Butterfly Proboscis Length and Pollen Load Charles McCanna Department of Biology, University of Wisconsin Madison _____________________________________________________________ ABSTRACT Pollination mutualisms between plants and pollinators facilitate in crea ses in genetic variability for plants while providing rewards for pollinators. Specialization of pollinators on specific plants has occurred to maximize benefits of these pollination mutualisms. With this specialization, pollination syndromes have ev colors, short to medium corollas, and strong scents that tend to attract butterflies. However, some of these flow ers have evolved long corollas to specialize on long proboscis butterflies. The purpose of this study was to examine the effect of corolla length on degree of pollen specificity in butterfly pollinated flowers. This was completed by observing proboscis l ength of butterflies and the corolla lengths of flowers visited . A total of 54 butterflies of 29 different species were collected from a fragmented Lower Montane Moist Forest in Ca Ã± itas, Costa Rica. Following removal and measurement, proboscises were obs erved under a microscope for pollen grains of 24 different species of flowering plants. Proboscises of lengths less than 10 mm were defined as short proboscises. Butterflies with proboscises greater than 17 mm were defined as long proboscis butterflies. Butterflies with long proboscises had a greater pollen species richness, often times twice that of small proboscis butterflies. Pollen from long corolla and short corolla flowers was found on all lengths of proboscises. Pollen loads on proboscises sugge st that pollination by butterflies is not specific in this community. This non specificity of butterflies on flowers has ecological and evolutionary effects on community composition and stability. RESUMEN Los mutualism o s de la polinizaciÃ³n entre l as pl antas y los poliniz adores facilitan la variabilidad genÃ©tica creciente para las plantas mientras que proporcionan recompensas para los pol in izadores. La especializaciÃ³n de los polinizadores en plantas especÃficas ha ocurrido para maximizar las ventajas de estos mutualism os de polinizaciÃ³n. Con esta especializaciÃ³n, los sÃndromes de la polinizaciÃ³n han evolucionado para aumentar la s posibilidades de la flor de ser visita por un pol in izador especÃfico. Por ejemplo, las "flores de mariposa" han desarrollado co lores rojos brillantes, corolas de tamaÃ±os cortos a medianos y olores fuertes que tienden a atraer mariposas. Sin embargo, algunas de estas flores han desarrollado corolas larga s para especializarse en mariposas con probÃ³scides largas. El propÃ³sito de este estudio fue examinar el efecto de la longitud de l a corola en el grado de especificidad del polen en flores polinizadas por mariposa s . Esto fue logrado por medio de la observacion de la longitud de la s probÃ³scide s de mariposas y las longitudes de l a corol a de las flores que visita ron . Un total de 54 mariposas de 29 especies diferentes fue colectado de un bosque hÃºmedo montano fragmentado en CaÃ±itas, Costa Rica. DespuÃ©s de ser removidas y medidas, bajo un microscopio para indentificar los granos de polen d e 24 especies diferentes de plantas florecientes. Las probÃ³scides con longitudes de menos de 10 mm fueron categorisadas como probÃ³scides cortas. Las mariposas con probÃ³scides de mÃ¡s de 17 mm fueron clasificadas como mariposas con proboscides largas. Las ma riposas con la s probÃ³scides larg a s tenÃan una mayor riqueza de especie s de polen, a menudo el doble de las mariposas con probÃ³scides pequeÃ±as. Se contrÃ³ polen tanto de flores de corola larga como de corola corta todas las longitudes de probÃ³scides. Las ca rgas de polen en las probÃ³scides sugieren que la polinizaciÃ³n por las mariposas no e s especÃfica en esta comunidad. Esta falta de especializaciÃ³n d e las mariposas a ciertas flores tiene efectos ecolÃ³gicos y evolutivos en la composiciÃ³n y la estabilidad de la comunidad.
