineol, bencyl acetato, metil salisilato/metil benzoato (un anlogo de la fragancia de Stanohopea tricornis), y cineol/metil salicilato/metil benzoato (un anlogo de la fragancia de Catasetum gnomus) fueron usadas para atraer quince abejas euglosinas en el Valle de San Luis, Monteverde, Costa Rica. La diversidad de las especies de abejas atradas al cineol/benzyl acetato fue significativamente menor que el cineole o benzyl acetato (Indice Shannon-Weiner index; cineol/benzyl acetato H =0.89; for cineol H= 1.91; for benzyl acetato H=1.82; for cineol/ benzyl acetato and cineol F=4.89, df=68; for cineol/benzyl acetato and benzyl acetate F=5.66, df=64,29; p<0.05). La diversidad de especies atradas al cineol/metil salicilato/metil benzoato, fue significativamente menor que el cineol (ndice de Shannon-Weiner para cineol metil salicilato/metil benzoato H =0.82; for cineol H=1.91; F=6.16 df=61.25, p<0.05). Los resultados presentan evidencia que los complejos de las fragancias de las orqudeas estn diseados para atraer las especies especficas de abejas euglosinas que sean polinizadores efectivos y suficientemente especializados para reducir el riesgo de ex cruzamiento interespecfico.
546
Text in English.
650
Orchids--Ecology
Bees--Behavior
Costa Rica--Puntarenas--Monteverde Zone--San Luis
4
OrqudeasEcologa
Abejas--Comportamiento
Costa Rica--Puntarenas--Zona de Monteverde--San Luis
653
Tropical Ecology Fall 2009
Ecologa Tropical Otoo 2009
655
Reports
720
CIEE
773
t Monteverde Institute : Tropical Ecology
856
u http://digital.lib.usf.edu/?m39.328



PAGE 1

1 Orchid Fragrance Complexity as a Mechanism for Euglossine Bee Pollinator Specialization Luci Moore Department of Biology, University of Puget Sound ABSTRACT Neotropical orchids of subtribes Stanhopeinae and Catasetinae are exclusively pollinated by e uglossine bees (Euglossini). This study determines whether orchids present specific combinations and proportions of chemical odors in order to attract a limited subset of euglossine bee species, thereby avoiding hybridization that may result from the attr action of too many pollinators. Chemical baits of cineole, benzyl acetate, methyl salicylate, cineole/benzyl acetate (a Stanhopea tricornis fragrance analogue), and cineole/methyl salicylate/methyl benzoate (a Catasetum gnomus fragrance analogue) were use d to attract fifteen species of euglossine bees in the San Luis Valley, Monteverde, Costa Rica. The diversity of bee species attracted to the cineole/benzyl acetate solution was significantly lower than either cineole or benzyl acetate (Shannon Weiner ind 68.00; for cineole/benzyl acetate and benzyl acetate: F = 5.66, df = 64.29, p < 0.05). The diversity o f species attracted to the cineole/methyl salicylate/methyl benzoate solution was significantly lower than cineole (Shannon p < 0.05). Th e results provide evidence that orchid fragrance complexes are designed to attract specific euglossine species that are both effective pollinators and specialized enough to reduce risk of interspecific outcorossing. RESUMEN Las orquideas neotropicales de las subtribus Stanhopeinae y Catasetinae son exclusivamente polinizadas por abejas euglosinas (Euglosini). Este estudio determin si las orquideas presentan combinaciones y proporciones de olores qumicos especficos con el fin de atraer un grupo limitado de abejas euglosinas, para evitar hibridizacin que puede resultar de la atraccin de muchos polinizadores. Sebos de cineol, bencyl acetato, metil salisilato/metil benzoato (un anlogo de la fragancia de Stanohopea tricornis ), y cineol/metil salicilato/me til benzoato (un anlogo de la fragancia de Catasetum gnomus ) were used to attract fifteen euglossine bees in the San Luis valley, Monteverde, Costa Rica. La diversidad de especies de abejas atraidas al cineol/benzyl acetato fue significativamente menor q ue el cineole o benzyl acetato (Indice Shannon F=5.66, df=64,29 ; p<0.05). La diversidad de especies atraidas al cineol/metil salicilato/metil benzoato, fue significativamente menor que el cineol (Indice de Shannon Los resultados presentan evidencia que los complejos de fragancias de las orquideas esta diseado para atraer especies especificas de abejas euglosinas que sean polinizadores efectivos y suficientemente especializados para reducir el riesgo de exocruza miento interespecfico. INTRODUCTION Many plants depend on their ability to attract pollinators so that outcrossing may occur, thereby increasing genetic variation in the population (Dressler 1981). In the tropics, plant species are typically rare with scattered distributions, making it necessary to specialize with specific animal pollinators to ensure that pollen will effectively reach conspecifics (Schatz 1990). As vessels of

