Thursday 24 October, 1st session
O.01 The role of floral traits as predictor of pollinators along elevation and season on Mount Cameroon
Yannick Klomberg1, J. Mertens1, Robert Tropek1,2, Jiří Hodeček1, Štěpán Janeček1 1 Department of Ecology, Faculty of Science, Charles University, Prague, Czechia 2 Institute of Entomology, Biology Centre CAS, Ceske Budejovice, Czechia
Flowering plants are expected to phenotypically converge in response to comparable selective pressures by the functional groups of pollinators, possibly resulting in pollination syndromes. However, there are still uncertainties about the conditions potentially influencing this predictability of pollination systems, as well as the role and efficiency of specific traits in attraction of specific pollinators. Therefore, we studied plant-pollinator interactions at four elevations from lowland to montane forest of Mount Cameroon in both wet and dry season. 217 plant species flowering in all strata and elevations were video-recorded resulting in a huge dataset of 1,209 recordings with >46,000 plant–visitor interactions. Showing for example shifts from predominantly bee to fly and bird visitation from dry to wet season in higher elevations. 25 floral traits (e.g. shape, colour and nectar characteristics) were measured in all flowering plants in order to distinguish the role of specific floral traits in the attraction of individual functional groups of visitors. Results of regression trees and non-parametric regression suggest that visitor limiting traits, being shape, size and tube width/length, play an important role in predicting potential pollinators. The best predicted were hymenopterans and lepidopterans. Using a trained regression tree model, the pollinators of specific plant species can be predicted relatively accurately across elevations and seasons. The general predictability of pollinators slightly increases towards higher elevations, while there is no systematic effect of season, although the predictability often differs inter-seasonally in each elevation.
O.02 Impact of invasive Rosa rugosa on pollinator species composition of coastal sand dunes
Elżbieta Rożej-Pabijan1, Irena Grześ2, Anna Stefanowicz3, Marcin Woch1 1 Institute of Biology, Pedagogical University of Kraków, Kraków, Poland 2 Department of Environmental Zoology, Institute of Animal Sciences, Agricultural University, Kraków, Poland 3 W. Szafer Institute of Botany, Polish Academy of Sciences, Kraków, Poland
Dune habitats are highly susceptible to change from both natural and human-induced factors. One of the factors is invasion of Rosa rugosa that results in fragmentation of dune landscapes and changes vegetation to monospecific shrubs rich in flowers long in season. Shrubs can attract pollinators, however many pollinators are specialized toward host plants and need an open sandy area as a nesting site. The aim of our study was to compare pollinator faunas of R. rugosa patches and adjacent native sand dunes. The study was performed at 25 matched pair plots (invaded, uninvaded) at the Polish coast of the Baltic Sea. We checked pollinator species composition (bees and butterflies), by sweep netting and pitfall trapping and determined visitation rates on both plot types during fieldwork. Higher overall pollinator diversity, 20 species, was found on uninvaded dune plots while invaded plots contained 14 species. Uninvaded dune plots had 65 % of species in common with invaded plots, indicating that dune pollinator species use R. rugosa as an important but not the only source of nectar and pollen. On uninvaded plots several species of dune specialist bee species were found. Increased abundance and visitation rate of generalist pollinators was observed on invaded plots and was on average three times higher than on dune plots. The results indicate that R. rugosa invasion affects pollinator sand dune communities by promoting generalist pollinators and eliminating sand dune specialists.
O.03 Some things we know about the bees of the Aegean Archipelago
Theodora Petanidou Departement of Geography, University of the Aegean, Greece
Despite its apparent simplicity, the Aegean is a very complex world. This complexity results basically from its extraordinarily high diversity in all kinds of beings, including plants and insects. Understanding the essence of the complex world of the Aegean constitutes a challenging multi-task effort involving many researchers worldwide. My lab’s systematic research on the pollinators (with particular focus on bees and hoverflies) of the Archipelago started in 2004, focusing mainly on their diversity, pollination services they provide, as well as the particular threats these pollinators face in the Aegean. The systematic work carried out so far covers >250 sites in 25 Aegean islands, with extra comparative work carried out on other islands and mainland areas. In my talk I will tackle a few main results of the above investigation which is still ongoing. Our approach, using classical observation and molecular ecology, as well as novel holistic tools (e.g. ecological network analysis), covers ecological, biogeographical, and conservation/restoration aspects. The talk will be organized in three umbrella-issues: (i) diversity and biogeography co-considering functional aspects of the plant–pollinator system; (ii) global threats (e.g. global warming threatening synchronization of pollination partners and nectar reward availability, as well as frequency and intensity of fires); and (iii) conservation vis-à-vis traditional management in the Aegean (e.g. grazing).
O.04 The role of bee diversity for size-specific interaction patterns in grasslands of South Africa
Saskia Klumpers, Steven Johnson School of Life Sciences, University of KwaZulu-Natal, Scottsville, Pietermaritzburg, South Africa
Size-matching between pollinator proboscis length and floral nectar tube depth of interacting species plays a crucial role in both pollination ecology and evolution. Community composition, specifically pollinator diversity, may affect species generalization degree and consequently the degree of size-matching. An increase in pollinator diversity may increase interspecific resource competition. Consequently, species may be more specialized which may lead to better size-matching. We investigated how bee diversity affected generalization degree and size-matching of Amegilla natalensis, a keystone long-tongued bee species in the grasslands of South Africa. We asked the following questions: (1) Is the average proboscis length of bees in a community related to the average nectar tube depth of flowers and are these traits related to bee diversity? (2) Are plants species that are visited by A. natalensis characterized by specific traits? And (3) does bee diversity affect generalization degree of A. natalensis and the average nectar tube depth of the flowers that it visits? We found that the average proboscis length of bees in a community was positively related to the average nectar tube depth of the flowers. Both traits were negatively related to bee diversity. A. natalensis visited flowers that were pink, blue or white and had deep nectar tubes (> 6 mm). In communities with a greater bee diversity, A. natalensis was more specialized and visited on average deeper-tubed flowers which better matched the length of its proboscis. Our results show that floral nectar tube depth distribution closely match bee proboscis length distribution within plant–bee communities. More diverse bee communities have a higher proportion of short-tongued bees. Therefore, for a long-tongued bee, an increase in bee diversity increases its specialization and leads to better size-matching. Consequently it highlights the possible role of species diversity for the facilitation of coevolution between plants and pollinators.
O.05 Understanding the functional role of migratory hummingbirds in plant-hummingbird interactions along a latitudinal gradient
Ainhoa Magrach Basque Centre for Climate Change, BC3, Leioa, Spain
Predicting how species, communities and ecosystems will respond to global environmental change remains a key scientific challenge. Much progress has been done in understanding how species interact and assemble into complex networks. However, the dynamic nature of these species assemblages and the role of biodiversity in shaping them remain poorly understood. To fill these gaps, we evaluated structural changes in plant-hummingbird interaction networks within communities harboring resident and migratory hummingbird species along a latitudinal gradient of increasing species diversity from central Mexico to Alaska. We surveyed plant and hummingbird communities in the presence and absence of the migratory species to evaluate changes in species functional roles and community structure. Our results show a greater niche overlap within the more diverse areas, which allows resident species to take over the role of the migrant species in its absence. Contrastingly, we find that the migrant, behaves as a more specialized species within less diverse areas, monopolizing the most abundant resources and leading resident species to feed on less abundant resources. In the absence of the migrant species, its role is however not fully covered by resident species, resulting in communities that are less robust to further perturbations. These results have important implications for community persistence given expected changes in the migratory ranges of many species.
O.06 Does different pollinator preference solve reproductive conflicts between co-flowering species?
