USF Libraries

Impact of reward on floral mimics Lantana camara (Verbenaceae) and Epidendrum radicans (Orchidaceae)

MISSING IMAGE

Material Information

Title:
Impact of reward on floral mimics Lantana camara (Verbenaceae) and Epidendrum radicans (Orchidaceae)
Translated Title:
Impacto de recompensa en los mimetismos florales Lantana cámara (Verbenaceae) y Epidendrum radicans (Orchidaceae) ( )
Physical Description:
Book
Language:
English
Creator:
Zwickey, Kara
Publication Date:

Subjects

Subjects / Keywords:
Orchids--Mimicry (Biology)--Costa Rica--Puntarenas--Canitas   ( lcsh )
Pollination by insects   ( lcsh )
Butterflies--Feeds and feeding--Costa Rica   ( lcsh )
Orquideas--mimetismo(Biología)--Costa Rica--Puntarenas--Cañitas
Polinización por insectos
Tropical Ecology 2008
Species fitness
Ecología Tropical 2008
Aptitud de especies
Genre:
Reports   ( lcsh )
Reports

Notes

Abstract:
The fitness of a non-rewarding plant species can be increased by the addition of another plant species with similar flowers and reward present if pollinators are unable to distinguish between the two species (Bierzychudek 1981). Floral mimicry occurs between species when three ecological conditions are met: the model and mimic species must coexist within flight range of an individual pollinator, plant species must share pollinator species, and pollinator individuals visiting one plant must also visit one or both of the other plant species (Bierzychudek 1981). Non-rewarding Epidendrum radicans (Orchidaceae) has been shown to be a Batesian mimic of rewarding Lantana camara (Verbenaceae). Although mimicry increases the fitness of Batesian species when plants are in close proximity, mixed patches are widely spaced, and patch sizes are small (Deacon 2000; Woo 2001), there is little known about floral reward’s affect on the relationship. This study uses L. camara and E. radicans to test the impact of increased reward availability in model plants (L. camara) on pollinator visitation rates to both model and mimic plants. Two observation plots were constructed containing L. camara and E. radicans, where L. camara flowers in one plot were injected with a 20% sucrose solution. Plots were observed for butterfly visitation rates and monitored for pollinia removal from E. radicans. Increased reward in L. camara resulted in fewer visits to E. radicans flowers and L. camara inflorescences within the treated patch (three-way ANOVA: F = 24.0506; df = 1; P < .0001). In addition, L. camara inflorescences were visited longer with increased reward present (three-way ANOVA: F = 5.88; df = 1; P < .05). However, no difference was observed between pollinia removal between patches (Chi-square = 3; df = 1; P > .05). Different butterfly species had varying rates of visitation to the mimic species in response to an increased reward. Danaus plexippus visited 0.75 fewer E. radicans flowers, Leptophobia aripa visited 0.184 more flowers, and Anartia fatima visited 0.039 fewer E. radicans flowers (two-way ANOVA: F = 3.7887; df = 2; P < .05). As a result of increased reward, model and mimic species are negatively affected, due to a decrease in pollinator visitation rates. Prior research suggested rewarding flowers benefit most from being in close proximity to one another, in clumped patches (Deacon 2000). The results of this study suggest that L. camara may have been selected to produce a reward that favors a large number of visits to many flowers, which benefits the mimic as well. Although mimic flowers were visited, there was no affect on the pollinia removal from E. radicans, suggesting that patch size and density ultimately affect pollination rates of deceptive plants.
Abstract:
El éxito reproductivo de una planta que no ofrece ninguna recompensa al agregarle una planta con flores similares y esta recompensa al presente si los polinizadores son incapaces de distinguir entre las dos especies. La planta que no ofrece recompensa Epidendrum radicans (Orchidaceae) es mimética de la especie Lantana camara (Verbenaceae), que si ofrece una recompensa. Este estudio utiliza L. camara y E. radicans para demostrar el impacto de un aumento en la disponibilidad de una recompensa en las plantas modelo (L. camara), en proporciones de visitas de polinizadores tanto a la planta modelo como a la mimética.
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 - M39-00048
usfldc handle - m39.48
System ID:
SFS0001224:00001