PAGE 2

2 both pollen distribution and reception, flowers are characterized by specifi c morphological and chemical traits that result from coevolution with their respective pollinators (Dressler 1993, Raguso 2008). By catering to the preferences of their pollinator organism, plants receive a means for outcrossing via the resulting pollen r emoval. In turn, pollinators use the flower rewards of nectar, pollen, etc. to maximize their own fitness, making this evolutionary relationship vital for both plant and animal (Raguso 2008). Some of the most interesting and highly evolved pollination sy ndromes are those of Family Orchidaceae. Orchids comprise the most species rich of any plant family, with 20,000 species known worldwide and 7,000 in the Neotropics (Roubik & Hanson 2004). Despite high biodiversity, orchids have particularly low biomass, scattered distributions, and inconspicuous microhabitats, necessitating some of the most highly coevolved plant pollinator relationships known (Raguso 2008). Orchid pollinators seek out conspecifics in search of nectar (the most common reward), oil, pseu dopollen, or perfume, and therefore provide a mechanism for the transfer of pollinia over long distances and varied microhabitats (Dressler 1993, Raguso 2008). The interaction between Neotropical orchids and male euglossine bees (Family Apidae; tribe Eug lossini), or orchid bees, is a well studied example of a specialized orchid pollinator relationship and is estimated to have arisen repeatedly over the last 20 million years (Engel 1999, Roubik & Hanson 2004). As a result of this long time period, male eu glossine bees have coevolved extensively with their orchid partners and currently pollinate up to ten percent of Neotropical orchid species from subtribes Catasetinae, Stanhopeinae, Lycanstinae, Zygopetalinae, and Oncidiinae (Ramirez et al 2002; Roubik & Hanson 2004). This highly evolved pollination syndrome is unique in that (1) each partner confers greater reproductive fitness to the other and (2) fragrance is the currency (and sole reward) involved in these mutualisms. Thus while a male bee is collect ing fragrances for use in mate attraction, the orchid is increasing its chances of outcrossing by transferring pollinia to the body of the bee (Dressler 1993; Roubik & Hanson 2004). Though outcrossing is the essential goal of pollination from the perspecti ve of the orchid, there is risk of hybridization via uptake of pollinia from dissimilar species (Roubik & Hanson 2004). Evolutionarily, hybridization is harmful in that it leads to homogenization of species, thereby decreasing genetic variation and leadin g to simpler systems that lower population resilience and decrease community function. Therefore, in order to coexist in the same habitat and remain separate species, orchids utilize a variety of reproductive isolating mechanisms, including temporal strat egies as well as complex and varying floral morphology (Dressler 1981). Given such mechanisms, only euglossine bee species that are of a certain size or that exhibit particular behavior can successfully transfer the pollen of specific orchids. However, s ince many species of euglossine bees may visit a variety of orchid species for fragrance collection, the chance of transgenic pollinia uptake and subsequent hybridization may be increased. To avoid hybridization, orchid species may produce different fragr ance combinations and/or proportions of combinations to further narrow the specificity of bee species, therefore using fragrance as yet another reproductive isolating mechanism (Hills et al 1972; Roubik & Hanson 2004, Raguso 2008). There is strong evidenc e to suggest that orchids produce and bees collect specific fragrance combinations. At least 60 compounds have been identified in the odors of euglossine orchids, but individual orchid species often contain fewer than 12 of these, with domination by one o r two chemical components (Gerlach & Schill 1991; Eltz et al 2005). Also, it is well documented that euglossine males display interspecific fragrance bouquet preference (Hills et al