Rocio Perez-Barrales School of Biological Sciences, University of Portsmouth, Portsmouth, United Kingdom
Pollinator preference, behaviour and flowering phenology regulate intra- and interspecific visitation in plant communities. In the Brazilian Cerrado, Palicourea species flower together between October and February during the rainy season. We investigated pollinator preference in co-flowering Palicourea species in the Cerrado in order to decipher mechanisms allowing co-flowering while reducing (potential) reproductive conflicts associated with pollinator sharing. We first studied pollinator behaviour to describe the level of pollinator sharing between P. coriacea and P. officinalis at different moments of the flowering season. In P. coriaceae, we then investigated pollen delivery and pollen deposition in single species and co-flowering patches. Our results showed that Bombus atratus was the main visitor to both species, but during the co-flowering period visitation increased in monospecific patches of P. coriacea as opposed to patches with both species. Avoidance of co-flowering patches might be related to nectar accessibility, which might be limited by the longer flower tube in P. officinalis compared to P. coriacea. This in turn resulted in an increase of pollen delivery in monospecific stands, while pollen deposition on the stigmas increased in co-flowering patches. Taken together, the results revealed a combination of strategies that reduce potential reproductive costs associated with co-flowering.
Friday 25 October, 2nd session
O.07 Extreme weather causes unexpected provision of pollination services
Björn Klatt Department of Biology, Biodiversity Unit, Lund University, Lund, Sweden
Climate change is threatening the structural and functional stability of both aquatic and terrestrial ecosystems. In particular, the increased frequency and intensity of extreme weather events can result in ecological surprises which makes it difficult to predict the consequences of climate change on biodiversity and thereby on ecosystem services. In a mesocosm experiment, we show how an extreme weather event resulted in unexpected interactions between bumblebees in an agricultural ecosystem and floating algal mats in an aquatic ecosystem – with consequences for bumblebee colony development and pollination services. More specifically, bumblebee colonies grew larger when bees could use algal mats as rafts to access water. In contrast, colonies were lighter and crop yield was higher when bees had no access to water due to the absence of floating algal mats. Bumblebees without access to water spent longer time visiting flowers when the access to water was limited. We hypothesise that by spending longer time collecting nectar, bumblebees also satisfied their water needs, which resulted in increased bee-flower interactions, better pollination and thus higher crop yield. Our findings exemplify how extreme weather events can drive complex and unpredictable responses of organismal behaviour, and that interactions between terrestrial and aquatic environments can be an important factor for the provision of ecosystem services under future climate change.
O.10 Why be equal? Ecological relationships between co-occurrent species with similar floral display
Maria Gabriela Gutierrez de Camargo1, Klaus Lunau2, Montserrat Arista Palmero3, Leonor Patrícia Cerdeira Morellato1 1 Institute of Biosciences, Department of Botany, Phenology Lab, UNESP - São Paulo State University, Rio Claro, São Paulo, Brazil 2 Department Biology, Institute of Sensory Ecology, Heinrich-Heine-University Düsseldorf, Düsseldorf, Germany 3 Faculty of Biology, Department of Vegetal Biology and Ecology, University of Sevilla, Sevilla, Spain
Within a community, while plants present flowering times and spatial distributions to favour pollination success and avoid interspecific competition, pollinators look forward optimizing their food search. In this context, bees’ floral constancy can be advantageous for both, plants and pollinators. Since floral constancy depends on the bees’ ability to discriminate floral signals, it is expected that co-occurrent plant species will diverge in their floral displays or present temporal segregation of flowering to avoid heterospecific pollen deposition. For groups of co-occurrent taxonomically related species with highly similar floral display we investigated if the flower colours can be distinguished by bees (colour loci distance in the bee colour space) and how their flowers are displayed in time (flowering phenology). We intend to infer possible ecological relationships among these species which allow their co-occurrence in space and look for cues of signal standardization. We analysed 12 groups of species, six from the Brazilian campo rupestre vegetation and six from Spanish Mediterranean shrublands. Among the species of each studied group almost no flowers can be discriminated by bees in order to maintain floral constancy, confirming the similar display for foraging bees. Based on our preliminary phenological results, different temporal patterns were observed between and among groups. Mainly for the temperate groups, which are more constrained by a rest season, some species overlapped their flowering peaks and can be under competition or facilitation selective pressures. However, some species presented a tendency of sequential flowering and, besides avoiding interspecific competition, may facilitate each other by signal standardization, maintaining a reliable visual signal for bees over time in the shared space. More detailed analysis is necessary to understand the ecological relationships allowing the coexistence of species with similar flower displays.
O.11 Linking soil environment to insect-plant mutualisms in a temperate, nitrophilous tree species
Alison Scott-Brown, M. -J.R. Howes, G.C. Kite, J.H. Martin, P.C. Stevenson Royal Botanic Gardens, Kew, United Kingdom
The concept of thrips as pollinating insects in temperate regions is rarely considered as thrips are more frequently regarded to be destructive florivores. Taking an integrative approach to studying thrips behaviour and floral chemistry we provide insights into the ambiguous pollination strategies of Sambucus nigra L. (common elder) and provide evidence that suggests that the relationship between S. nigra and Thrips major is mutualistic. Understanding what drives mutualisms between plants and insects can provide the basis for studying more complex relationships between land-use change and plant health and reproduction strategies, particularly as mutualism can rapidly revert to antagonism if ecological conditions change. For example, availability of NO3- and NH4+ can alter the expression of specialised floral secondary metabolites and volatile organic compounds that influence a range of ecological functions such as pollinator behaviour, herbivore defence and adaptation to environmental stress. Using gas- and liquid-chromatography coupled with mass spectrometry in conjunction with insect monitoring and morphological analysis of floral tissues we explored whether soil chemistry alters floral traits of S. nigra at three sites; arable, dairy and a conservation site of special scientific interest (SSSI), each with a distinct soil chemistry profile. Our study showed that while pollen chemistry favoured pollinating thrips in plants in arable and SSSI locations, floral morphological traits associated with resource availability were enhanced in dairy farm plants, resulting in similar numbers of thrips visiting flowers in arable and dairy sites. The methods developed through this study enable an efficient and timely mechanism for identifying and communicating how environmental changes such as application of soil improvers used in agriculture can potentially alter floral traits of common native hedgerow species which are important food sources for insect pollinators, impacting their behaviour and populations at a time when some key pollinating species are in decline.
O.12 Spectral tuning of flower coloration to pollinator vision
Casper van der Kooi University of Groningen, The Netherlands
Flowers display vivid colours to entice pollinators. Whereas the pigmentary aspects of flower coloration (e.g. synthesis, molecular biology and evolution of floral pigments) have been widely studied, the optics, i.e. the interaction of light with the flower’s interior components, remain largely unstudied. It remains further unknown how floral visual signals are tuned to pollinators with different visual systems, such as bees, butterflies and birds. Here, I present results from a large-scale comparative study on flowers of species-pairs with contrasting pollination systems. First results suggest convergent evolution of floral visual signals to pollinator vision along three different routes, which all result in high visibility to respective pollinators: (i) the type of floral pigment evolves to yield hues that are most conspicuous to the main pollinator; (ii) the amount of pigment evolves such that it yields the highest saturation (“purity”) possible for the specific pigment; (iii) the amount of light reflected (“brightness”) increases upon switches from diurnal to nocturnal pollination. These three lines of evidence suggest that the optical properties of flowers convergently evolve via different means to yield the highest salience to pollinators.