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 M39-00048
040
FHM
0 041
eng
049
FHmm
1 100
Zwickey, Kara
242
Impacto de recompensa en los mimetismos florales Lantana cmara (Verbenaceae) y Epidendrum radicans (Orchidaceae)
245
Impact of reward on floral mimics Lantana camara (Verbenaceae) and Epidendrum radicans (Orchidaceae)
260
c 2008-11
500
Born Digital
3 520
The fitness of a non-rewarding plant species can be increased by the addition of another plant species with similar flowers and reward present if pollinators are unable to distinguish between the two species (Bierzychudek 1981). Floral mimicry occurs between species when three ecological conditions are met: the model and mimic species must coexist within flight range of an individual pollinator, plant species must share pollinator species, and pollinator individuals visiting one plant must also visit one or both of the other plant species (Bierzychudek 1981). Non-rewarding Epidendrum radicans (Orchidaceae) has been shown to be a Batesian mimic of rewarding Lantana camara (Verbenaceae). Although mimicry increases the fitness of Batesian species when plants are in close proximity, mixed patches are widely spaced, and patch sizes are small (Deacon 2000; Woo 2001), there is little known about floral rewards affect on the relationship. This study uses L. camara and E. radicans to test the impact of increased reward availability in model plants (L. camara) on pollinator visitation rates to both model and mimic plants. Two observation plots were constructed containing L. camara and E. radicans, where L. camara flowers in one plot were injected with a 20% sucrose solution. Plots were observed for butterfly visitation rates and monitored for pollinia removal from E. radicans. Increased reward in L. camara resulted in fewer visits to E. radicans flowers and L. camara inflorescences within the treated patch (three-way ANOVA: F = 24.0506; df = 1; P < .0001). In addition, L. camara inflorescences were visited longer with increased reward present (three-way ANOVA: F = 5.88; df = 1; P < .05). However, no difference was observed between pollinia removal between patches (Chi-square = 3; df = 1; P > .05). Different butterfly species had varying rates of visitation to the mimic species in response to an increased reward. Danaus plexippus visited 0.75 fewer E. radicans flowers, Leptophobia aripa visited 0.184 more flowers, and Anartia fatima visited 0.039 fewer E. radicans flowers (two-way ANOVA: F = 3.7887; df = 2; P < .05). As a result of increased reward, model and mimic species are negatively affected, due to a decrease in pollinator visitation rates. Prior research suggested rewarding flowers benefit most from being in close proximity to one another, in clumped patches (Deacon 2000). The results of this study suggest that L. camara may have been selected to produce a reward that favors a large number of visits to many flowers, which benefits the mimic as well. Although mimic flowers were visited, there was no affect on the pollinia removal from E. radicans, suggesting that patch size and density ultimately affect pollination rates of deceptive plants.
El xito reproductivo de una planta que no ofrece ninguna recompensa al agregarle una planta con flores similares y esta recompensa al presente si los polinizadores son incapaces de distinguir entre las dos especies. La planta que no ofrece recompensa Epidendrum radicans (Orchidaceae) es mimtica de la especie Lantana camara (Verbenaceae), que si ofrece una recompensa. Este estudio utiliza L. camara y E. radicans para demostrar el impacto de un aumento en la disponibilidad de una recompensa en las plantas modelo (L. camara), en proporciones de visitas de polinizadores tanto a la planta modelo como a la mimtica.
546
Text in English.
650
Orchids--Mimicry (Biology)--Costa Rica--Puntarenas--Canitas
Pollination by insects
Butterflies--Feeds and feeding--Costa Rica
4
Orquideas--mimetismo(Biologa)--Costa Rica--Puntarenas--Caitas
Polinizacin por insectos
653
Tropical Ecology 2008
Species fitness
Ecologa Tropical 2008
Aptitud de especies
655
Reports
720
CIEE
773
t Monteverde Institute : Tropical Ecology
856
u http://digital.lib.usf.edu/?m39.48



