Effects of community composition on plant-pollinator interaction networks across a spatial gradient of oak-savanna habitats.

Distance between habitats may impact the composition and corresponding interactions between trophic levels. Mutualistic networks, such as those of plants and pollinators tend to have a core set of properties that often relate to the resilience of the community, or the ability of the community to retain function and structure after a disturbance. Furthermore, network structure is highly dependent on the number of specialists and generalists; however, it is unclear how different groups of species with various life-history strategies influence network structure. In this study, we evaluated how the composition of plants and pollinators within 16 oak-savanna sites changed across a latitudinal gradient. In addition, we evaluated how the abundance of different groups of plants and pollinators affected network metrics related to resilience. We found that the composition of plants and pollinators varied between ecoregions, while pollinator composition further varied with habitat characteristics. Network metrics displayed no spatial pattern but were related to the abundance of several pollinator groups. Above-ground nesting insects had a positive relationship with nestedness and a negative relationship with modularity, while predatory larvae had a negative relationship with modularity. Thus, above-ground nesting insects and predatory larvae could be expected to increase network resilience. This study emphasizes how spatial scales can influence species compositions, which in turn affects the structure of interactions in the community with implications for resilience.

Global biogeography of mating system variation in seed plants.

Latitudinal gradients in biotic interactions have been suggested as causes of global patterns of biodiversity and phenotypic variation. Plant biologists have long speculated that outcrossing mating systems are more common at low than high latitudes owing to a greater predictability of plant-pollinator interactions in the tropics; however, these ideas have not previously been tested. Here, we present the first global biogeographic analysis of plant mating systems based on 624 published studies from 492 taxa. We found a weak decline in outcrossing rate towards higher latitudes and among some biomes, but no biogeographic patterns in the frequency of self-incompatibility. Incorporating life history and growth form into biogeographic analyses reduced or eliminated the importance of latitude and biome in predicting outcrossing or self-incompatibility. Our results suggest that biogeographic patterns in mating system are more likely a reflection of the frequency of life forms across latitudes rather than the strength of plant-pollinator interactions.

A global synthesis of the effects of diversified farming systems on arthropod diversity within fields and across agricultural landscapes.

Agricultural intensification is a leading cause of global biodiversity loss, which can reduce the provisioning of ecosystem services in managed ecosystems. Organic farming and plant diversification are farm management schemes that may mitigate potential ecological harm by increasing species richness and boosting related ecosystem services to agroecosystems. What remains unclear is the extent to which farm management schemes affect biodiversity components other than species richness, and whether impacts differ across spatial scales and landscape contexts. Using a global metadataset, we quantified the effects of organic farming and plant diversification on abundance, local diversity (communities within fields), and regional diversity (communities across fields) of arthropod pollinators, predators, herbivores, and detritivores. Both organic farming and higher in-field plant diversity enhanced arthropod abundance, particularly for rare taxa. This resulted in increased richness but decreased evenness. While these responses were stronger at local relative to regional scales, richness and abundance increased at both scales, and richness on farms embedded in complex relative to simple landscapes. Overall, both organic farming and in-field plant diversification exerted the strongest effects on pollinators and predators, suggesting these management schemes can facilitate ecosystem service providers without augmenting herbivore (pest) populations. Our results suggest that organic farming and plant diversification promote diverse arthropod metacommunities that may provide temporal and spatial stability of ecosystem service provisioning. Conserving diverse plant and arthropod communities in farming systems therefore requires sustainable practices that operate both within fields and across landscapes.

Traits and phylogenetic history contribute to network structure across Canadian plant-pollinator communities.

Interaction webs, or networks, define how the members of two or more trophic levels interact. However, the traits that mediate network structure have not been widely investigated. Generally, the mechanism that determines plant-pollinator partnerships is thought to involve the matching of a suite of species traits (such as abundance, phenology, morphology) between trophic levels. These traits are often unknown or hard to measure, but may reflect phylogenetic history. We asked whether morphological traits or phylogenetic history were more important in mediating network structure in mutualistic plant-pollinator interaction networks from Western Canada. At the plant species level, sexual system, growth form, and flower symmetry were the most important traits. For example species with radially symmetrical flowers had more connections within their modules (a subset of species that interact more among one another than outside of the module) than species with bilaterally symmetrical flowers. At the pollinator species level, social species had more connections within and among modules. In addition, larger pollinators tended to be more specialized. As traits mediate interactions and have a phylogenetic signal, we found that phylogenetically close species tend to interact with a similar set of species. At the network level, patterns were weak, but we found increasing functional trait and phylogenetic diversity of plants associated with increased weighted nestedness. These results provide evidence that both specific traits and phylogenetic history can contribute to the nature of mutualistic interactions within networks, but they explain less variation between networks.