INTRODUCTION Mutualisms between animal pollinators and plants facilitate plant outcrossing while rewarding pollinators (Proctor et al. 1996). During feeding, pollen grains stick to pollinators allowing their trans portation to conspecific p lan ts thus increasing genetic variability (Price 1984). In exchange for pollen transportation, animal pollinators often receive a nectar reward. Specialization of pollinators on specific plants is common , resulting in maximization of potential benefits o f the mutualism (Pellmyr 2002). With these specific pollinators, plants have greater chances of moving pollen to conspecific stamens. As a result of this specificity , pollination syndromes have evolved in which morphological characteristics of flowers ar ise in attempt to elicit specific pollinator s colors, strong scents and nectar sources that are highly accessible through slender corolla tubes (Proctor et al. 1996). S ome of t hese flowers have evolved long coro llas to increase visitation of long tongued or long proboscis butterflies. Although flower visitation may occur by short proboscis butterflies, they are not able to obtain nectar from these long corolla flowers. Coevolvi ng morphological features of both flowers and butterflies aim to increase pollination specificity ( Feinsinger 1983 ). Butterflies with longer proboscises are presumed to be specialized on longer corolla flowers (Proctor et al. 1996). These flowers should have high visitation by long proboscis butterflies due to consistent nectar reward and decreased competition from short proboscis butterflies for resources. L ong proboscis butterflies can also obtain nectar and pollinate short corolla flowers, in turn dec reasing pollination specificity. If specialization is prevalent in communities , a majority of pollen load on long proboscis butterflies should be from long corolla flowers. Butterflies with short proboscises can only access nectar from small corolla flo wers (May 1992) thus their proboscis is expected to only have small corolla pollen. In both cas es , pollination efficiency of plant s should increase and butterflies should exhibit increased effective foraging . On the other hand, if long proboscis butterfl ies are opportunistic feeders and forage on a wide range of corolla lengths, specialization is not taking place. The purpose of my study was to examine the effects of coroll a length on degree of pollen specificity of butterflies. Proboscises with length s less than 10 mm were defined as short proboscises. Butterflies with proboscises greater than 17 mm were defined as long proboscis butterflies. It was hypothesized that as length of proboscis increases, butterflies would feed more opportunistically on a wider array of corolla lengths, conferring a higher richness of pollen found on proboscises . With these increases in proboscis length, specificity is expected to decrease. MATERIALS AND METHODS Data collection took place in the town of CaÃ±itas, Costa Rica from October 25 th to November 15 th . The study site consisted of pasture bordered by fragmented Lower Montane Moist Forest (Hayes and Laval 1989). This property is owned by DoÃ±a Engracia Arguedas. Fifty four total individuals of 29 different specie s of butterflies were collected using a butterfly net. All butterflies were identified to species (DeVries 1987).
Proboscises were removed from butterflies with tweezers and placed on a piece of rolled Scotch tape. Using a metal probe, proboscises were e xtended to full length on the tape. The tape was then placed on a microscope slide and length of the proboscis was measured. Pollen located on these proboscises was observed with a Carl Zeiss compound light microscope at 400X. Using a self compiled poll en database of 24 flowering plants within 300 m 2 of the study site, pollen on proboscises was identified to plant species. Corolla lengths of these specific flowering plants were measured to examine the relationship between proboscis length of butterflies and corolla lengths of visite d flowers (Table 1). Pollen of the 24 species was easily distinguishable. Unknown pollen species found of proboscises were only included in pollen richness c ounts on proboscises. RESULTS Twelve Anartia f atima , four Helic onius clysonmus, three Paplio polyxines , three Siproeta epaphus epaphus and three Pieriballia mandela noctipennis were collected. Of the 24 remaining butterfly species, one to two individuals were collected (Table 2). Pollen from 12 out of the 24 surveye d flowering plants was fou nd on the 54 proboscises. Polli nia from Asclepias curassavica (Apocynaceae) were found on two thirds of all butterfly species collected, while Lantana camara (Verbenaceae) pollen was found on about one fourth of butterfly species . The ten other species of flowering plants were visited by 6 or less butterfly species (Fig. 1). A significant correlation was observed between length of butterfly proboscis and the number of pollen species on proboscises (Linear regression, r = .56, p < .05, N = 54 individuals). Butterflies with long proboscises had greater pollen species richness than those with short proboscises (Fig. 2). Longer proboscises carried up to twice the n umber of pollen species as short proboscises. L ong and short p roboscis butterflies carried pollen of both long and short corolla flowers. No significant trend was observed between proboscis length and length of corollas visited by butterflies (Linear regression, r = .13, p > 0.05, N = 84) . Oerstedella centrodenia ( Orchidaceae), Pentas lanceolata (Rubiaceae), Kohleria sp. (Gesneriaceae), and Malvaviscus penduliflorus (Malvaceae) with corolla lengths of 7.5 mm, 16.1 mm, 22.5 mm and 34.5 mm respectively, were visited by butterflies that had proboscises whose