O.13 Sunbirds of a feather: Pollinator partitioning within the morphologically diverse Cotyledon orbiculata L. complex (Crassulaceae) in South Africa
Alexander de Gouveia, Tracey L. Nowell, Craig I Peter Department of Botany, Rhodes University, Grahamstown, South Africa
Cotyledon L. (Crassulaceae) is a small genus (11 species) of succulent shrubs that is largely endemic to South Africa. The most widespread of the species, Cotyledon orbiculata L., exhibits considerable morphological variation across its range, particularly amongst its flowers, signifying pollinator partitioning. In terms of reproductive traits, the pendulous red, orange or yellow flowers vary in corolla tube length and width, the degree to which the stamens and styles are exerted, and in the volume of nectar available. Peak flowering occurs in spring and summer, and different forms of C. orbiculata appear to have non-overlapping flowering times, especially when sympatric. The suite of floral traits found in C. orbiculata, is consistent with ornithophily, more specifically, sunbird pollination, and the variation therein suggests partitioning within thesunbird guild based on their size and behaviour. We studied the pollination biology of different forms of C. orbiculata in order to test a hypothesis of pollinator-mediated differentiation in this species complex. We aimed to address the following specific questions: 1) Is the variation in floral morphology across C. orbiculata continuous or is there evidence that discrete floral clusters are maintained? 2) Are different of floral forms of C. orbiculata separated temporally? 3) Do different sunbird species visit different floral forms preferentially? Field observations and data collection were carried out at multiple sites across the Eastern Cape, Western Cape, and Northern Cape provinces in South Africa. Sunbird visitation was quantified using Bushnell® HD Trophy CamTM camera traps and by direct observation. Variation in floral morphology was measured using multivariate analysis of measurements of the gynoecium, androecium, corolla, and calyx. Information on flowering phenology was collated from herbarium specimens and field visits and analysed as circular data to assess degree of overlap between forms. Nectar volume was recorded using microcapillary tubes (ringcaps®, Hirschmann® Laborgerate), and nectar concentration was determined using an Atago® refractometer. Malachite (Nectarinia famosa L.), Greater Double-Collared (Cinnyris afer L.), and Southern Double-Collared (Cinnyris chalybeus L.) Sunbirds were found pollinating C. orbiculata. Analyses of floral morphometric data identifies three discrete clusters within the forms of C. orbiculata sampled - corolla tube length, corolla lobe length, and filament length, account for most of the variation. All three mentioned-species of sunbirds pollinated flowers having the longest corolla tube length (19.28–36.67 mm), while the Greater Double-Collard Sunbird was found pollinating flowers with medium (13.65–20.70 mm) and small (9.09–12.81 mm) corolla tubes. Average nectar volume was most abundant in flowers with the longest corolla tubes (22.65 μL), compared to flowers with the smallest corolla tubes (1.81 μL). Phenology patterns do not correlate with morphometric and nectar data sets and is year-round and non-overlapping. Findings from this study provides a new insight into the pollination biology of C. orbiculata: 1) That nectar-producing flowers are readily available to entice a variety of sunbirds year-round. 2) Flowers are non-competing, thus, as the flowering period ends for one flower, the next one starts to flower, thus, ensuring constant sunbird visitation. 3) That sunbird diversity, for a flower, is correlated to nectar volume and flower size. 4) That there may be more species within the C. orbiculata complex than initially anticipated.
O.17 Bees are inefficient pollinators in cloud forests - investigating pollinator shifts in Merianieae
Agnes Dellinger1, Rocio Pérez-Barrales2, Diana Fernández-Fernández3, Jürg Schönenberger1 1 Department of Botany and Biodiversity Research, University of Vienna, Vienna, Austria 2 University of Portsmouth, United Kingdom 3 Instituto Nacional de Biodiversidad, Ecuador
Changes in pollination efficiency are thought to be the main driver of shifts in reproductive strategies in plants relying on pollination by animals. Such changes may occur when plants colonize new habitats, abiotic environmental conditions change or new pollinators appear. Environmental conditions and pollinator communities differ along altitudinal gradients and species which grow in lowland habitats often differ in their pollination strategy from closely related species which have colonized montane habitats. In particular, numerous Neotropical plant clades show a consistent shift directionality from bee to vertebrate pollination associated with growth at high elevations. Surprisingly, the underlying assumption that these pollinator shifts arose from a reduced efficiency of bee pollinators in mountain ecosystems has rarely been tested in the field. Using Merianieae (Melastomataceae) as study system, we test the hypothesis that bees are less efficient pollinators at high elevations than vertebrates, hence leading to repeated shifts from bee to vertebrate pollination. We employ manual pollination experiments, pollinator observations and measure pollen transfer (pollen export and pollen deposition) in six Merianieae species pollinated by either bees, passerine birds or mixed assemblages of vertebrates. Furthermore, we comparatively study selected floral traits crucial for pollen transfer in 47 Merianieae species to understand how pollen transfer is optimized in the different pollination systems. Our results confirm that bees are highly inefficient pollinators in montane cloud forests as bee-pollinated plants were strongly pollen limited. Vertebrate pollinators were significantly more efficient in transferring pollen, even when visitation rates were low. Employing different vertebrate pollinator groups (e.g. hummingbirds and bats) maximized pollen transfer in Merianieae. Certain floral adaptations to increase pollen transfer, such as enlarged stamen pores, bigger stigmatic areas or changes in pollen release mechanisms, associated with pollinator shifts.
O.18 Poricidal anthers as pollen dispensers
Jurene Kemp University of Stirling, Stirling, United Kingdom
The buzz pollination syndrome, where sonicating bees extract pollen from poricidal anthers, has evolved independently across 65 plant families. Despite the prevalence of this syndrome, the function of poricidal anthers remains poorly understood. One possible function is that poricidal anthers act as dispensing mechanisms that gradually release pollen, and thus influence pollinator visitation and pollen deposition rates. We assessed whether poricidal anthers of various Solanum species gradually dispense pollen by applying simulated bee vibrations to flowers. If the gradual release of pollen is an adaptation to pollinators, we would expect the pollen release rate to vary with species’ reliance on pollinators. We tested this and found that species with low pollinator reliance (i.e. selfing species) released most of their pollen after one or few simulated bee visits, whereas those reliant on pollinators released their pollen more gradually across multiple simulated visits. This suggests that the dispensing mechanism is an adaptation to pollinators. Because different bee taxa produce different vibrations, we tested how changes in vibration velocity influence pollen dispensing rates. Low velocity vibrations resulted in a reduction in total pollen released, and pollen was dispensed more gradually than when high velocity vibrations were applied. Male fitness modelling revealed that low velocity vibrations do not necessarily reduce fitness, as long as visitors have low diminishing returns and high visitation rates. In contrast, high velocity vibrations resulted in high male fitness after few visits, but only when there were no or few diminishing returns. Additionally, high velocity vibrations always resulted in higher pollen reward for pollinators than low velocity vibrations. This suggests that bees that produce either high or low velocity vibrations can act as effective pollen exporters of buzz pollinated plants, as long as visitation rates are high, but only bees that produce high velocity vibrations receive large pollen rewards.
O.19 Pollination from the bees’ perspective: do bees use nutritional cues to select pollen?