PAGE 1

Impact of Reward on Floral Mimics Lantana camara (Verbenaceae) and Epidendrum radicans (Orchidaceae) Kara Zwickey Department of Environmental Science, University of Oregon ABSTRACT The fitness of a non-rewarding plant species can be increased by the addition of another plant species with similar flowers and reward present if pollinators are unabl e to distinguish between the two species (Bierzychu dek 1981). Floral mimicry occurs between species when three ec ological conditions are met: the model and mimic sp ecies must coexist within flight range of an individual pollin ator, plant species must share pollinator species, and pollinator individuals visiting one plant must also visit one or both of the other plant species (Bierzychudek 19 81). Nonrewarding Epidendrum radicans (Orchidaceae) has been shown to be a Batesian mimi c of rewarding Lantana camara (Verbenaceae). Although mimicry increases the fit ness of Batesian species when plants are in close proximity, mixed patches are widely spaced, and pat ch sizes are small (Deacon 2000; Woo 2001), there i s little known about floral reward’s affect on the relations hip. This study uses L. camara and E. radicans to test the impact of increased reward availability in model plants ( L. camara ) on pollinator visitation rates to both model and mimic plants. Two observation plots were constructed con taining L. camara and E. radicans where L. camara flowers in one plot were injected with a 20% sucrose solution. Plots were observed for butterfly visitation rate s and monitored for pollinia removal from E. radicans Increased reward in L. camara resulted in fewer visits to E. radicans flowers and L. camara inflorescences within the treated patch (three-way ANOVA: F = 24.0506; df = 1; P < .0001). In addition, L. camara inflorescences were visited longer with increased reward present (three-way ANOVA: F = 5.88; df = 1; P < .05). However, no difference was obser ved between pollinia removal between patches (Chi-s quare = 3; df = 1; P > .05). Different butterfly species had varying rates of visitation to the mimic species in response to an increased reward. Danaus plexippus visited 0.75 fewer E. radicans flowers, Leptophobia aripa visited 0.184 more flowers, and Anartia fatima visited 0.039 fewer E. radicans flowers (two-way ANOVA: F = 3.7887; df = 2; P < .05). As a result of increased reward, model and m imic species are negatively affected, due to a decr ease in pollinator visitation rates. Prior research sugges ted rewarding flowers benefit most from being in cl ose proximity to one another, in clumped patches (Deacon 2000). The results of this study suggest that L. camara may have been selected to produce a reward that favors a large nu mber of visits to many flowers, which benefits the mimic as well. Although mimic flowers were visited, there was no a ffect on the pollinia removal from E. radicans suggesting that patch size and density ultimately affect pollinatio n rates of deceptive plants. RESUMEN El exito reproductivo de una planta que no ofrece n inguna recompensa al agregar una planta con flores similares que ofrezca una recompensa si el polinizador no es capa z de diferenciarlas. La planta que no ofrece recomp ensa Epidendrum radicans (Orchidaceae) es mimtica de la especie Lantana cmara (Verbenaceae), que si ofrece recompensa. Este estudio utiliza L. camara y E. radicans para demostrar el impacto de un aumento en la disponibilidad de recompensa en plantas modelo ( L. camara ), en proporciones de visitas de polinizadores tant o a la planta modelo como a la mimtica. Dos parcelas de observacin fueron construidas conteniendo ambas es pecies de plantas, en donde las flores de L. camara en una parcela fueron inyectadas con una solucin a l 20% de sacarosa. Se determinaron las proporciones de visitas por parte de mariposas y la remocin de polinia en E. radicans Aumento en el factor de recompensa en L. camara resulta en menos visitas a flores de E. radicans e inflorescencias de L. camara dentro del parche tratado (ANOVA: F = 24.0506; df = 1; P < .0001). En adicin las inflorescencias de L. camara fueron visitadas por periodos ms largos cuando se aument la recompensa (ANOVA: F = 5.88; df = 1; P < .05). Sin embargo no se observ ninguna diferencia en la remocin de polinia entre parches (X = 3; d f = 1; P >