lengths we re less than that of their corresponding corollas (Fig. 3 ) (10 instances). Eighty six percent of pollen grains found on proboscises were from flowers with corolla lengths 10 mm or less. DISCUSSION The hypothesis that long proboscises would have grea ter pollen richness than short proboscises was supported. This result may be do to the fact that butterflies with long proboscises have more flower feeding options and therefore visit flowers of all corolla lengths. These l ong proboscis but terflies feed o pportunistically in the community . In further support of this conclusion , no correlation was observed between length of proboscis and lengths of corollas visited. Butterflies with short proboscises have the potential to pollinate long corolla flowers ev en though nectar sources are unreachable. This may possibly be due to the fact that long corolla flowers have different placement of pollen and nectar than small corolla flowers (Endress 1994). In this case, actual corolla
length may be an overestimate o f distance to nectar source. Kohleria sp. and M. penduliflorus ( hummingbird pollinated plants) possibly evolved long anthers to maximize chances of pollen dispersal when hummingbird visitation is limited (Feinsinger 1983). These long anthers increase cha nces of pollen transfer; butterflies, insects and wasp s that land on these anthers can act as pollen transporters . O. centrodenia and P. lanceolata lack jutting anthers and therefore their pollen was probably gathered by short proboscis butterflies throug h random pollen transmission. This is likely to occur in communities with butterflies that feed on a wide range of corolla lengths . L ong proboscis butterflies may leave long corolla pollen on short corolla flowers, allowing later transfer of that pollen by short proboscis butterflies. With opportunistic feeding of butterflies , both l ong and short corolla flowers are expected to receive a high amount of pollen from foreign species. In this study, short proboscis butterflies probably failed to realize tha t nectar was unreac hable in long corolla flowers. Over time, butterflies should learn that nectar rewards are not reachable resulting in decreased visitation to that long corolla species. This may account for the small amount of s hort proboscis butterfly visitation to long corolla plants. A. curassavica and L. camara may possibly have been visited by the greatest number of butterfly species due to its high density in the pasture. The high accessibility to nectar of short corolla flowers may also contri bute to high butterfly species visitation. Other physiological characteristics of the flower such as nectar output could also play a role in high species visitation (Proctor et al. 1996). With long and short corolla pollen carried by all lengths of prob oscis es , specialization of butterfly pollinators on plants is not prominent in the community . This non specificity results in h igher community stability due to plant species having a variety of butterfly species pollinators (Bawa 1990) . Negative effects on an ecological community due to lack of specialization may also be observed . With less specialization, a decrease in pollen transfer to conspecifics is observed. H igher competition for potential pollinators also increases within plant species . Lack o f specialization of butterflies on flowering plants in this study ha s both positive and negative effects on community composition and stability. Pollination loads on proboscises suggest that butterfly pollination of flowers in the studied community is no t specific. Long proboscis butterflies do have a nectar foraging advantage over short proboscis butterflies. But why have these long proboscis butterflies evolved when short proboscis butte rflies can, as shown, pollinate long corolla flowers ? Further st udy may show that long proboscis butterflies have high frequency of visit ation to long corolla flowers. This possible result increases specificity and gives rise to explanations of long probosc is evolution in this community. ACKNOWLEDGEMENTS I would sincerely like to thank Alan and Karen Masters for their inspirational teaching throughout the greatest semester ever. Thanks to Javier Mendez for his cheerful attitude and laughter, adding infinite memories to an already unforgettable time. Gasner and H up with. Special thanks to Willow Zuchowski who helped identify numerous flowering plants. I give it up to my CIEE confidants for constant humor and support through thick and thin. And a long -
lasting fa rewell and thank you to my family tica who provided me with a dream study site and the experience of a lifetime. LITERATURE CITED Bawa, K. S. 1990. Plant Polli nator Interactions in Tropical Rain Forests. Annu. Rev. Ecol. System 21: 399 422 DeVries, P. J. 1987. The butterflies of Costa Rica and their natural history. Princeton University Press, Princeton, New Jersey. Endress, P. K. 1994. Diversity and evolutionary biology of tropical plants. Cambridge University Press, Cambridge, England. Feinsing er, P. 1983. Coevolution and Pollination. In: Coevolution , D. J. Futuyma and M. Slatkin, ed. Sinauer Publishers, Sunderland, Massachusetts. Pellmyr, O. 2002. Pollination by Animals. In: Plant Animal Interactions , C. M. Herrera, and O. Pellmyr, ed. Blackw ell Publishing, Malden, Massachusetts Proctor, M., P. Yeo, and A. Lack. 1996. The Natural History of Pollination. Timber Press, Portland, Oregon. Hayes, M. and R. Laval. 1989. The Mammals of Monteverde. Tropical Science Center, San Jose, Costa Rica. May, P. G. 1992. Flower selection and the dynamics of lipid reserves in two nectarivorous butterflies. Ecology 73(6): 2181 2191 Zar, J.H. 1984. Biostatistical Analysis. Prentice Hall, Englewood Cliffs, New Jersey.