Nutrients from spawning salmon influence leaf area, tissue density, and nitrogen-15 in riparian plant leaves.

Nutrient subsidies have significant impacts on ecosystems by connecting disjunct habitats, often through long-distance animal migrations. Salmon migrations on the North Pacific coasts provide these kinds of nutrient subsidies from senescent fish at the end of their life cycle, which can have significant ecological effects on terrestrial species. This can include impacts on individuals, populations, and communities, where shifts in community composition towards plant species that indicate nitrogen-rich soils have been documented. We investigated the effects of variation in salmon spawning density on the leaf traits of four common riparian plant species on the central coast of British Columbia, Canada. We found that all plant species had higher foliar salmon-derived nitrogen on streams with a higher spawning density. Three of the four species had larger leaves, and one species also had higher leaf mass per area on streams with more salmon. However, we found no differences in leaf greenness or foliar percent nitrogen among our study streams. These results demonstrate that nutrient subsidies from spawning salmon can have significant impacts on the ecology, morphology, and physiology of riparian plants, which lends support to a mechanism by which certain plants are more common on productive salmon streams.

A global quantitative synthesis of local and landscape effects on wild bee pollinators in agroecosystems.

Bees provide essential pollination services that are potentially affected both by local farm management and the surrounding landscape. To better understand these different factors, we modelled the relative effects of landscape composition (nesting and floral resources within foraging distances), landscape configuration (patch shape, interpatch connectivity and habitat aggregation) and farm management (organic vs. conventional and local-scale field diversity), and their interactions, on wild bee abundance and richness for 39 crop systems globally. Bee abundance and richness were higher in diversified and organic fields and in landscapes comprising more high-quality habitats; bee richness on conventional fields with low diversity benefited most from high-quality surrounding land cover. Landscape configuration effects were weak. Bee responses varied slightly by biome. Our synthesis reveals that pollinator persistence will depend on both the maintenance of high-quality habitats around farms and on local management practices that may offset impacts of intensive monoculture agriculture.

A data set for pollinator diversity and their interactions with plants in the Pacific Northwest.

Pollinator populations have declined substantially in recent years. The resulting loss in pollination services has both ecological and economic consequences, including reductions in plant diversity and crop production and lower food security. Data sets that identify pollinators and their plant hosts are of the utmost importance for the light can shed on the main causes of pollinator declines. Here we present a data set that contains 67,954 individual pollinator records. The data were collected across the Pacific Northwest, primarily focused in British Columbia (Canada), with 182 individual sites over 11 years, between 2005 and 2017. This data set comprises multiple studies that aimed to collect information on pollinator abundance, diversity, and their interactions with plants. Overall, the data set includes 937 morphospecies (of which 482 were identified to the species level) of pollinators across 105 families, including data for bees, wasps, butterflies, moths, and flies. We also present information on the interactions of these species, with 473 species of plants. The data set is being released for noncommercial use only. Credits should be given to this paper (i.e., proper citation).

CropPol: A dynamic, open and global database on crop pollination.

Seventy five percent of the world's food crops benefit from insect pollination. Hence, there has been increased interest in how global change drivers impact this critical ecosystem service. Because standardized data on crop pollination are rarely available, we are limited in our capacity to understand the variation in pollination benefits to crop yield, as well as to anticipate changes in this service, develop predictions, and inform management actions. Here, we present CropPol, a dynamic, open, and global database on crop pollination. It contains measurements recorded from 202 crop studies, covering 3,394 field observations, 2,552 yield measurements (i.e., berry mass, number of fruits, and fruit density [kg/ha], among others), and 47,752 insect records from 48 commercial crops distributed around the globe. CropPol comprises 32 of the 87 leading global crops and commodities that are pollinator dependent. Malus domestica is the most represented crop (32 studies), followed by Brassica napus (22 studies), Vaccinium corymbosum (13 studies), and Citrullus lanatus (12 studies). The most abundant pollinator guilds recorded are honey bees (34.22% counts), bumblebees (19.19%), flies other than Syrphidae and Bombyliidae (13.18%), other wild bees (13.13%), beetles (10.97%), Syrphidae (4.87%), and Bombyliidae (0.05%). Locations comprise 34 countries distributed among Europe (76 studies), North America (60), Latin America and the Caribbean (29), Asia (20), Oceania (10), and Africa (7). Sampling spans three decades and is concentrated on 2001-2005 (21 studies), 2006-2010 (40), 2011-2015 (88), and 2016-2020 (50). This is the most comprehensive open global data set on measurements of crop flower visitors, crop pollinators and pollination to date, and we encourage researchers to add more datasets to this database in the future. This data set is released for non-commercial use only. Credits should be given to this paper (i.e., proper citation), and the products generated with this database should be shared under the same license terms (CC BY-NC-SA).