Sara Leonhardt1, Fabian Ruedenauer1, Johannes Spaethe2 1 Department of Zoology III, University of Würzburg, Würzburg, Germany 2 Department of Zoology II, University of Würzburg, Würzburg, Germany
Bees are highly important pollinators of many plant species. In return for their service, they obtain their main food resources, i.e. pollen and nectar. While nectar represents an essential source of sugar and thus energy, pollen provides bees with all other essential macro- and many micronutrients, e.g. protein, fat or sterols. The nutritional quality of pollen therefore has a major influence on the bees’ development, health and reproductive fitness. Given the high importance of pollen nutritional quality, we assumed that bees should pay close attention to nutritional cues in pollen in order to select and consume pollen which best meet their nutritional needs. We therefore investigated the role of different nutritional cues for pollen collection in bumblebees (Apidae: Bombus terrestris) and honeybees (Apis mellifera) through a combination of different behavioral assays and chemical analyses of resources. We found that bumblebees prioritized lipid perception when assessing pollen appropriateness and generally selected pollen of comparatively lower fat content, while totally ignoring its amino acid and sterol content. In contrast, honeybees payed attention to both pollen fatty acid and pollen amino acid content. Interestingly, foraging decisions appeared to ultimately affect colony fitness, as fatty pollen severely decreased the survival and reproductive fitness of bumblebees.
O.20 Pollination, paternity and mating portfolios in a hermaphroditic plant
Jeffrey D W Karron Department of Biological Sciences, University of Wisconsin-Milwaukee, United States of America
When pollen carryover is limited, a hermaphroditic plant may mate with different partners through male and female function. We characterized mating portfolios in Mimulus ringens, a hermaphroditic, bumblebee-pollinated plant native to central and eastern North America. Nearly every sampled seed could be assigned to a single sire. We used this information to quantify the overlap in mating through dual sex functions, which can be presented as a pair of bipartite networks. Mate diversity is high within fruits, but there was almost no overlap in mates between the two sexual functions. This finding might help explain the maintenance of hermaphroditism under conditions that would otherwise favor the evolution of separate sexes.
O.24 High land--use intensity in grasslands constrains wild bee species richness in Europe
Johan Ekroos, Henrik G Smith Centre for Environmental and Climate research (CEC), Lund University, Sweden
An increasing number of studies find negative relations between bee species richness and simplification of agricultural landscapes, but the role of land-use intensity in focal grasslands, and its relative importance compared to landscape simplification is less well-known. In particular, it is not known if common species, which are the dominant crop-visiting species, are more robust to increasing land-use intensity than rare species. We compared the effects of nitrogen inputs, as a proxy for land-use intensity, and proportion of natural and semi-natural habitat, as a measure of landscape complexity, on total bee species richness, rare species richness and dominant crop-visiting species richness. We used data from five European countries, consisting of 282 grasslands, covering the entire range of low intensity, no-input systems, to high-input grasslands, with more than 400 kg N/ha applied per year. We found strong negative impacts of increasing land-use intensity on total bee species richness across Europe. The richness of rare bee species was not significantly related to increasing land-use intensity, whereas dominant crop-visiting species richness was significantly reduced by increasing land-use intensity. Based on species accumulation curves, grasslands with no nitrogen inputs had higher total bee richness and higher shares of rare species compared with sites with high nitrogen inputs (>125 kg N/ha/year). Finally, we found no effects of increasing landscape complexity on bee species richness. Our results highlight the importance of retaining grasslands characterised by low land-use intensity across agricultural landscapes to promote the conservation of wild bees.
O.25 Honey bees vs non-Apis bees: pollination performance and response to bee abundance in sweet cherry orchards
Maxime Eeraerts, Guy Smagghe, Ivan Meeus Department of Plant and Crops, Ghent University, Belgium
Previous studies have highlighted the contribution of wild pollinators to sweet cherry production (Eeraerts et al., 2017; Eeraerts et al., 2019). Pollination performance of honey bees and other pollinators and interspecific interactions might explain the added value of pollinator diversity. In our study we focused on the foraging behaviour (flower visitation rate, probability of tree change, probability of row change and contact with the floral stigma) and pollination efficiency (fruit set of flowers that received only one visit) of honey bees and non-Apis bees in sweet cherry orchards in Belgium. The influence of honey bee abundance and non-Apis bee abundance on the foraging behaviour was also investigated. Single visit pollination efficiency on sweet cherry was higher for both mason bees and solitary bees compared to bumble bees and honey bees. Visitation rate of bumble bees and mason bees was higher compared to other solitary bees and honey bees, the latter also visiting more flowers per minute than solitary bees. Mason bees and bumble bees showed a higher probability of changing trees in the same row and a higher probability of changing trees between rows, respectively, compared to both solitary bees and honey bees. We also found that the probability that honey bees change trees between rows increased with increasing non-Apis bee abundance. Foraging behaviour of non-Apis bees was not influenced by honey bee abundance of non-Apis bee abundance. Our results highlight the higher pollination performance of non-Apis bees, especially that of mason bees and other solitary bees. We also conclude a facilitative component of non-Apis bees to crop pollination. Management to support species with high pollination efficiency and effective foraging behaviour for specific crops will promote crop pollination.
O.26 Investigating the effects of the neonicotinoid pesticide clothianidin on bumblebee foraging using an automated monitoring system
Pawel Kolano1, Katrine Borgå2, Anders Nielsen1 1 Centre for Ecological and Evolutionary Synthesis (CEES), Department of Biosciences, University of Oslo, Norway 2 Section for Aquatic Biology and Toxicology (AQUA), Department of Biosciences, University of Oslo, Norway
Bumblebees are important pollinators at high latitudes. Recent studies have shown that currently 46% of European bumblebee species have declining populations. This has caused concerns for the sustainability of bumblebee populations and the ecosystem service they provide both in wild plant communities and to entomophilous crops. The aim of this project was to design a monitoring system for tracking foraging bouts of individual bumblebees. The system consisted of a nest box and a dedicated camera box as an entrance/exit. Each worker bumblebee was equipped with a 2mm x 2mm data matrix (bCode, Tim Gernat) on its back. A data matrix much like a QR-code is a variation of a two-dimensional barcode. A computer with a camera and motion detection software controlled the system, taking a series of pictures each time it detected motion, i.e. when a bumblebee left or entered the hive. I used a tailored software (bTools) to scan each picture for bCodes. The software returns a text string containing the ID of the bumblebee(s) found in the picture and the exact time the picture was taken. By using these timestamps I was able to generate data on activity patterns, i.e. number and lengths of foraging bouts, on an individual level. Using this system I tested how different sub-lethal doses of the neonicotinoid pesticide clothianidin affected foraging behaviour on individual bumblebees. Preliminary results suggest that exposure to field-realistic doses of clothianidin significantly increases foraging bout length. More through results will be presented, both on the individual and hive level.
O.27 Assessing the risks and impacts of exposure to systemic insecticides for solitary, ground-nesting squash bees
D. Susan Willis Chan1, Ryan S. Prosser1, Jose L. Rodríguez-Gil2, Nigel E. Raine1 1 School of Environmental Sciences, University of Guelph, Guelph, Ontario, Canada 2 Department of Biology, University of Ottawa, Ottawa, Ontario, Canada
Ground-nesting solitary bees comprise around 70% of bee species in temperate climates. In these species, female bees contact relatively large amounts of soil as they excavate their nests. Using the hoary squash bee (Peponapis pruinosa) as a model species, we evaluated the risk to adult female ground-nesting bees of exposure to lethal doses of systemic insecticide residues (three neonicotinoids: clothianidin, thiamethoxam and imidacloprid, and the anthranilic diamide: chlorantraniliprole) in agricultural soil in Ontario, Canada. To do this, we sampled agricultural soil prior to insecticide application and during the bee-active period (July-August). These samples were analyzed for insecticide residues, and residue concentrations were plotted to produce an environmental exposure distribution for each insecticide. The probability of exceeding lethal endpoints was determined by comparing three LD50 benchmarks to the distribution curve. We found substantial risk to ground-nesting bees of exposure to lethal doses of clothianidin, thiamethoxam, and imidacloprid residues in agricultural soil using our squash bee model. Minimal exposure risk was found for chlorantraniliprole. In parallel to our risk assessment, we investigated the impacts of insecticide application practices for cucurbit growers on the behaviour and reproduction of squash bees. We introduced mated adult female squash bees into twelve net-covered hoop-houses containing blooming squash. Squash plants in each hoop-house were treated with one insecticide (either imidacloprid, thiamethoxam or chlorantraniliprole) or were untreated controls. We found significant sublethal impacts of exposure to imidacloprid on pollen foraging, nesting behaviour and reproduction of female squash bees, with no significant differences apparent for bees exposed to squash plants treated with either thiamethoxam or chlorantraniliprole. Our results show that in future risk assessments should include exposure impacts from pesticides in soil to be protective for ground-nesting bees, and that field-realistic levels of imidacloprid exposure could be significantly affecting foraging and reproduction of these important pollinators.