PAGE 2

.05). Diferentes especies de mariposa presentan di ferentes proporciones de visitacin a la especie mi mtica en respuesta al aumento en la recompensa; Danaus plexippus visita 0.75 menos E. radicans flores, Leptophobia aripa visita 0.184 ms flores, y Anartia fatima visita 0.039 menos E. radicans flores (ANOVA: F = 3.7887; df = 2; P < .05). Como resultado del aumento de la recompensa, las especies modelo y mimticas se ven afectadas negativamente, debido al aumento en la proporcin d e visitas por parte de los polinizadores. Resultad os posteriores sugieren un beneficio al aumentar la recompensa flo ral al estar cerca unos de otros, en parches agrega dos (Deacon 2000). Los resultados de este estudio sugieren que L. camara puede estar seleccionada a producir recompensa que favorece un gran nmero de visitantes a varias flor es, lo cual beneficia al mimtico tambin. Aunque flores mimticas fueron visitadas, no hay un efecto en la remocin de polinia de E radicans sugiriendo que el el tamao y la densidad del parche afectan finalmente las tasas de polinizacin de plantas engaosas. INTRODUCTION Plant-pollinator associations are not always mutual istic. Plants can attract pollinators using deceptive signals without rewarding them with necta r or pollen (Schluter 2008). Floral mimicry occurs under specific conditions between plants wit h larger inflorescence size, within small patches, and when model and mimic are in close prox imity to one another (Deacon 2000; Dupre 2004; Woo 2001). However, Bierzychudek (1981) and Deacon (2000) suggest mimicry does not occur when plants are within dense patches, because pollinators are able to detect the unrewarding nature of the stand and leave before ma ny flowers are pollinated. Additionally, mimicry is dependent on the foraging habits and pre sence of pollinators, with some flowering plants relying on deception or the lack of a collec table or consumable substance (Ackerman et al. 1994). Rewards may attract visitors and maintain h igh levels of pollination, but the cost of the reward production may outweigh the benefits, as nec tar production can be expensive (Ackerman et al. 1994). A non-rewarding species that mimics a rewarding mod el species and obtains a one-sided advantage through imitation is said to be a Batesia n mimic (Dafni 1984; Roy and Widmer 1999). Batesian mimics do not reward pollinators and rely on morphological similarities including floral coloration and other shared attractive floral signa ls of the model species to successfully attract pollinators (Craig and Johnson 2008; Jersakova 2006 ; Roy and Widmer 1999; Schluter 2008). Floral mimicry is highly conditional, requiring spe cific ecological conditions necessary for mimicry to occur between a mimic and model species. These conditions include shared phenologies such as flowering time and color simila rities, shared pollinators, and overlapping geographical ranges (Bierzychudek 1981; Schluter 20 08). Lantana camara (Verbenaceae) and Epidendrum radicans (Orchidaceae) are common weedy plants of Costa Rica. They have overlapping geographic ranges and prefer highly disturbed regions along roadsides or in pastures fr om 0-2000 meters in elevation (Gargiullo 2008; Schemske 1983; Todiza 1983). Further, they s hare similar floral phenologies in that both flower year round with red-orange and yellow inflor escence. These floral species are primarily pollinated by butterfly species, Anartia fatima and Danaus spp (Schmeske 1983; Todzia 1983; Wolfe 1987). Studies suggest that the deceptive or chid E. radicans achieves higher pollinia removal through resemblance of rewarding Mllerian mimic species L. camara and Asclepias curassavica (Apocynaceae), and is a Batesian mimic (Deacon 2001 ; Dupre 2004; Jersakova 2006; Todzia 1983). Studies of Batesian mimicry suggest deceptive plant species benefit in close proximity to rewarding plants because pollinators are more abund ant near rewarding plants and are more likely to leave a plot when non-rewarding deceptive flowers are visited first within patches (Ackerman et al. 1994; Craig and Johnson 2008). Ho wever, it remains unclear whether