PAGE 3

3 1972; Roubik & Hanson 2004; Ackerman et al 2005; Eltz et al 2005). H owever, past studies have shown contradictory results in testing the preference by male bees for chemical baits of single or multiple fragrances (Dodson et al 1969; Schiestl & Roubik 2003). Modifier chemicals (which are found in many orchid fragrance com plexes, but do not by themselves act as attractants) such as alpha pinene are known to reduce attractant potential, but it is not as certain whether a combination of successful attractants also alters the composition of pollinators. This study set out to determine whether orchids, specifically Stanhopea tricornis and Catasetum gnomus of Costa Rica, present specific combinations and proportions of chemical odors to attract a limited subset of euglossine bee species. These two orchid species serve as repres entatives of the subtribes Stanhopeinae and Catasetinae, the only subtribes pollinated exclusively by male euglossine bees (Williams & Whitten 1983). I predicted that more complex combinations of fragrances would attract fewer euglossine bee species as a mechanism to attract a limited set of effective pollinators, thereby increasing chances of outcrossing and reducing chances of orchid hybridization. METHODS General This study was conducted in the Puntarenas province of Costa Rica in the San Luis Valley Monteverde, Costa Rica (1000 meters above sea level). Sampling was done in primary forest from 800 to 1200 (period of greatest fragrance collection by male euglossine bees, Williams & Whitten 1983) for 15 days from October 27 to November 16, 2009. My s ampling consisted of two separate parts (1) Fragrance and (2) Visitation of species. Fragrance Combinations In order to test for specificity of fragrances and pollinator attraction, I mimicked the fragrance combination of two orchid species, Stanhopea t ricornis and Catasetum gnomus Stanhopea tricornis produces a fragrance that is one part cineole and 39 parts benzyl acetate (Dodson et al. 1969), so I mixed 0.2 mL cineole and 7.8 mL benzyl acetate to create its fragrance analogue. Catasetum gnomus prod uces a fragrance that is 59% methyl salicylate, 10% cineole, and 9% methyl benzoate (Hills et al. 1972), so I created the analogue with 1.5 mL methyl salicylate, 0.25 mL cineole, and 0.23 mL methyl benzoate. For each orchid species I conducted trials for each separate fragrance and for the specific combination of fragrances. Visitation of Species Only one fragrance or combination was tested in a day. For each trial, I saturated cotton balls with three drops of the appropriate fragrance or solution, hu ng them from shrubs, and identified the species that visited the bait. Baits were replenished with fragrance halfway through the trial period. Some euglossine bee species pairs were difficult to distinguish in the field and were recorded as either/or, inc luding Euglossa cybelia and maculilabris Euglossa deceptrix and tridentata Euglossa dodsoni and erythrochlora and Eulaema meriana and bombiformis Each bee observed was marked with a paint pen to avoid recounts. Abiotic conditions, including