__________________ ____________________________________________________________________ Table 1. Corola lengths of 12 different flowering plants within 300m 2 of the study site. The 12 different pollen types corresponding to these species were found on the 54 butterfly probos cises sampled in this study. ________________________________________________________________________ Species of Flowering Plant Length of Corolla (mm) Asclepias curassavica (Apocynaceae) 0.2 Unidentified Aster (Fig. 5) 3 Rubus roseifolius (Rosaceae) 4 Lantana camara (Verbenaceae) 4.5 Epidendrum radicans (Orchidaceae) 7 Hidalgo ternate (Asteraceae) 7.5 Oerstedella centrodenia (Orchidaceae) 7.5 Duranta erecta (Verbenaceae) 8.5 Impatiens walleriana (Balsaminaceae) 10 Pentas lanceolata (Rubiaceae) 1 6.1 Kohleria sp. (Gesneriaceae) 22.5 Malvaviscus penduliflorus (Malvaceae) 34.5
________________________________________________________________________ Table 2. Number of individuals collected from 29 different butterfly species i n pasture bordered by fragmented Lower Montane Moist Forest in CaÃ±itas, Costa Rica. ________________________________________________________________________ Butterfly Species Numbers of Individuals Collected Anartia fatima 12 Heliconius clysonmus 4 Pap lio polyxines 3 Siproeta epaphus epaphus 3 Pieriballia Mandela noctipennis 3 Anthanassa sp. 2 Consul electra 2 Dircenna relata 2 Catasticta strigosa actinotis 2 Dircenna klugii 1 Mechanitis polymnia isthmia 1 Cissia similis 1 Dircenna chiriquensi s 1 Phoebis sannae 1 Eretris suzannae 1 Lycorea cleobaea atergatis 1 Ithomia patilla 1 Euptychia millis 1 Metal mark sp. 1 Enantia melite amalia 1 Pereute charops 1 Eurema proterpia 1 Eueides lineate 1 Chloreuptychia arnaea 1 Actinote leucomela s 1 Phoebis argante 1
________________________________________________________________________ FIGURE 1: Number of butterfly species containing pollen from specific flowering plants in a fragmented Lower Montane Moist Forest in CaÃ±itas , Costa Rica. Butterflies were collected and pollen samples were observed from October 25 th to November 15 th . Asclepias cur rasivica (Apocynaceae) showed visitation by the greatest number of butterfly species (19) followed by Lantana camara (Verbenaceae) (8). The ten other flowering plants were visited by six or fewer butterfly species. ________________________________________________________________________
________________________________________________________________________ FIGURE 2: Effect of butterfly proboscis length on species richness of pollen found on proboscises. Fifty four individuals of 29 different butterfly species were collected from a fragmented Lower Montane Moist Forest in CaÃ±itas , Costa Rica from mid October to mid November. Increases in length of proboscis conferred increases in pollen richness (Linear regression, r = 56, p < .05, N = 54 individuals). ________________________________________________________________________ __________________________________________ ______________________________ FIGURE 3: Relation between length of butterfly proboscis and length of corolla flowers visited. A total of 84 pollen samples from 54 butterfly proboscises from Lower Montane Moist Forest in CaÃ±itas , Costa Rica were identified. Colle ction took place from October 25 th to November 15 th . No significant correlation was observed between proboscis length and corolla length of flowers visited (Linear regression, r = .13, p > 0.05, N = 84). ____________________________________________________ ____________________
________________________________________________________________________ FIGURE 4: Unidentified flowering plant of the Asteraceae family. Pollen from this flower was sampled from the orchid garden of Licho and Danis Perez, owners of Ros ewood Souvenirs in CaÃ±itas , Costa Rica. ________________________________________________________________________
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 M39-00367
Largo de la probscides de las mariposas y el peso del polen
Butterfly proboscis length and pollen load