Investigating bee dietary preferences along a gradient of floral resources: how does resource use align with resource availability?

Bee dietary preferences, or the floral resources that they consistently collect, likely impact where a species can persist. For this reason it is likely that bee dietary preferences are dependent upon the composition of the plant community. In this study, we evaluated floral visits and pollen loads of the mining bee, Andrena angustitarsata Viereck, across a 630 km north-south range to understand dietary preferences along a floral resource gradient. Previous research, in a more geographically limited area, suggested this species was an eclectic oligolege on predominantly Apiaceae and in part Rosaceae. In the present study we found the species predominately visited and collected pollen from Apiaceae and Rosaceae, but visited 12 flower families and collected pollen from 32, distinguishing them as generalist foragers. The frequency of Apiaceae pollen on the bees and the species-level specialization index (a measure of visit specialization) were higher in regions with higher Apiaceae abundance. In addition Apiaceae and Rosaceae were the only plant families significantly preferred for pollen collection, regardless of floral abundance. We conclude that across our study region A. angustitarsata has a generalist dietary breadth, but also has dietary preference for Apiaceae and Rosaceae. Our study indicates that while bees may overall make generalist foraging decisions they may still prefer and likely benefit from selecting fewer flower taxa.

Network robustness and structure depend on the phenological characteristics of plants and pollinators.

Many structural patterns have been found to be important for the stability and robustness of mutualistic plant-pollinator networks. These structural patterns are impacted by a suite of variables, including species traits, species abundances, their spatial configuration, and their phylogenetic history. Here, we consider a specific trait: phenology, or the timing of life history events. We expect that timing and duration of activity of pollinators, or of flowering in plants, could greatly affect the species' roles within networks in which they are embedded. Using plant-pollinator networks from 33 sites in southern British Columbia, Canada, we asked (a) how phenological species traits, specifically timing of first appearance in the network and duration of activity in a network, were related to species' roles within a network, and (b) how those traits affected network robustness to phenologically biased species loss. We found that long duration of activity increased connection within modules for both pollinators and plants and among modules for plants. We also found that date of first appearance was positively related to interaction strength asymmetry in plants but negatively related to pollinators. Networks were generally more robust to the loss of pollinators than plants, and robustness increased if the models allow new interactions to form when old ones are lost, constrained by overlapping phenology of plants and pollinators. Robustness declined with the loss of late-flowering plants, which tended to have higher interaction strength asymmetry. In addition, robustness declined with loss of early-flying or long-duration pollinators. These pollinators tended to be among-module connectors. Our results point to networks being limited by early-flying pollinators. If plants flower earlier due to climate change, plant fitness may decline as they will depend on early emerging pollinators, unless pollinators also emerge earlier.

Correlated evolution of mating system and floral display traits in flowering plants and its implications for the distribution of mating system variation.

Reduced allocation to structures for pollinator attraction is predicted in selfing species. We explored the association between outcrossing and floral display in a broad sample of angiosperms. We used the demonstrated relationship to test for bias against selfing species in the outcrossing rate distribution, the shape of which has relevance for the stability of mixed mating. Relationships between outcrossing rate, flower size, flower number and floral display, measured as the product of flower size and number, were examined using phylogenetically independent contrasts. The distribution of floral displays among species in the outcrossing rate database was compared with that of a random sample of the same flora. The outcrossing rate was positively associated with the product of flower size and number; individually, components of display were less strongly related to outcrossing. Compared with a random sample, species in the outcrossing rate database showed a deficit of small floral display sizes. We found broad support for reduced allocation to attraction in selfing species. We suggest that covariation between mating systems and total allocation to attraction can explain the deviation from expected trade-offs between flower size and number. Our results suggest a bias against estimating outcrossing rates in the lower half of the distribution, but not specifically against highly selfing species.