O.28 Scale-dependent mitigation of pollination
Henrik Smith Centre for Environmental and Climate Research (CEC), Lund University, Sweden
Ongoing agricultural intensification and landscape simplification negatively impact wild pollinators and the pollination service they provide to both crops and wild plants. As a result, there is currently a strong focus on how to benefit pollinator populations in agricultural landscapes, by e.g. preserving semi-natural habitats or providing supplemental flower resources in the form of flower strips. However, not all pollinators are the same, neither in terms of how they react to landscape change and mitigation measures, nor in what services they provide. Using recent research in our group, we demonstrate the implication for pollinators and pollination. Combining modelling and empirical field studies, we demonstrate (i) how spatio-temporal availability of food and nesting resources act as spatial ecological filter for bees, (ii) the scales at which mitigation of loss of flower resources affect pollinators, (iii) how competitive interactions modify the responses of individual species, and (iv) how this may explain how wild flower pollinated by generalist and specialist pollinators, respectively, are differentially affected by contemporary landscape simplification. We discuss our results in relation to where and when pollinator mitigation measures should be implemented to preserve pollination as a service.
O.29 VIP guests on flower scent parties at night: more than just moths and bats
Stefan Dötterl1, Isabel Alves dos Santos2, Guaraci Duran Cordeiro1, Florian Etl3, Robert R. Junker1, Andreas Jürgens4, Jette Knudsen5, Cristiane Krug6, Artur Maia7, Reisla Oliveira8, Samuel Prieto Benitez9, Clemens Schlindwein8 1 Paris-Lodron-University of Salzburg, Austria 2 University of São Paulo, Brazil 3 University of Vienna, Austria 4 Technische Universität Darmstadt, Germany 5 Nattaro Labs AB, Lund, Sweden 6 Empresa Brazileira de Pesquisa Agropecuária (Embrapa) Amazônia Ocidental, Manaus, Brazil 7 Universidade Federal de Pernambuco, Recife, Brazil 8 Universidade Federal de Minas Gerais, Belo Horizonte, Brazil 9 Universidad Rey Juan Carlos-ESCET., Madrid, Spain)
Nocturnal pollinators mainly use olfactory cues to find appropriate host plants. Floral scents of plants pollinated by bats and moths as well as specific compounds attractive for such pollinators are studied since decades. Much less is known about the chemical ecology of pollination systems involving other nocturnal pollinators, such as scarab beetles (Cyclocephalini) and nocturnal bees. In the last few years, we extensively studied the chemical communication of scarab and nocturnal bee pollination systems and in the present paper, we will ask whether the floral scents are similar or different to scents described from plants pollinated by moths and bats. More specifically, we test the hypothesis that there are distinct floral scent syndromes at night, with scents grouping according to the four groups of pollinators (moths, bats, scarab beetles, bees). We also tested for phylogenetic effects on scent patterns. We overall found differences in scent according to pollinator group, but also some obvious overlap between some groups, and that phylogeny explains a part of overall and within syndrome variations. Our data also show, that, despite the recent advances, more data are needed to better understand the ecology and evolution of floral scent.
O.30 Hearing insects with light
Salena Helmreich FaunaPhotonics, Denmark
We present a new technique for automated monitoring of pollinators and other flying insects using a light-based sensor. Quantifying insect behavior has always been very labor-intensive work requiring netting, trapping and/or visual observations. This severely limits the amount of information that can be collected for scientific research or base farming decisions on. We have developed an optical sensor capable of collecting, analyzing, and uploading non-intrusive insect recordings in real-time over extended periods. The instrument illuminates an air volume with invisible infra-red light and records the back-scattered light from passing insects. From the recorded signal, we can extract multiple parameters such as wing beat frequency, size, body to wing ratio, etc. The technique works at distances from 1 to 500 meters and allows unprecedented time resolution when studying insect behavior. The sensor is currently able to count as well as cluster and identify insects down to genus or even species level by using an automated intelligence system. We have shown that on a typical evening in Tanzania, 10 000 insect flight observations can be recorded in an hour. Our primary focus, as of late, has been concluding a recent bee field trial in Denmark – further demonstrating our ability to distinguish between honeybees and several Nordic bumblebee populations – as well as our continued work with oilseed rape insect populations.
O.31 Do plants eavesdrop on floral volatiles? Plant-plant communication beyond herbivore-induced volatiles
Jonas Kuppler1, Amy L Parachnowitsch2 1 Institute of Evolutionary Ecology & Conservation Genomics, Ulm University, Ulm, Germany 2 Department of Biology, University of New Brunswick, Fredericton, New Brunswick, Canada
Volatile organic compounds (VOC) emitted in response to herbivory can be detect by neighbouring plant individuals which may lead to phenotypic changes and increase herbivore resistance. For plant individuals, it could be also advantageous to detect other VOC such as floral scent. Detecting the presence of neighbouring flowering conspecifics could allow plant individuals to precisely synchronize flowering and increase attractiveness for pollinators and thus increasing the likelihood of cross-pollination. To test this idea, we explored if conspecific floral volatiles can induce phenotypic changes in flowers such as flower opening rate, nectar volume and sugar concentration using a diverse set of model organisms. We could show that exposure to flowering conspecifics resulted in phenotypic changes such as increased flower opening rate while plants exposed to floral scent of conspecifics showed no changes. Therefore, we discuss ecological implications for flowering plants of our findings and future directions.
Saturday 26 October, 7th session
O.32 What makes a good year for fruit set in an early-flowering lily?
James D. Thomson, Barbara A. Thomson Department of Ecology and Evolutionary Biology, University of Toronto, Canada
We have conducted supplemental pollinations in a single subalpine population of Erythronium grandiflorum for 26 years. Contrary to an earlier publication, there is no tendency for pollination to deteriorate over this period. Within seasons, early-flowering cohorts consistently have lower fruit set, due both to inadequate pollination by bumble bee queens and to frost damage. The best years are those that follow previous summers with plentiful rain and previous winters with deep snowpack that produce late springs.
O.35 Looking up for bees: pollination in smallholder legume crops, and the importance of trees in field margins
Sarah E J Arnold1, Filemon Elisante2, Yolice Tembo3, Prisila A. Mkenda2, Steven R Belmain1, Patrick A Ndakidemi1, Geoff M Gurr4, Philip C. Stevenson1,5 1 Natural Resources Institute, University of Greenwich, United Kingdom 2 Nelson Mandela African Institution of Science and Technology, Tanzania 3 Lilongwe University of Agriculture and Natural Resources, Malawi 4 Charles Sturt University, Australia 5 Royal Botanic Gardens, Kew, United Kingdom
Common bean (Phaseolus vulgaris) is an essential source of protein for many low-income households in Eastern and Southern Africa, and also benefits from insect pollination. We investigated the extent of its pollinator-dependence and the identity of its flower visitors on smallholder bean farms in Tanzania and Malawi. Using flower-visitor surveys we also evaluated the interactions between those potential and other plants growing on smallholder farms and particularly in the field margins. We anticipated that plant species richness on and around smallholder farms would positively influence abundance and diversity of flower visiting guilds and support increased visitation by key groups to bean flowers. Plants had around 25% more pods per plant and beans per pod when open pollinated compared to pollinator-excluded. Tanzanian fields in general had higher plant species richness and more complex flower visitor networks. Many of the most frequently visited field margin plants were non native species, though some have medicinal or natural pesticidal properties that may be beneficial to households. However, overall there was not a significant relationship between plant species richness on farms and most network parameters, flower visitor measures or visits to the crop flowers. However, the species richness of trees in and around bean fields was positively predictive of higher visitation rates to the crop. Trees on East African farms are often nectar-rich species and may provide a food source at times of year that the herb layer has been cut, burned or ploughed in. As the trees are often also leguminous species they may confer multiple benefits for smallholder farmers and wider biodiversity.
