Reduced floral fidelity translates functionally into less conspecific pollen carried by bees and transferred between individual plants of the same species, which was ultimately reflected in decreased plant seed production in our manipulated sites. Thus, our results show that the ecosystem functional contributions of bee species in our system are not fixed, but instead are dynamic and dependent on interactions with competing species. These findings highlight a new mechanism for how biodiversity can shape ecosystem services and functions. Complementarity—in which different species perform distinctfixed roles in different components of ecosystem function—is the primary accepted mechanism for biodiversity-ecosystem function patterns . In contrast, we show a role for biodiversity per se in ecosystem function, in shaping dynamic specialization and its functional consequences over short timescales. While we focused on specialization and did not assess dynamic complementarity in our experiments, complementarity and specialization are often ecologically intertwined and there is evidence for interspecific competition shaping specialization in ways that likely increase dynamic complementarity . The mechanism of interspecific competition driving dynamic specialization is widespread both taxonomically and geographically , supporting the idea that our results likely extend to other BDEF relationships more generally. If dynamic specialization and/or complementarity drive some ecosystem functions, however, planting blueberries in containers why has this result not been uncovered in the hundreds of studies on BDEF relationships? Three factors might contribute. First, the vast majority of BDEF studies have focused on plants and other sessile autotrophs , while much of the work on competition and dynamic specialization has focused on animals .

Given the rapid behavioral plasticity of animals in response to interspecific competition , dynamic specialization and/or complementarity may be especially important for animal-driven ecosystem functions. Still, there is recent evidence of dynamic specialization in plant communities driven by morphological phenotypic plasticity . Second, most BDEF studies cover relatively short time scales; but longer-timescale experiments, especially for plants, may allow for more dynamic specialization to occur, and may explain in part the result that greater biomass “overyielding” in species-rich treatments relative to monocultures often occurs only after months or years in long-term BDEF experiments . Third, the bulk of BDEF experiments are designed such that species identities and abundances are tightly controlled. While such studies are undoubtedly of great value for mechanistically untangling BDEF relationships, manipulative experiments under natural field conditions , especially in terms of interspecific competition regimes, may allow for stronger dynamic responses and thus greater impacts on ecosystem function. As we have demonstrated, losses of a single pollinator species lead to reductions in dynamic specialization, driven by interspecific interactions, that in turn drive significant negative effects on ecosystem functioning in terms of plant reproduction. Our work thus suggests that ongoing pollinator declines could already be causing negative impacts on plant populations, in contrast to the network-based simulation models that predict plant communities will be robust to pollinator species losses. To prevent disruptions of pollination and other critical ecosystem functions and services, we must move beyond assuming that the functional roles of species are static and work to understand how and under what conditions phenotypic plasticity and interspecific competition can interact to drive dynamic changes in ecosystem services and functions.We assessed the effects of our manipulations on stigmatic pollen deposition and seed production in Delphinium barbeyi , a common, long-lived perennial herbaceous wildflower in our plots that is pollinated by several species of bumble bees.

We pre-bagged racemes containing immature floral buds of D. barbeyi in each plot 48-72 hours before experiments. We opened 15 separate bags containing mature, virgin flowers in the control and the manipulation experimental periods, and re-closed the bags at the end of a standardized 4-hr period. We returned to the site 3-4 days after the treatment to harvest stigmas , and 7-15 days later to harvest fruits after they had matured. Stigmas were removed from the flower and mounted on a slide with fuschin jelly in the field to avoid any loss of pollen in transit. We counted both total number of pollen grains as well as proportion of delphinium and non-delphinium pollen grains on the stigmas. We dissected mature fruits and counted developed and undeveloped seeds in the laboratory.Data collected within a study site are not independent: bees and plants are likely to be closely related genetically, and environmental conditions, floral resources, and competitive community context are all similar. To address this lack of independence and prevent pseudoreplication, we used generalized linear mixed-effects models with site as random effect. Thus, the site represents the level of experimental replication across all of our statistical models. We included three fixed effects in all models: experimental state , Bombus species richness, and Bombus abundance, allowing us to statistically control for the changes in abundance in our manipulations, as well as the differing levels of initial Bombus species richness in each plot. We assessed full models to maintain consistency and comparability among the different analyses. Our data on floral fidelity, pollen carriage, and stigmatic pollen deposition were measured in a binomial fashion: individual bees either displayed fidelity or infidelity; foraging transitions were either to a conspecific or to aheterospecific plant; pollen loads were either “pure” or else “mixed” ; and pollen grains on stigmas were identified as either conspecific or heterospecific. We used GLMMs with binomial errors to model these response variables. For all of these outcomes, the data were overdispersed , and so we included an individual-level random effect in the model to compensate for the over-dispersion .