Conservation of a rare plant requires different methods in different habitats: demographic lessons from Actaea elata.

Understanding species decline and conserving endangered species requires demographic information, and variation in the environment may affect demography. Actaea elata is a globally rare, perennial herb found in a range of Pacific Northwest forest stand types that differ in canopy openness. Canopy openness increases reproductive output in this species and so was expected to have demographic impact. We performed a demographic analysis of A. elata in contrasting forest stands (broadleaved vs. coniferous) over two annual intervals, and predicted that population growth rate would be higher in the open-canopy broadleaved stand. Population growth was determined using stage-based matrix models, and the most influential transitions were identified using elasticity analyses. The finite rate of population increase (λ) was lower for the two transition periods at the broadleaved stand than at the coniferous stand (λ = 0.86 and 0.87 vs. 0.94 and 0.98), even though the former population was more fecund. The decline in the broadleaved stand reflects greater mortality and retrogression to previous stages, partly as a consequence of herbivory. In contrast, lower recruitment occurred in the coniferous stand, but there was also less mortality and retrogression. Our results suggest that management decisions for conservation of A. elata should be tailored to differing habitats, with a focus on preventing mortality in some populations and increasing recruitment in others.

Correlations among fertility components can maintain mixed mating in plants.

Classical models studying the evolution of self-fertilization in plants conclude that only complete selfing and complete outcrossing are evolutionarily stable. In contrast with this prediction, 42% of seed-plant species are reported to have rates of self-fertilization between 0.2 and 0.8. We propose that many previous models fail to predict intermediate selfing rates because they do not allow for functional relationships among three components of reproductive fitness: self-fertilized ovules, outcrossed ovules, and ovules sired by successful pollen export. Because the optimal design for fertility components may differ, conflicts among the alternative pathways to fitness are possible, and the greatest fertility may be achieved with some self-fertilization. Here we develop and analyze a model to predict optimal selfing rates that includes a range of possible relationships among the three components of reproductive fitness, as well as the effects of evolving inbreeding depression caused by deleterious mutations and of selection on total seed number. We demonstrate that intermediate selfing is optimal for a wide variety of relationships among fitness components and that inbreeding depression is not a good predictor of selfing-rate evolution. Functional relationships subsume the myriad effects of individual plant traits and thus offer a more general and simpler perspective on mating system evolution.

Disruption of a belowground mutualism alters interactions between plants and their floral visitors.

Plants engage in diverse and intimate interactions with unrelated taxa. For example, aboveground floral visitors provide pollination services, while belowground arbuscular mycorrhizal fungi (AMF) enhance nutrient capture. Traditionally in ecology, these processes were studied in isolation, reinforcing the prevailing assumption that these above- and belowground processes were also functionally distinct. More recently, there has been a growing realization that the soil surface is not a barrier to many ecological interactions, particularly those involving plants (who live simultaneously above and below ground). Because of the potentially large impact that mycorrhizae and floral visitors can have on plant performance and community dynamics, we designed an experiment to test whether these multi-species mutualisms were interdependent under field conditions. Using benomyl, a widely used fungicide, we suppressed AMF in a native grassland, measuring plant, fungal, and floral-visitor responses after three years of fungal suppression. AMF suppression caused a shift in the community of floral visitors from large-bodied bees to small-bodied bees and flies, and reduced the total number of floral visits per flowering stem 67% across the 23 flowering species found in the plots. Fungal suppression has species-specific effects on floral visits for the six most common flowering plants in this experiment. Exploratory analyses suggest these results were due to changes in floral-visitor behavior due to altered patch-level floral display, rather than through direct effects of AMF suppression on floral morphology. Our findings indicate that AMF are an important, and overlooked, driver of floral-visitor community structure with the potential to affect pollination services. These results support the growing body of research indicating that interactions among ecological interactions can be of meaningful effect size under natural field conditions and may influence individual performance, population dynamics, and community structure.

Pollen Foraging Differences Among Three Managed Pollinators in the Highbush Blueberry (Vaccinium corymbosum) Agroecosystem.