PAGE 3

increased visits by pollinators to mimic species wi thin patches results from increased availability of reward. This study returns to the Batesian flor al mimicry complex to test the impact of increased reward availability in model plants ( L. camara ) on pollinator visitation rates to model and mimic plants. Based on earlier studies by Acke rman et al. (1994) and Craig and Johnson (2008), I predict that increasing reward in model p lants will result in a greater pollinator abundance to mimicry plants, therefore increasing v isitation rates to adjacent mimic species and increasing the potential for pollination to occur i n mimic plants. Batesian mimicry has both costs and benefits. While not producing a reward is more cost efficient for the plant, Batesian plants are dependent on the density and patch composition of model plants to attract pollinators (Ackerman et al. 1994). MATERIALS AND METHODS Study Site This study took place from October 29, 2008 to Nove mber 13, 2008 on an inclined embankment along the roadside near the Miramontes Hotel in Ca itas, Costa Rica. Two sites, A and B, each composed of 11 L. camara inflorescences, containing 10-40 individual flower s per inflorescence, and eight E. radicans flowers, with one or two flowers per inflorescence were observed. E. radicans species were interspersed among the native L. camara plants on each site, with plants positioned so flowers were at similar heights. E. radicans were moved from a site in San Luis to Caitas. The entire plant was uprooted and placed directly within the site. Because E. radicans has aerial roots, transplanted individuals did not require extra care (Gargiullo 2008). L. camara flowers in site B were injected with a 20% sucrose solution, using a 3 mL syringe, to increase the reward volume within the plot site at the beginning of each observation period. Observations Sites A and B were observed for the number and spec ies of butterfly pollinators, flower species visited, and length of stay per flower to determine pollinator visitation rates. E. radicans flowers’ pollen is packaged into pollinia, which at taches onto the base of the butterfly’s proboscis when the flower is pollinated (Bierzychud ek 1981). To determine the rate of visitation for E. radicans flowers, pollinia removal was also monitored withi n plot A and B. The presence or absence of pollinia was recorded at the beginnin g and end of each observation period. E. radicans flowers with pollinia removed were removed from th e site and replaced with a new flower containing pollinia at the beginning of each observation period. Observation periods were roughly 30-90 minutes long. Sites A and B were loc ated approximately 60 meters apart and observed between the hours of 8:30 am and 1:00 pm f or a total of 1000 minutes. Relevant Natural History L. camara inflorescences were observed in this study as a si ngle compound flower. Because 67% of inflorescences are contiguous or almost comp osite (Schemske 1976; Schemske 1983), consisting of several small flowers, many butterfly pollinators do not treat flower probing as an individual event. Rather, pollinators probe the ma jority of the flowers in an inflorescence at each feeding period. One L. camara inflorescence (21.8 mm +/0.9) is approximately e qual to one E.

PAGE 4

radicans flower (20mm) in diameter, resulting in equivalent floral displays per individual unit (Gargiullo 2008; Schemske 1976). If flowers of L. camara were observed for individual probing incidents by butterfly pollinators, differences in expected results would have been greater. L. camara has axillary inflorescence that are broadly rounde d and composed of yellow and orange-red flowers (Schemske 1983). Infloresce nce diameter of L. camara is 21.8 mm +/0.9 (Schemske 1976). L. camara inflorescences in the study contain10-40 flowers p er inflorescence. Similarly, E. radicans are comprised of yellow/orange and red inflorescen ce, with a flower diameter of approximately 20 mm (Gargiullo 2008; Todzia 1983). RESULTS Considering all visiting butterfly species, more ne ctar did not change visitation rates between patches. Nearly equal numbers of butterflies visit ed both patches, with 235 butterflies observed in patch A and 222 butterflies observed in patch B (Fig 1: Chi-square = 0.37; df = 1; P < .001). Although Plot A and B were visited equally by butte rflies, pollinators did show a preference between model and mimic species between plots. In general, inflorescences of L. camara were visited more frequently than flowers of E. radicans 182 flowers of E. radicans were visited by butterflies in patch A, and 102 flowers were visite d in patch B with increased reward. Additionally, 64 more inflorescences of L. camara were visited in patch A than in patch B (Fig 2: Chi-square = 6.49; df = 1; P < .05; Chi-square = 26.04; df = 1; P < .001). Consequently, increasing reward availability within a plot negati vely affected model and mimic floral species, as both were visited fewer times within the treated plot, reducing the possibility of pollination to occur. Although pollinators visited both types of plants m ore in patch A, butterflies on average spent longer periods of time probing plants in patc h B. The explanation of fewer visits to inflorescences and flowers with added nectar is dep icted in Figure 3, by average time spent per individual within plots. When nectar was added to patch B, the time spent on inflorescences dramatically increased. Inflorescences of L. camara were visited on average 33.072 +/SE seconds longer in patch B with additional reward pr esent, while flowers of E. radicans were observed to be probed for less time in the treated patch B, by 0.329 +/SE seconds (Fig 3; threeway ANOVA: F = 24.0506; df = 1; P < .0001). In plo t B, butterflies were observed probing L. camara inflorescences for longer periods of time, and visi ted fewer flowers. This suggests that fewer inflorescences were required by butterflies t o be full. Although three species of butterflies of varying si zes were commonly observed visiting patches, Anartia fatima (Nymphalidae: Nymphalinae), Danaus plexippus (Nymphalidae: Danainae), and Leptophobia aripa (Pieridae), there was no significant difference be tween the average amount of time spent within the patches or on individual plant species by butterflies (Fig 3; three-way ANOVA: F = 2.8492; df = 1; P > .05; th ree-way ANOVA: F = 0.9127; df = 2; P > .05). D. plexippus was observed spending on average the longest time within plot A and B, 37.471 +/SE, and 86.759 +/SE, respectively (Fig 6). Overall, fewer visits to E. radicans flowers resulted in a decreasing trend in the numb er E. radicans pollinia removed from patched with nectar added to L. camara flowers (B)(Fig 4: Chi-square = 3; df = 1; P > .05). Patch A, without additional nectar had 52 pollinia removed, while patch B had 35 pollinia removed from flowers (Fig 4). When examining the most common butterfly visitors o bserved within patches A and B by species, A. fatima, D. plexippus and L. aripa each responded differently to the increasing