Pollination mutualisms between plants and pollinators facilitate increases in genetic variability for plants while providing rewards for pollinators. Specialization of pollinators on specific plants has occurred to maximize benefits of these pollination mutualisms. With this specialization, pollination syndromes have evolved that increase a flowers chance of eliciting visitation of a specific pollinator. For example, butterfly flowers have evolved bright colors, short to medium corollas, and strong scents that tend to attract butterflies. However, some of these flowers have evolved long corollas to specialize on long proboscis butterflies. The purpose of this study was to examine the effect of corolla length on degree of pollen specificity in butterfly pollinated flowers. This was completed by observing proboscis length of butterflies and the corolla lengths of flowers visited. A total of 54 butterflies of 29 different species were
collected from a fragmented Lower Montane Moist Forest in Caitas, Costa Rica. Following removal and measurement, proboscises were observed under a microscope for pollen grains of 24 different species of
flowering plants. Proboscises of lengths less than 10 mm were defined as short proboscises. Butterflies with proboscises greater than 17 mm were defined as long proboscis butterflies. Butterflies with long
proboscises had a greater pollen species richness, often twice that of small proboscis butterflies. Pollen from long corolla and short corolla flowers was found on all lengths of proboscises. Pollen loads on proboscises suggest that pollination by butterflies is not specific in this community. This non-specificity of
butterflies on flowers has ecological and evolutionary effects on community composition and stability.
Los mutualismos de la polinizacin entre las plantas y los polinizadores facilitan la variabilidad gentica creciente para las plantas mientras que proporcionan recompensas para los polinizadores. La especializacin de los polinizadores en plantas especficas ha ocurrido para maximizar las ventajas de estos mutualismos de polinizacin. Con esta especializacin, los sndromes de la polinizacin han evolucionado para aumentar las posibilidades de la flor de ser visitada por un polinizador especfico. Por ejemplo, las "flores de mariposa" han desarrollado colores rojos brillantes, corolas de tamaos cortos a medianos y olores fuertes que tienden a atraer mariposas. Sin embargo, algunas de estas flores han desarrollado corolas largas para especializarse en mariposas con probscides largas. El propsito de este estudio fue examinar el efecto de la longitud de la corola en el grado de especificidad del polen en flores polinizadas por mariposas. Esto fue logrado por medio de la observacin de la longitud de las probscides de mariposas y las longitudes de la corola de las flores que visitaron. Un total de 54 mariposas de 29 especies diferentes fue colectado de un bosque hmedo montano fragmentado en Caitas, Costa Rica. Despus de ser removidas y medidas, bajo un microscopio para identificar los granos de polen de 24 especies diferentes de plantas florecientes. Las probscides con longitudes de menos de 10 mm fueron categorizadas como probscides cortas. Las mariposas con probscides de ms de 17 mm fueron clasificadas como mariposas con probscides largas. Las mariposas con las probscides largas tenan una mayor riqueza de especies de polen, a menudo el doble de las mariposas con probscides pequeas. Se encontr polen tanto en flores de corola larga como de corola corta todas las longitudes de probscides. Las cargas de polen en las probscides sugieren que la polinizacin por las mariposas no es especfica en esta comunidad. Esta falta de especializacin de las mariposas en ciertas flores tiene efectos ecolgicos y evolutivos en la composicin y la estabilidad de la comunidad.
Text in English.
Pollination by insects
Polinizado por insectos
Diversidad de especies
Tropical Ecology Fall 2004
Ecologa Tropical Otoo 2004
t Monteverde Institute : Tropical Ecology