Highbush blueberry, Vaccinium corymbosum (Gray), production in British Columbia is dependent upon insect pollination for fruit yield with particular cultivars demonstrating low yields due to poor pollination. New managed species of pollinators are being developed to provide farmers with managed pollinator options beyond Apis mellifera (Linnaeus). Pollinators in highbush blueberry agricultural systems encounter a variety of nontarget floral resources that may affect the pollination received by the crop. Our study analyzed the differences in pollen foraging of honey bees and two species of managed bumblebees across nine farm sites. Corbicular pollen loads from pollen foraging workers were removed and identified to the lowest taxonomic level possible. Of the three managed pollinators, the corbicular pollen loads of Bombus huntii (Greene) contained the most blueberry pollen (52.1%), three times as much as the two other managed bee species. Fifteen morphotypes of pollen were identified from all foraging workers with Rosaceae being the most frequently gathered overall pollen type (n = 74). The noncrop pollen identified in our samples derived from plant species not common as weedy species in the agroecosystem suggesting that floral resource diversity outside of the farm boundaries is important to pollinators. The three managed species in our blueberry fields utilized floral resources differentially underscoring the importance of pollinator species' characteristics and large-scale floral resource landscape in developing new managed pollinators and pollination strategies.

Hummingbirds and bumble bees exposed to neonicotinoid and organophosphate insecticides in the Fraser Valley, British Columbia, Canada.

To measure exposure to neonicotinoid and other pesticides in avian pollinators, we made novel use of cloacal fluid and fecal pellets from rufous (Selasphorus rufus) and Anna's (Calypte anna) hummingbirds living near blueberry fields in the Fraser River Valley and Vancouver Island, British Columbia, Canada. To examine on-farm exposure to pesticides in invertebrate pollinators, we also collected bumble bees native to Canada (Bombus mixtus, Bombus flavifrons, and Bombus melanopygus), their pollen, and blueberry leaves and flowers from within conventionally sprayed and organic blueberry farms. By sites and sample type, the results reported in the present study represent pooled samples (n = 1). In 2015 to 2016, the combined concentration of the neonicotinoid insecticides imidacloprid, thiamethoxam, and clothianidin detected in hummingbird cloacal fluid from sites near conventionally sprayed blueberry fields was 3.63 ng/mL (ppb). Among the 18 compounds measured in fecal pellets, including one neonicotinoid (imidacloprid), only piperonyl butoxide was detected (1.47-5.96 ng/g). Piperonyl butoxide is a cytochrome P450 inhibitor applied with some insecticides to increase their toxic efficacy. Only diazinon was detected in bumble bees (0.197 ng/g), whereas diazinon (1.54-1.7 ng/g) and imidacloprid (up to 18.4 ng/g) were detected in pollen collected from bumble bees including the bees from organic sites located near conventionally sprayed blueberry farms. Imidacloprid was also detected at 5.16 ng/g in blueberry flowers collected 1 yr post spray from 1 of 6 conventionally sprayed blueberry farms. Environ Toxicol Chem 2018;37:2143-2152. © 2018 SETAC.

Contrasting Pollinators and Pollination in Native and Non-Native Regions of Highbush Blueberry Production.

Highbush blueberry yields are dependent on pollination by bees, and introduction of managed honey bees is the primary strategy used for pollination of this crop. Complementary pollination services are also provided by wild bees, yet highbush blueberry is increasingly grown in regions outside its native range where wild bee communities may be less adapted to the crop and growers may still be testing appropriate honey bee stocking densities. To contrast crop pollination in native and non-native production regions, we sampled commercial 'Bluecrop' blueberry fields in British Columbia and Michigan with grower-selected honey bee stocking rates (0-39.5 hives per ha) to compare bee visitors to blueberry flowers, pollination and yield deficits, and how those vary with local- and landscape-scale factors. Observed and Chao-1 estimated species richness, as well as Shannon diversity of wild bees visiting blueberries were significantly higher in Michigan where the crop is within its native range. The regional bee communities were also significantly different, with Michigan farms having greater dissimilarity than British Columbia. Blueberry fields in British Columbia had fewer visits by honey bees than those in Michigan, irrespective of stocking rate, and they also had lower berry weights and a significant pollination deficit. In British Columbia, pollination service increased with abundance of wild bumble bees, whereas in Michigan the abundance of honey bees was the primary predictor of pollination. The proportion of semi-natural habitat at local and landscape scales was positively correlated with wild bee abundance in both regions. Wild bee abundance declined significantly with distance from natural borders in Michigan, but not in British Columbia where large-bodied bumble bees dominated the wild bee community. Our results highlight the varying dependence of crop production on different types of bees and reveal that strategies for pollination improvement in the same crop can vary greatly across production regions.