PAGE 5

reward availability of L. camara in patch B. Butterfly species A. fatima visited 178 flowers and inflorescences in patch A and 195 individuals in pa tch B. Butterfly species L. aripa visited 45 flowers and inflorescences in patch A and 3 individ uals in patch B. Lastly, D. plexippus visited 12 individuals in patch A and 18 flowers and inflor escences in patch B (Fig 5: Chi-square = 37.992; df = 2; P < .0001). It appears the D. plexippus is the major species responsible for pollinia removal from E. radicans flowers, as is visited the most E. radicans flowers per individual per butterfly species (Fig 7). Addition ally, D. plexippus was the only species observed visiting more E. radicans flowers when the reward to L. camara flowers was not enhanced in patch A (Fig 7). Considering only E. radicans flowers, which accounted for 33.3% of total visits butterfly species differed in the number of flowers visited p er individual (Fig 7; two-way ANOVA: F = 2.8188; df = 5; P < .05). Trends suggest that D. plexippus responded to increased nectar by visiting slightly fewer flowers, 0.75, while L. aripa responded to increases in nectar by visiting slightly more flowers, 0.184. Lastly, little chan ge was observed between the number of flowers A. fatima visited, 0.039 fewer flowers visited (Fig 7; two-w ay ANOVA: F = 3.7887; df = 2; P < .05; two-way ANOVA: F = 2.8791; df = 2; P > .05). Again, only considering L. camara inflorescences, pollinators were observed visiting similar numbers of inflorescences within patches. A. fatima and D. plexippus were observed visiting on average more L. camara inflorescences in patch B, 0.372 and 0.143 more inflorescences respectively. However, L. aripa was observed visiting on average 0.373 fewer inflorescences in patch B. Trends were not observe d between butterfly species and number of inflorescences visited in response to increased rew ard (Fig 8; two-way ANOVA: F = 4.9485; df = 5; P < .001; two-way ANOVA: F = 0.0552; df = 1; P > .05; two-way ANOVA: F = 2.267; df = 2; P > .05; two-way ANOVA: F = 1.6144; df = 2; P > .05). DISCUSSION Previous floral mimicry studies suggest that mimicr y occurs under very specific and controlled conditions (Bierzychudek 1981). Dupre (2004) obser ved a floral mimicry complex occurring between Lantana camara Epidendrum radicans and Asclepias curassavica in which patch sizes and densities were controlled. The data is t his study also suggests that mimicry is occurring between Lantana camara and Epidendrum radicans when plants were combined in small patches, as butterflies visited both model an d mimic plants. Unlike the theory presented by Ackerman et al. (1994) in which pollinators are mor e likely to leave the patch after visiting a non-rewarding flower, pollinators were observed vis iting multiple flowers and inflorescences within a patch after visiting the non-rewarding mim ic. However, pollinator species prefer the model species containing a reward to the non-reward ing mimic species. In response to increased nectar availability in inf lorescences of L. camara in patch B, butterflies visited significantly fewer E. radicans flowers and L. camara inflorescences. This was due to increases visitation length to L. camara inflorescences, reducing foraging activity within patch B. Consequently, increasing nectar wi thin model plants negatively affected visitation rates to model and mimic species, reduci ng the possibility of pollination to occur. Although no significant differences were observed i n pollinia removal between the normal reward patch A and increased reward patch B, trends were observed in which pollinia removal from E. radicans flowers decreased with increase reward.