O.36 Diversified farming systems at field- and landscape scales for pollination in faba beans
Chloé Raderschall1, Ola Lundin1, Riccardo Bommarco1, Sandra Lindström2 1 SLU, Uppsala, Sweden 2 Hushållningssällskapet, Kristianstad, Sweden
The intensification of food production has led to transformation of natural habitats into agricultural fields. The resulting homogenisation of the landscape has been accompanied by a stark decline in pollinators, which provide an essential ecosystem service to agriculture. During my PhD I assess diversification strategies at different spatial scales for their potential to reverse the negative impacts of intensive agriculture on pollinator abundances and to increase pollination and yield. At landscape scale, we assessed if crop diversity (i.e. increasing the number and evenness of crops grown) benefits pollinators and faba bean pollination by providing more food and habitat resources. We selected 14 faba bean fields in Scania along independent gradients of crop diversity and proportion semi-natural habitat (SNH). We found higher pollinator densities with increasing SNH. Furthermore, insect pollinated plants produced 27% more yield, yet this pollination benefit was reduced with increasing SNH. In landscapes with higher crop diversity, pollinators were observed pollinating legitimately more often compared to robbing nectar or visiting extra floral nectaries. At field scale, we assessed the effects of annually sown flower strips and managed honey bee hives on wild pollinators, crop pollination and yield. Flower strips sown along fields can act as a resource bridge when flowering crops are scarce. Yet, a better understanding is needed of the dynamics of pollinators between flower strips and flowering focal crops such as faba beans. Similarly, added honeybee hives potentially augment pollination and yield but the impact of managed honey bees on wild pollinators requires further research. I will introduce our experimental design and discuss preliminary data on pollinator abundances in fields with/without flowers trips and/ or honey bee hives.
O.37 Grassland management for meadow birds in the Netherlands is unfavourable for pollinators
Lisette van Kolfschoten Naturalis Biodiversity Center, The Netherlands
Agricultural intensification and loss of semi-natural grassland have caused widespread loss of pollinator species in open habitats around the world. To reverse the decline, management regimes have been implemented, varying widely in effectiveness. In addition, the Netherlands has established nature reserves in which semi-natural grasslands are restored and are often managed for specific groups of species, e.g. meadow birds or plants. The effects of such measures on insect biodiversity are not well known. This study assesses the relationships between management regime, floral abundance and diversity and pollinator communities in three common semi-natural grassland management regimes: (1) wet hay meadows, (2) herb rich grasslands and (3) meadow bird grasslands. The results show that, first of all, meadow bird grasslands have lower pollinator abundance and diversity and a less unique pollinator assemblage than both other grassland-types. Second, flower abundance has a positive effect on pollinator abundance. Third, complete mowing has a strong negative effect on both pollinator abundance and species richness while partly mowing does not. These results show that meadow-bird grasslands are a comparatively unfavourable habitat for bees, hoverflies and butterflies but with applying a different mowing regime the imbalance could be remediated.
O.38 Does management at a local or landscape scale impact pollinator communities in semi-natural grasslands?
Michelle Larkin1, Dara A. Stanley2 1 Botany & Plant Science, Ryan Institute and School of Natural Sciences, National University of Ireland Galway, Ireland 2 School of Agriculture and Food Science, University College Dublin, Ireland
Agricultural intensification is one of the primary drivers of global pollinator decline, leading to concerns over the pollination of many wild flowers and commercially important crops. Agri-environmental schemes were introduced to mitigate biodiversity declines by offering financial incentives to farmers to protect and enhance biodiversity on their farms. However, these schemes have had varying levels of success as a) traditionally farmers are paid for applying measures regardless of environmental outcomes and b) schemes are often applied at a local scale when biodiversity conservation may need a wider landscape scale approach. A potential solution is to use results based agri-environment schemes where farmers are paid for conservation outcomes, and try to implement these schemes at a landscape scale. The aim of this study was to investigate local and landscape drivers of pollinator diversity in species rich grasslands in Ireland, using a results based agri-environment scheme (The Burren Programme) as a case study in the Burren region in the west of Ireland. Insect pollinators were sampled in 23 fields with varying conservation scores within the scheme in either high intensity (≥65% improved grassland) or low intensity (≥ 65% semi-natural grassland) landscapes using transects and pan traps. We found that local factors were driving bumblebee community composition while hoverfly and butterfly communities were influenced at the landscape scale. The results from this study will help understand the role of local and landscape factors in pollinator conservation and inform agri-environmental planning, and will advise whether there are any modifications to the existing Burren Programme which would benefit pollinators.
O.39 The cholinergic pesticide imidacloprid impairs motion-sensitive neurons in the pollinator fly Eristalis tenax
Elisa Rigosi, David C. O’Carroll Department of Biology, Lund University, Lund, Sweden
Insect pollinators are key species for natural ecosystems and the human economy. Common agrochemicals, such as imidacloprid (IMI), are currently major threats for the survival of pollinators both in rural and urban environments. This insecticide acts pharmacologically as an agonist of acetylcholine receptors (nAChR). The majority of studies on the sub-lethal effects of IMI conducted in the last decade have focused on bee species. Among other effects, bees exposed to sub-lethal doses of IMI revealed altered foraging behaviours and navigation. But while the impairment in foraging and flight activities during navigation might be affected by a disruption in visual processing, the effect of neonicotinoids on the visual system of pollinators has been largely overlooked. Nevertheless, immunostaining has revealed widespread nAChR expression in proximity to a well-known population of motion sensitive neurons in the third optic ganglia, the lobula plate tangential cells (LPTCs). We thus performed in vivo recording from LPTCs in a widely used model for insect visual neurophysiology, the hoverfly, Eristalis tenax, which is also an important pollinator in its own right. We performed electrophysiological recordings while the animal was constantly perfused and the brain haemolymph exposed to either 3.9 μM IMI or its vehicle. Presenting wide-field moving stimuli in different directions and with increasing contrast we found that IMI exposure caused increased spontaneous responses, impaired contrast sensitivity and reduced direction selectivity of LPTCs. These results provide new insights into the neurophysiological impact of cholinergic pesticides and open the door for employing the same experimental approach to other part of the visual system to unravel where along the visual processing these impairments occur.