Delivery of crop pollination services is an insufficient argument for wild pollinator conservation.

There is compelling evidence that more diverse ecosystems deliver greater benefits to people, and these ecosystem services have become a key argument for biodiversity conservation. However, it is unclear how much biodiversity is needed to deliver ecosystem services in a cost-effective way. Here we show that, while the contribution of wild bees to crop production is significant, service delivery is restricted to a limited subset of all known bee species. Across crops, years and biogeographical regions, crop-visiting wild bee communities are dominated by a small number of common species, and threatened species are rarely observed on crops. Dominant crop pollinators persist under agricultural expansion and many are easily enhanced by simple conservation measures, suggesting that cost-effective management strategies to promote crop pollination should target a different set of species than management strategies to promote threatened bees. Conserving the biological diversity of bees therefore requires more than just ecosystem-service-based arguments.

Costs of glandular trichomes in Datura wrightii: a three-year study.

Models accounting for genetic variation for resistance to herbivores within plant populations often postulate a balance between the costs of that resistance and its benefits. The production of glandular trichomes by Datura wrightii was shown to be costly in a previous one-year study because plants producing glandular trichomes (sticky plants), a factor conferring resistance to some insect herbivores, also produced 45% fewer seeds than plants producing nonglandular trichomes (velvety plants) when grown in a common garden. Because sticky plants tended to be larger than velvety plants but produced fewer seed capsules, we postulated an allocation trade-off in which velvety plants are more reproduction-dominated whereas sticky plants are more growth-dominated. If a greater commitment to vegetative growth eventually allows sticky plants to compensate for reduced seed production, we would expect a reduction or elimination of the cost of resistance over time in this perennial plant. We monitored growth, survival, and seed production of plants from defined crosses of local populations for three years in a common garden when exposed to and protected from herbivores, and with and without supplemental water. The majority of plants exposed to herbivores had died by the end of the study. We used standard life-table methods to determine the net reproductive rate (R0) and the finite rate of increase (lambda) of plants of each trichome type. After three years, when plants were protected from herbivores, sticky plants were 187-245% larger than velvety plants, depending upon irrigation treatment, but sticky plants continued to be less efficient in producing seeds per unit of canopy volume. Even though the total seed production of sticky plants eventually equaled that of velvety plants, the advantage of earlier reproduction by velvety plants increased lambda by 55-230% over that of sticky plants, depending upon herbivore and irrigation treatment. Exposure to herbivores reduced lambda by 69-83%, depending upon plant type and irrigation treatment, whereas supplemental irrigation increased lambda by 29-175%, depending upon plant type and exposure to herbivores. Although there was a large allocation trade-off between growth and reproduction, the benefits of such a trade-off did not emerge before most plants were killed by herbivores. The cost of producing glandular trichomes strictly for herbivore resistance continued to exceed its benefits, and in the absence of other, unmeasured benefits from the suite of life-history characters associated with glandular trichome production, natural selection is expected to eliminate this costly resistance trait from D. wrightii populations.

Sex Allocation and Reproductive Success in the Andromonoecious Perennial Solanum carolinense (Solanaceae). II. Paternity and Functional Gender.

According to Bateman's principle, male fitness in entomophilous plant species should be limited by mating opportunity, which is influenced by the size or number of flowers. We determined male-specific fitness consequences of floral phenotype in andromonoecious Solanum carolinense, examined the relationship between male and female reproductive success within plants, and evaluated the distribution of functional gender among plants. A maximum likelihood-based paternity analysis, based on multilocus allozyme phenotypes of parents and offspring from four experimental plots, was used to determine male reproductive success and its relationship to floral phenotype. Male success was enhanced by an increase in the proportion of male flowers produced but not by an increase in total flower number, even though all flowers contain male parts. Larger flower size increased male success in only one plot. Male and female reproductive success were negatively correlated, and plants varied in functional gender from completely female to completely male. This gender specialization may occur because hermaphroditic and male flowers differ in their ability to contribute to male and female success. Although sex allocation theory predicts a positive relationship between the size or number of plant parts and reproductive success, this study indicates that aspects of floral morphology that affect gender specialization should also be considered.