PAGE 6

Additionally, trends in butterfly preferences were observed for mimic species. D. plexippus appears to be the major species responsible for pol linia removal from E. radicans flowers, as it visited the most flowers per individ ual per butterfly species in both patch types. D. plexippus was the only species observed visiting more E. radicans flowers when the reward to L. camara flowers was at normal levels in patch A. Because Batesian species do not produce a reward, t hus relying on models to aid in the attraction of pollinators (Craig and Johnson 2008; Jersakova 2006; Roy and Widmer 1999; Schluter 2008), Batesian species are dependent on t he phenology and attractive characteristics of their models. Therefore, Batesian species pollinat ion success is also dependent on the model species. If model flowers produce high floral dens ities within a patch and consequently increase the nectar content within the patch, the mimic is a lso impacted. As observed in this data, mimics are negatively impacted due to increased nectar con centration within patches. Despite the benefits associated with floral mimicry, including less energy invested into reward production, mimics are highly vulnerable, as they have no contr ol over reward production or pollinator attraction (Ackerman et al. 1994). In this study, both L. camara and E. radicans benefited from the normal concentration of reward offered by L. camara the model species. This suggests that selection of the model for optimal pollination based on nectar levels offered to pollinators, also evolutionarily assured optimal pollination to the mimic species. However, only one set of mimics was observed within a controlled environment. Therefore, no conclusion s can be formed about reward and pollinator interaction for all floral mimics. To further investigate the study of floral mimicry and the impact of floral reward on pollination, further studies should concentrate on observing more paired patches, with normal reward and with increased reward, as well observing individual flower probing on L. camara Because the Batesian mimic, E. radicans is widely accepted as part of a three-plant mimicr y complex with Mllerian mimics L. camara and A. curassavica it would be interesting to repeat the studying including model species A. curassavica Lastly, observing the impact of floral abundance and density on mimic pollination rate wou ld further the study in the field of mimicry. ACKNOWLEDGMENTS This study would not have been possible without the permission of “Licho” Argedas who kindly allowed m e to conduct research in his garden. Additionally I wou ld like to thank the owners of the horse ranch near the Miramontes Hotel for allowing me to observer flower s on their land. I especially would like to thank Pablo Allen for all of his patience and assistance in answering my numerous statistical questions and helping me i dentify butterflies. Without his help, my results section would have been non-existent. I would also like to thank Moncho Caldern for translating my abstract. Lastly, than k you to my advisor, Allen Masters for guiding me i n my research. LITERATURE CITED Ackerman, J.D., J.A. Rodriguez, and E.J. Melendez. 1994. A Meager Nectar Offering By an Epiphytic Orch id is Better then Nothing. Biotropica 26(1): 44-49. Bierzychudek, P. 1981. Asclepias, Lantana, and Epidendrum: A Floral Mimicry Complex? Biotropica 13: 54-58. Craig, P.I. and S.D. Johnson. 2008. Mimics and Magn ets: The Importance of Color and Ecological Facilit ation in Floral Deception. Ecology 89(6): 1583-1595.

PAGE 7

Deacon, Nicholas. Pollinia Removal and Visitation i n Epidendrum radicans (Orchidaceae) and Asclepias Currassavica (Asclepiadaceae). CIEE Program. Fall 2000. Dafni, A. 1984. Mimicry and Deception in Pollinatio n. Annual Review of Ecology and Systematics. 15: 259-278. Dupre, Sarah. Evidence for Floral Mimicry in Epidendrum radicans (Orchidaceae) with Asclepias curassavica (Apocynaceae) and Lantana camara (Verbenaceae). CIEE Program. Fall 2004. Gargiullo, M.B. 2008. A Field Guide to Plants of Co sta Rica. Oxford University Press, New York, page 324. Jersakova, J. 2006. Mechanisms and Evolution of Dec eptive Pollination in Orchids. Biological Reviews 81(2): 219235. Roy, B.A. and A. Widmer. 1999. Floral Mimicry: A Fa scinating Yet Poorly Understood Phenomenon. Trends in Plant Science 4(8): 325-330. Schemske, D.W. 1976. Pollinator Specificity in Lantana camara and L. trifolia (Verbenaceae). Biotropica 8(4): 260-264. Schemske, D.W. 1983. Lantana camara. In: Costa Rica Natural History, D.H. Janzen, ed. The University of Chicago Press, Chicago, IL, pp. 266-268. Schlter, P.M. 2008. Molecular Mechanisms of Floral Mimicry in Orchids. Trends in Plant Science 13(5): 228-235. Todzia, C.A. 1983. Epidendrum radicans In: Costa Rican Natural History, D.H. Janzen, ed. The University of Chicago Press, Chicago, IL, pp. 243-244. Wolfe, L.M. 1987. Inflorescence Size and Pollinaria Removal in Asclepias curassavica and Epidendrum radicans. Biotropica 19(1): 86-89. Woo, Jennifer. The Effects of Spatial Distribution on the Proportion of Pollinia Removed in Epidendrum radicans (Orchidaceae). CIEE Program. Spring 2001.