PAGE 4

4 temperatu re, humidity, and general weather conditions (sunny, windy, raining, etc.) were also recorded with each observation. Statistical Analysis A template was used to calculate Shannon Weiner diversity index for the euglossine bee community visiting each chemi cal fragrance. These values were compared using modified t tests. RESULTS Statistical Results A total of 191 euglossine bees were observed across 15 species of all four genera (Table 1). The most frequent visitors to the fragrance cineole were Euglos sa imperialis and Euglossa cybelia/maculilabris and the most frequent visitors to benzyl acetate were E. imperialis Euglossa deceptrix / tridentata and Eulaema meriana Eulaema meriana was the most frequent visitor of cineole/benzyl acetate and, along wit h E. imperialis to methyl salicylate and cineole/methyl salicylate/methyl benzoate. Euglossa championi and Euglossa dodsoni/erythrochlora were both attracted to cineole and benzyl acetate when these chemicals were presented alone, but not in combination. Euglossa championi was similarly attracted to methyl salicylate alone, but not to the cineole/methyl salicylate/methyl benzoate solution (Table 1). For the test of S. tricornis orchid fragrances (cineole, benzyl acetate, and the cineole/benzyl acetate an alogue solution), the diversity of species that visited cineole alone was not different from the diversity of species that visited benzyl acetate alone (Shannon Weiner 0.54; Figure 1). Both cineole and benzyl acetate alone attracted a greater diversity of species than the cineole/benzyl acetate solution (Shannon or cineole and solution: F = 4.89, df = 68.00; for benzyl acetate and solution: F = 5.66, df = 64.29, p < 0.05). For the test of C. gnomus orchid fragrances (cineole, methyl salicylate, methyl benzoate, and the cineole.methyl salicylate/methyl benzoate a nalogue solution), the diversity of species that visited cineole alone was greater than the diversity that visited methyl salicylate alone (Shannon 64.86, p < 0.05; Figu re 1). Diversity of visitors to cineole was also greater than the diversity of visitors to the cineole/methyl salicylate/methyl benzoate solution (Shannon Weiner index; for 6.16, df = 61.25, p < 0.05). Methyl salicylate and the cineole/methyl salicylate/methyl benzoate solution did not attract different species diversities (Shannon 0.96; for cineole/methyl salicylate/methyl benzoate: For other comparisons, the diversity of species attracted to benzyl acetate alone was greater than the diversity attracted to methyl salicylate alone as well as to the cineole/methyl salicylate/methyl benzoate s olution (Shannon acetate and methyl salicylate: F = 6.07, df = 74.84; for benzyl acetate and cineole/methyl

PAGE 5

5 salicylate/methyl benzoate: F = 7.63, df = 86.30, p < 0.05; Figure 1). There was no difference between the diversity of species attracted to methyl salicylate and the cineole/benzyl acetate solution (Shannon = 0.89; F = 2.31, df = 65.05, p = 0.15). The diversity of euglossine bee visitors to the cineole/benzyl acetate solution was greater than the diversity for the cineole/methyl salicylate/methyl benzoate solution (Shannon Wei Figure 1). The range of ambient temperatures with the highest species richness was between 21 and 24 C (Figure 2) and with the highest species abundances was between 20 and 24 C (Figure 3). Additional Observations Of the 191 bees observed, ten were carrying pollinia. All bees with pollinia were E. meriana/bombiformis except for one Eufrisea surninamensis Seven b ees carried the pollinia attached to their cleypeus, which could have been from orchid genera Aspasia Notylia or Thrichophilia (Roubik & Hanson 2004) Three bees, including E. surninamensis carried the pollinia on the frontal thorax, indicating that th e visited orchids could have been Sobralia Houlettia, and/or Kegeliella genera. DISCUSSION Each orchid analogue experiment supported my original hypothesis. For the Stanhopea analogue, cineole and benzyl acetate by themselves attracted a greater divers ity of bees than when the chemicals were combined in specified proportions found in S. tricornis Eulaema meriana the known pollinator of this orchid, was the most frequent visitor to the analogue solution. Four Euglossa species also visited the cineole /benzyl acetate baits, but these are all too small to be effective pollinators for S. tricornis (Hills et al 1972). It is likely that these bees visited because the true fragrance composition in nature is much more complex and includes modifiers such as alpha and beta pinene that may normally ward off such visitors. Furthermore, given that cineole and benzyl acetate attract the same diversity of euglossine bee species, it is unlikely that fewer bees were attracted to the combined solution due to a diffe rence in concentration. For Catasetum the combination of cineole, methyl salicylate, and methyl benzoate found in C. gnomus attracted a lower diversity of bees than cineole alone. As with the Stanhopea analogue, the most frequent visitor was E. merian a Euglossa imperialis was also a relatively frequent visitor. What is surprising is that the diversity of bees attracted to methyl salicylate was not statistically different from the cineole/methyl salicylate/methyl benzoate solution. This could be due to the frequency of methyl salicylate in nature. Fourteen of the 39 compounds produced by Catasetum orchids have been identified with gas liquid chromatography. Of the 22 Catasetum species analyzed by Hills et al. (1972), cineole is present in 64 percen t of the species compared to methyl salicylate, which is only present in 14 percent of the species. Cineole is also the dominant component in 14 percent of the species while methyl salicylate is the dominant fragrance in 9 percent. Furthermore, in a meta analysis on orchid floral fragrances as euglossine bee attractants done by Williams and Whitten (1983), 60 chemicals were identified and rated in attractiveness and abundance. Cineole was marked as a good (as opposed to moderate)