O.40 Influence of spatio-temporal pollen availability on pollinators and their function in agricultural landscapes
Philipp Wolfgang Eckerter1, Lars Albus1, Farnaz Faramarzi1, Sharumathi Natarajan1, Erika Gobet2, Willy Tinner2, Christopher Bause3, Thomas Eltz3, Matthias Albrecht4, Felix Herzog4, Martin H. Entling1 1 iES Landau, Institute for Environmental Sciences, University of Koblenz-Landau, Landau/Pfalz, Germany 2 Institute of Plant Science and Oeschger Centre for Climate Change Research, University of Bern, Bern, Switzerland 3 Department of Animal Ecology, Evolution and Biodiversity, Ruhr-Universität Bochum, Bochum, Germany 4 Agricultural Landscapes and Biodiversity, Agroscope, Zürich, Switzerland
Wild pollinators depend on pollen and nectar for their survival and reproduction. Thus, a high availability of floral resources in crop and non-crop habitats is expected to enhance pollinators and their services to insect-pollinated crops. However, the specific use of floral resources and their availability at the landscape scale have rarely been considered in studies on crop pollination. In addition, while resources available before the focal crop are expected to boost beneficial insect densities and hence ecosystem services, flowers available simultaneously to the focal crop may attract beneficial insects away from it and thus reduce ecosystem services. To investigate effects of the type, amount and timing of alternative resources on pollinators and their services, we conducted experiments in 24 landscape around Landau (Germany) between 2016 to 2019. Pollen use over the cropping season was determined for Bombus terrestris, Osmia cornuta and Osmia bicornis, three important pollinators. We measured yield of Vicia faba phytometers, colony development of B. terrestris and colonization rate of Osmia nests along the landscape gradient. The diet of each pollinator species was dominated by a sequence of specific pollen types over the season. However, the analyses so far suggest that the availability of these dominant resources did not provide a strong explanation of pollinator density or fitness. Nevertheless, we found that higher pollen availability early in the season lead to more seeds per pod in Vicia faba. For Bombus terrestris, early pollen availability increased both with the amount of flowering crops and with the amount of semi-natural habitat. Overall, our results confirm that the timing of alternative floral resource availability is important for crop pollination services. However, the lack of a simple relationship of pollinators and the availability of specific resources complicates efforts to manage landscapes for pollination services.
O.45 Testing the role of floral neighborhood density and phenology on floral trait evolution
Kate Gallagher, Yuval Sapir Tel Aviv University, Israel
In Israel, the Royal Irises (Iris section Oncocyclus) come in all colors of the rainbow. These self-incompatible flowers rely primarily on pollination by male Eucera bees in a night-sheltering system. In this system, bees that sleep in irises emerge earlier in the morning to forage than ground-sleeping bees (Sapir et al., 2006). Therefore, we asked whether the strength of selection on floral traits could vary depending on the floral neighborhood density and phenology of a given focal plant? We hypothesized that if irises are rare in an area, bees may not exert strong selection on floral traits because they would rather sleep in any flower regardless of its size or color. However, if irises are common, as they would be during peak flowering or in dense clusters, then we might detect stronger selection on floral traits because bees have the opportunity to be more selective. Here we present results from our study testing this hypothesis, using the Iris petrana population in Yeruham. In 2019, we set up 40 plots using stratified random sampling based on iris population density. Throughout season, we selected two flowering focal plants per plot each week, and for each focal plant we measured floral traits including color and size, as well as the density of co-flowering irises and overall diversity of the co-flowering floral neighborhood. At the end of the season, we collected fruit and seed set data as measures of fitness. This study provides a novel insight to the evolution of flower color, a key trait in the interaction of plants with their environment, and in particular whether fine-scale temporal and spatial variation in selection on floral traits could be a mechanism maintaining continuous floral color polymorphism in the Royal Irises.
O.46 Different pollination approaches to compare seed set of diploid and tetraploid red clover (Trifolium pratense L.)
Shuxuan Jing, Birte Boelt, Per Kryger Aarhus University, Denmark
Tetraploid red clover offers high forage production performances, but the low seed yield is limiting the commercial exploitation. We investigated the causes of the low seed yield of tetraploid red clover by comparing diploid and tetraploid red clover at the level of pollination and seed set per floret. Different pollination approaches were studied covering hand pollination, honey bees in confined environment and bumble bees under open field conditions. In all experiment the plant material was cv. Rajah (2x) and cv. Amos (4x). We found that bumble bees (especially Bombus pascuorum) are the dominant pollinators in open field conditions with high abundance and floret-visit speed. In the cage study, surprisingly, similar or even higher frequencies of honey bees were observed visiting tetraploid red clover compared to diploid red clover. We further suggested that increasing honey bees to high numbers may decrease the seed set of red clover, due to the preference of honey bees visiting already tripped florets.
O.47 Changes in pollinator behaviour under different plant spatial aggregation
Jakub Štenc, Klára Koupilová, Zdeněk Janovský Department of Botany, Charles University, Czech Republic
Flowering plants are typically aggregated in space. Moreover, the individuals within a cluster of flowers are often closely related due to either clonal growth or limited seed dispersal. Pollinator foraging strategies differ among major pollinator functional groups and are likely to generate different visitation behaviour under different plant spatial aggregations. The main aim of this study was to investigate differences in visitation behaviour of pollinator functional groups under different levels of spatial aggregation of flowers. We observed pollinator behaviour in arrays of potted plants arranged in four clusters. We manipulated distance between clusters (near/far) and spacing of plants within clusters (loose/dense). Five common meadow species were used for observations in order to cover all major pollinator groups. Butterflies and hymenopterans were most prone to fly between clusters and on average every fifth or seventh visited plant was from a different cluster than the previous ones and probability of flying between the clusters increased, if the clusters were near to each other. On the other hand, syrphids and other dipterans, tended to fly between clusters much less (on average only every twentieth or tenth visited plant, respectively). Probability of syrphids and other diptera flying between plant clusters increased if the plants in clusters were loosely spaced, but did not differ with the distance of clusters from each other. Our results indicate that pollinator functional groups differ not only in their preferences for flowers and carry-over effectiveness, but also in probability of carrying pollen outside the local plant cluster. This could have profound impacts on plant mating structure and also reproductive success of self-incompatible plant species.
O.48 The ecological significance of phenotypic differentiation in floral attraction traits in populations of Eruca sativa in Israel
Oz Barazani1, Sharoni Shafir2 1 Institute of Plant Sciences, Agricultural Research Organization, Israel 2 B. Triwaks Bee Research Center, Department of Entomology, The Hebrew University of Jerusalem, Faculty of Agriculture, Food, and the Environment, Rehovot, Israel
Plants of Eruca sativa (Brassicaceae) from desert and Mediterranean populations in Israel differ in flower color and size. In the conspecific desert habitat, the population has higher abundance of flowers with cream color and longer petals, whereas in the Mediterranean habitat, with a more heterogeneous annual plant community, the population has higher abundance of flowers with yellow and shorter petals. Choice experiments with honey bees, the main pollinator in the natural habitat in Israel, confirmed our hypothesis that they are more attracted to the yellow flower morph than to the cream one. In addition, a proboscis extension response test indicated that honey bees are able to discriminate between flower scents of the two morphs. The advantage of the yellow color morph in attracting pollinators may explain its dominance among plants of the Mediterranean population, where plants are more exposed to inter-specific competition over pollinators.
O.49 Flowers respond to pollinator sound within minutes by increasing nectar sugar concentration
Lilach Hadany, Marine Veits, Itzhak Khait, Uri Obolski, Eyal Zinger, Arjan Boonman, Aya Goldshtein, Kfir Saban, Rya Seltzer, Udi Ben‐Dor, Paz Estlein, Areej Kabat, Dor Peretz, Ittai Ratzersdorfer, Slava Krylov, Daniel Chamovitz, Yuval Sapir, Yossi Yovel Tel Aviv University, Israel
Can plants hear their pollinators and respond to them rapidly? We show that Oenothera drummondii flowers, exposed to playback sound of a pollinator or to synthetic sound signals at similar frequencies, produce sweeter nectar within 3 min, potentially increasing the chances of cross pollination. We found that the flowers vibrated mechanically in response to these sounds, suggesting a plausible mechanism where the flower serves as an auditory sensory organ. Both the vibration and the nectar response were frequency‐specific: the flowers responded and vibrated to pollinator sounds, but not to higher frequency sound. Our results document for the first time that plants can rapidly respond to pollinator sounds in an ecologically relevant way. Potential implications include plant resource allocation, the evolution of flower shape and the evolution of pollinators sound. Finally, our results suggest that plants may be affected by other sounds as well, including anthropogenic ones.