PAGE 8

FIGURE 1. Total butterfly visits to inflorescences of Lantana camara and flowers of Epidendrum radicans in patches with no added reward to L. camara flowers (A) compared to those whose flowers had additional 20% sucrose adde d (B). nn r FIGURE 2. Total butterfly visits to flowers of Epidendrum radicans (orange) and inflorescences of Lantana camara (red) in patches with no added reward (A) compared to L. camara inflorescences with an additional 20% sucrose solu tion (B). The patch with no added reward (A) had 80 more butterfly visits to flowers of E. radicans and 64 more butterfly visits to inflorescences of L. camara.

PAGE 9

nnnrr r FIRGURE 3. Average length of butterfly visitation to flower of Epidendrum radicans and inflorescences of Lantana camara in patches with no added reward (A) compared to pa tches where 20% sucrose was added to inflorescences of L. camara (B). Inflorescences of L. camara within patch (B) were visited on average 33.072 sec onds longer. (Error bars represent +/SE) n n FIGURE 4. Total pollinia removed from flowers of E. radicans within patch of no added reward (A) compared to patches with an additional 2 0% sucrose added to flowers of Lantana camara (B). 52 pollinia were removed from flowers of E. radicans in patch A, while 35 pollinia were removed from patch B with increased reward. N = 182 N = 317 N = 102 N = 253

PAGE 10

! n n FIGURE 5. Total visits by most common butterflies t o flowers of Epidendrum radicans and inflorescences of Lantana camara in patches without added reward (A) compared to pat ches with 20% sucrose added to flowers of L. camara (B). The most common pollinators within both patch A and B were Danaus plexippus Anartia fatima and Leptophobia aripa. Butterfly species A. fatima visited 178 flowers and inflorescences in patch A and 195 individuals in patch B. Butterfly species L. aripa visited 45 flowers and inflorescences in patch A an d 3 individuals in patch B. Lastly, D. plexippus visited 12 individuals in patch A and 18 flowers an d inflorescences in patch B. n n FIGURE 6. Average time spent probing flowers of Epidendrum radicans and inflorescence of Lantana camara in patches without added reward (A) compared to pa tches with 20% sucrose added to flowers of L. camara (B). Average time is divided for the three most c ommon butterfly species within both patch A and B: Danaus plexippus Anartia fatima and Leptophobia aripa. D. plexippus spent the on average the longest time within plot A and B, in comparison to the other species, 37.471 +/SE and 86.759 +/SE seco nds, respectively. (Error bars represent +/SE) N = 272 N = 247 N = 17 N = 29 N = 72 N = 10

PAGE 11

! FIGURE 7. The average number of flowers of Epidendrum radicans visited by butterfly species Anartia fatima (red), Danaus plexippus (orange), and Leptophobia aripa (white) in patches without added sucrose reward to Lantana camara flowers (A) compared to patches with a 20% sucrose solution added to flowers of L. camara (B). In response to the added sucrose solution in patch B, A. fatima visited on average 0.039 fewer flowers, D. plexippus visited on average 0.75 fewer flowers, and L. aripa visited on average 0.184 more flowers within patch B. (Error bars represent +/SD) FIGURE 8. The average number of inflorescences of Lantana camara visited by butterfly species, Anartia fatima (red), Danaus plexippus (orange), and Leptophobia aripa (white) in patches without added sucrose reward (A) compared t o patches with a 20% sucrose solution added to L. camara flowers (B). (Error bars represent +/SD) N = 6 N = 73 N = 5 N = 4 N = 99 N = 38 N = 173 N = 13 N = 34 N = 174 N = 24 N = 4