We used the lme4 package for the R statistical programming language to conduct binomial-errors GLMMs. Seed production is a count variable, and plants with insufficient pollination do not produce seeds, resulting in zero counts. Seed count data were both highly overdispersed and zero-inflated relative to a Poisson distribution, so we modeled seed production with a zero-inflated negative binomial distribution using the glmmADMB library for R.Most pollinators are generalist foragers that can switch between plant species within a single foraging bout . This sharing of pollinators among plant species within a community can lead to the transfer of heterospecific pollen to plant stigmas. Such heterospecific pollen deposition is highly variable in nature , and can represent a substantial percentage of total pollen on a stigma, often more than 50% of grains . Stigmatic heterospecific pollen can negatively impact plant reproductive function , but we continue to have a limited understanding of the magnitude and mechanisms of those impacts, particularly under field conditions. Most of what we know about the reproductive effects of heterospecific pollen comes from hand-pollination studies, which have primarily focused on mechanisms of reproductive disruption . Heterospecific pollen can reduce reproductive output by physically blocking conspecific pollen from adhering to the stigma ; by driving stigma closure ; by producing allelochemicals that limit subsequent pollen germination ; by interfering with pollen tube growth in the style , and by usurping ovules, especially among closely related plant species .There is reason to suspect that these hand-pollination results, which come primarily from greenhouse or potted-plant studies, may not reflect the reality of field situations. In nature, we would expect a range of proportions of heterospecific pollen deposited on stigmas, as well as a range of pollen from different plant species, with different amounts of diversity on stigmas. Most hand-pollination experiments have applied fixed heterospecific:conspecific pollen ratios to stigmas, and the results on plant reproduction are variable, with some finding detrimental effects of heterospecific pollen and others showing no impact . Most of these studies used only one heterospecific pollen species and in proportions not necessarily common in nature, making it difficult to apply inference from these results to plant reproduction in the field. Moreover, we know of only one study that assessed the impact of heterospecific pollen over a continuous range of experimental heterospecific pollen proportions—a situation that is likely common in nature—and that was a hand-pollination study that held conspecific pollen quantity constant, using a single heterospecific pollen donor species. In addition, the greenhouse or potted-plant context of most or all of these studies may not translate to the field in terms of possible interactions between resource limitation and heterospecific pollen deposition. For example, container growing raspberries heterospecific pollen may have larger impacts on seed set in plants that are water- or nutrient- stressed relative to plants that are not facing serious resource limitation. Thus, while we know from field studies that heterospecific pollen deposition is common in nature, and from hand-pollination studies understand some of the mechanisms by which it can disrupt plant reproduction, the extent to which heterospecific pollen impacts plant reproduction in the field remains poorly understood. Another gap in the literature on heterospecific pollen is assessment of possible interactions between heterospecific and conspecific pollen, i.e. if the impact of a fixed amount of heterospecific pollen has varying impacts on plant reproduction depending on conspecific pollen deposition.

Such interactions could arise from mechanisms either driven by the heterospecific pollen or driven by the plant on which the heterospecific pollen is deposited. Mechanisms mediated by heterospecific pollen include stigma clogging, stylar clogging, allelopathic inhibition, and ovule usurpation . One common feature of these mechanisms is that heterospecific pollen is likely to have stronger impacts if it is deposited before conspecific pollen, especially if it is in more contact with the receptive stigmatic surface . In most realistic scenarios of pollen deposition, the more heterospecific pollen that is proportionally present on a stigma, the greater the chance that it arrived early, enhancing its negative impact, whereas with more conspecific pollen the chance of an early arrival and concomitant deleterious effects is reduced. Second, there is at least one documented mechanism driven by the plant receiving the heterospecific pollen, which is stigma closure, i.e. stigmatic lobes closing in response to heterospecific pollen, effectively ruling out subsequent seed production in that flower , and plants could also hypothetically drive active inhibition of pollen tube growth ; or ovule or carpel abortion, in response to heterospecific pollen. To effect such active mechanisms, flowers must be able to detect the presence of heterospecific pollen. If they can also detect quantity or proportion of heterospecific pollen grains, plants could potentially use that signal when actively disrupting pollination at various points in the process , ultimately as a means to conserve resources by not investing in flowers that may have low-quality seed production or quality. In this study, to begin to understand the impact of heterospecific pollen in natural systems, we used a field approach linking stigmatic pollen deposition to seed set in the same individual carpels in wild plants that had been naturally pollinated . In contrast to hand-pollination studies, this approach allowed us to assess stigmatic pollen loads varying greatly in conspecific pollen and heterospecific pollen quantities , while achieving a relatively large sample size and replication across space. This approach also allowed us to assess interactions between conspecific and heterospecific pollen in assessing plant reproduction. We examined how the total amount of naturally deposited heterospecific pollen co-varied with the reproductive output of Delphinium barbeyi Huth , a common subalpine flower species in the Rocky Mountains of Colorado, USA. D. barbeyi receives visits from several species of bumble bees that are known to visit many co-flowering species within a community . We asked the following specific questions: How variable is heterospecific pollen deposition in naturally occurring D. barbeyi populations? Is conspecific pollen and/or heterospecific pollen deposition related to whether or not a carpel will abort? How are the amounts of conspecific pollen and heterospecific pollen deposition related to seed production in carpels? We hypothesized that there would be a positive relationship between heterospecific pollen deposition and carpel abortion rates and a negative relationship between heterospecific pollen deposition and seed set. In particular, we predicted that the effect of heterospecific pollen would vary depending on conspecific pollen deposition, with heterospecific pollen having a larger negative impact on stigmas with lower conspecific pollen deposition.We used generalized linear mixed-effects models because of hierarchical lack of independence in our data . Similarly, flowers within a plant clearly do not represent independent samples. Thus, we included nested random effects in the model, with flower nested within plant nested within site. We assessed carpel abortion as counts of aborted vs.