O.50 Pollinator mediated evolution of floral Traits in Digitalis purpurea after range expansion
Christopher R Mackin1, Balfour NJ, Peña JF, Blanco MA, Castellanos MC 1 University of Sussex, United Kingdom
Evolution of novel floral form can occur when plants experience a change in their pollinators, such as when they extend their range into novel environments. Understanding how variation in the pollinator community visiting a plant and exerting natural selection on its floral traits can help us understand floral evolution and could help predict how plants will respond to environmental change. This study investigates floral variation and patterns of natural selection in the biennial herb Digitalis purpurea (common foxglove) by comparing native populations in the UK with naturalised populations in tropical mountains in Colombia and Costa Rica. Pollinator censuses in 11 populations and tests for the pollinating ability of each species indicate differences in the pollinator assemblage between continents, including visitation by hummingbirds and long-tongued bumblebees. In parallel, we found variation in floral morphology, particularly in the size of the proximal portion of the corolla tube, where a constriction restricts access to nectar to visitors with long mouth parts. We found consistent directional selection for larger proximal corolla tubes in the naturalised non-native populations, while no selection on corollas (directional, or otherwise) was present in two native European localities. We discuss potential explanations for these patterns in the light of adaptation to new pollinator environments.
O.51 Pollinator behavior and resource limitation maintain honest floral signalling
Anina Knauer Agroscope, Switzerland
In many communication systems, signal-receivers profit from honest signals that indicate the signaller’s quality, whereas low quality signallers should profit from cheating. Under such a conflict of interests between signallers and signal-receivers, the maintenance of honest signals presents a puzzle. In theory, honesty can represent an optimal strategy or be maintained by a constraint, but the actual mechanisms have been studied in few systems only. Here, we investigate honest signalling in a plant species, Brassica rapa, that advertises nectar amounts to pollinators by two honest floral signals; corolla size and the floral volatile phenylacetaldehyde. In a series of seven experiments we tested for physiological constraints and pollinator behaviors related to honest floral signals and rewards. While honest floral signals were associated with pollinator attraction, bees’ flower visitation time depended on nectar amounts and increased the number of seeds that flowers developed. Further, honest floral signals and the seed set after hand pollination both increased after soil fertilization indicating nutrient limitation in the traits and a potential trade-off in resource allocation. Finally, by incorporating these results into a mathematical model, we showed that honest signalling in B. rapa is maintained by a combination of pollinator behavior and resource limitation causing differential benefits of nectar production.
O.52 Phenotypic selection on floral scent in deceptive Arum maculatum
Eva Gfrerer1, Danae Laina1, Marc Gibernau2, Anja C Hörger1, Hans-Peter Comes1, Stefan Dötterl1 1 University of Salzburg, Department of Biosciences, Salzburg, Austria 2 Université de Corse Pascal Paoli, Laboratoire des Sciences Pour l’Environnement, Ajaccio, Corsica
Floral scent is a key mediator in plant-pollinator interactions, and as such sensitive to selection imposed by its pollinators. When environmental conditions, such as the pollinator community, differ in the range of the plant´s distribution, the selection pressures might vary geographically. By chemically mimicking an oviposition site, the common woodland plant Arum maculatum L. (Araceae) attracts and deceives its pollinating moth flies (Psychodidae) with its floral scents. The species composition and sex ratio of the attracted flies exhibit a strong geographical pattern north vs. south of the Alps, and the different fly species are known to have different olfactory preferences. Thus, our goals are i) to assess if floral scent in A. maculatum differs among populations north and south of the Alps, ii) if floral scent is under selection, and iii) if this selection is geographically structured. To address these goals, we collected floral scent and fruit/seed set from a total of 240 plants from six populations north and five populations south of the Alps. We show that floral scents of Arum maculatum are highly complex and differ within and among populations north and south of the Alps. Further, we will present results on correlations between floral scent and fitness data, either considering or not the regional context. Overall, this is one of the few studies that tested for β-selection on floral scents and contributes to our understanding of intraspecific variation in floral signals.
O.53 Don’t forget about the flies! Public perceptions around pollinator conservation in Ireland
Katherine L.W. Burns1,2, Úna Fitzpatrick3, Dara A. Stanley1,2 1 School of Agriculture and Food Science, University College Dublin, Belfield, Dublin 2, Ireland 2 Earth Institute, University College Dublin, Belfield, Dublin 2, Ireland 3 National Biodiversity Data Centre, Beechfield House, Waterford Institute of Technology West Campus, Carriganore, Co. Waterford, Ireland
Media coverage of pollinator decline has resulted in a growing amount of public concern and interest in the topic of pollinator conservation. This emerging interest in pollinators presents an opportunity for future citizen involvement in the protection and promotion of global pollinator populations. However, despite a growing interest in pollinator conservation, there may be a lack of public understanding about the importance and identification of insect pollinators, especially with relation to non-honeybees. In an effort to support current conservation initiatives and action plans for pollinator conservation, and further inform new educational materials and engagement strategies, we used Ireland as a case study to determine how pollinators, pollination services, and pollinator decline are currently perceived by Irish citizens. We designed and distributed a citizen survey, “What's the Buzz? Public Views of Pollinating Insects in Ireland,” to 613 participants to determine perceptions and understanding around the identification, importance, decline, and conservation of Irish pollinators. Our findings indicate that the Irish public is aware that pollinators are declining and understand the main causes, and most participants indicated that they are already carrying out actions to protect pollinators. However, the majority of survey participants underestimated the number of bee species in Ireland and were unable to identify certain common pollinators, such as flies and solitary bees. Many were also unaware of the importance of non-bee pollinators, such as flies and wasps, to the pollination of Irish crops and wildflowers. Our findings indicate that future educational measures should highlight the identification of solitary bees and non-bee pollinators and their importance to pollination services, to ensure a holistic approach to pollinator conservation.
O.54 Pollinator preference for intermediate floral size in a night-sheltering pollination system is and its association with heat reward
Yuval Sapir1, M. Kate Gallagher1, Omer Bar1, Ilan Koren2 1 Tel Aviv University, Israel 2 Weizmann Institute of Science, Israel
Animal pollinators exert selective pressure on plants they pollinate. Pollinator-mediated selection is typically associated with traits that increase the visibility, attractiveness and utility of flowers to their pollinators. Although Oncocyclus irises are large and showy, they do not produce a nectar reward. They are, however, completely self-incompatible and depend on night-sheltering Eucera bees for pollination. This night-sheltering system is putatively associated with a heat reward, which enables bees that sleep in the pollination tunnels to emerge earlier in the morning than their ground-nesting counterparts. In this study, we used artificial iris flowers in six size classes to ask (1) whether pollinators select on flower size in irises and (2) if so, do their choices among flowers of different sizes correspond with the potential heat reward? We found that pollinators prefer intermediate-sized flowers and that intermediate-sized flowers warm faster at sunrise than other flower sizes. These data suggest that pollinator preference for intermediate sized flowers may be associated with the heat reward offered by these dark-colored irises to night-sheltering bees. This work offers new avenues of research using artificial flowers in natural systems to disentangle the traits under pollinator-mediated selection, and specifically whether a preference for intermediate sized flowers, in this system, is a learned preference based on the heat reward.