PAGE 6

6 attractant, is known to be present in a relatively large number of orchid species, and makes up 21 to 90 percent of the fragrance composition of most species in which it is present. Therefore cineole is more commonly found in nature than methyl salicylate and acts as a stronger attractant. It is reasonable to conclude that the inclusion of methyl salicylate in a floral presentation already acts to attract a limited subset of euglossine species in comparison to cineole. Further studies should be done to test whether the assumpti on that frequency in nature and attractive ability are indeed positively correlated. For the chemical baits presented, simple and complex alike, E. meriana and E. imperialis made up a large portion of the bee visitors. It is possible that the orchid fra grances are not only selectively attracting species, but that these bees species are simply more generalist. For E. meriana individuals with pollinia, the placement on the body suggests that these individuals received the pollen from at least two and at m ost six different orchid genera, none of which are Catasetum or Stanhopea Furthermore, it is not unusual for certain euglossine species to be attracted to a wide variety of odors. In nature, euglossine males have been observed collecting odors from a di versity of locations, including rotting fruit and wood, tree bark, and exposed roots (Ackerman 1989). Furthermore, researchers often successfully attract euglossines using chemicals that have not been found in natural orchid fragrance complexes ( e.g. van illin, skatole, p crestyl acetate) (Williams and Whitten 1983). One study even reports technical grade DDT used for malaria control as an attractant for male Eufriesea purpurata in Amazonian Brazil (Roberts et al. 1982). Along with possibly being more ge neralist in fragrance preference, E. meriana and E. imperialis bees are larger in body size and may be able to better withstand abiotic conditions of rain, wind, or temperature. It would be interesting to test this by controlling for scent preference and observing whether fewer of the small Euglossa bees are attracted in harsher weather conditions. Both E. championi and E. dodsoni/erythrochlora were attracted to cineole and benzyl acetate separately, but not to the combined solution. Euglossa championi wa s also attracted to methyl salicylate, but not to the cineole/methyl salicylate/methyl benzoate solution. This is not necessarily a rare occurrence. In a previous study, Williams and Calaway (1972) found that E. meriana is attracted to cineole and methyl salicylate separately, but is attracted to the combination in very low numbers. Further study is needed to determine by what mechanism a combination of fragrances reduces attractiveness. Currently, the most convincing hypothesis is that reduction in attr action is due to a synergistic inhibition of receptor responses at the peripheral and central nervous system (Williams & Calaway 1972; Schiestl & Roubik 2003). This could occur via binding interactions, where each chemical inhibits the other from binding to a receptor (Schiestl & Roubik 2003). The fact that two attractive chemicals become unappealing to certain bee species when combined demonstrates the importance of chemical blends and proportions in orchid fragrances. This study provides support for the proposal that euglossine orchids produce specific recipes and proportions of chemical odors to (1) increase chances of successful pollen dispersal and thus outcrossing and (2) serve as a reproductive isolating mechanism to prevent hybridization. A fragra nce complex is designed to attract a limited subset of euglossines that are both effective as pollinators for that species and specialized enough that visits would not increase risk of hybridization. Further studies should be done in testing the variabili ty of fragrance themselves act as general attractants. However, modifier chemicals such as alpha pinene may be more important to specifying the attractive ability of an orchid. The identification and

PAGE 7

7 classification of orchid fragrance composition is also lacking. However, once chemicals are identified, it would be interesting to investigate the stereochemistry and test whether different isomers of chiral compounds attra ct different species. Finally, it would also be useful to test intraspecific variability of fragrance composition in orchids in relation to attraction of the optimal pollinator species set. ACKNOWLEDGEMENTS I would like to thank Anjali Kumar for her indispensable help and positive attitude as my advisor. I also thank Alan Masters and Pablo Allen for advice and brainstorming, Yimen Araya for help with statistical analysis, Jos Carlos Caldern for his help with materials and random needs at Estacin B iolgica, and Frank Joyce for use of his fragrances. Resources were provided by CIEE. LITERATURE CITED A CKERMAN J. D. 1989. Geographic and seasonal variation in fragrance choices and preferences of male euglossine bees. Biotropica 21: 340 347. D OD SON C.H., R.L. D RESSLER H.G. H ILLS R.M. A DAMS AND N.H. W ILLIAMS 1969. Biologically active compounds in orchid fragrances. Science 164: 1243 1249. D RESSLER R .L. 1993. Field Guide to the Orchids of Costa Rica and Panama. Comstock Publishing Assoc iates, Ithaca, pp. 26 28. 1981. The Orchids: Natural History and Classification. Harvard University Press, Cambridge, pp. 121 134. E LTZ T., D.W. R OUBIK AND K. L UNAU 2005. Experience dependent choices ensure species specific fragra nce accumulation in male orchid bees. Behav. Ecol. Sociobiol. 59: 149 156. E NGEL M. S. 1999. The first fossil Euglossa and phylogeny of the orchid bees (Hymenoptera: Apidae; Euglossini). American Museum Novitates 3274: 1 14. G ERLACH G. AND R. S CHILL 1991. Composition of orchid scents attracting euglossine bees. Bot. Acta. 104: 379 391. H ILLS H. G., N.H. W ILLIAMS AND C.H. D ODSON 1972. Floral fragrance and isolating mechanisms in the genus Catasetum (Orchidaceae). Biotropica 4: 61 76. R AGUSO R.A 2008. Wake up and smell the roses: the ecology and evolution of floral scents. Annu. Rev. Ecol. Syst. 39: 549 569. R AMIREZ S., R.L. D RESSLER AND M. O SPINA 2002. Euglossine bees (Hymenoptera: Apidae) from the neotropical region: a species ch ecklist with notes on their biology. Biota Colombiana 3: 7 118. R OBERTS D.R., W.D. A LECRIM J.M. H ELLER S.R. E HRHARDT AND J.B. L IMA 1982. Male Eufrisia purpurata a DDT collecting euglossine bee in Brazil. Nature 297: 62 63. R OUBIK D.W. AND P.E. H ANSON 2004. Orchid bees of tropical America: Biology and Field Guide. Instituto Nacional de Biodiversidad, Costa Rica. S CHATZ G.E 1990. Some aspects of pollination biology in Central American forests. In: Reproductive Ecology of Tropical Forest Plants K.S. Bawa and M. Hadley, ed. UNESCO/Parthenon, Paris, pp. 69 84. S CHIESTL F.P. AND D.W. R OUBIK 2003. Odor compound detection in male euglossine bees.Journal of Chemical Ecology 29: 253 258. W ILLIAMS N.H. AND H.D. C ALAWAY 1972. Selective attraction of male euglossine bees to orchid floral fragrances and its importance in long distance pollen flow. Evolution 26: 84 95. AND W.M. W HITTEN 1983. Orchid floral fragrances and male euglossine bees: methods and advances in the last sesquidecade. Biology Bulletin 164: 355 395.

PAGE 8

8 TABLES AND FIGURES TABLE 1. Abundances of male euglossine bee species visiting each chemical bait in the San Luis Valley, Monteverde, Costa Rica. a a b b, c c FIGURE 1. Total euglossine bee species richness attracted to each chemical fragrance. SE bars are shown. The letters above the bars denote significant difference in Shannon Weiner index of the chemical. A t otal of 191 bees were observed, comprising 15 different species.

PAGE 9

9 FIGURE 2. Euglossine bee species richness from 19.3 C to 26.6 C. Trials were run from 8 AM to 12 from October 27 th to November 16 th in the San Luis Valley, Costa Rica. FIGURE 3. Euglossine bee abundance from 19.3 C to 26.6 C. Trials were run from 8 AM to 12 from October 27 th to November 16 th in the San Luis Valley, Costa Rica.