Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.

Author ORCID Identifier


Campus-Only Access for One (1) Year

Document Type


Degree Name

Doctor of Philosophy (PhD)

Degree Program

Organismic and Evolutionary Biology

Year Degree Awarded


Month Degree Awarded


First Advisor

Lynn S. Adler

Subject Categories

Biodiversity | Botany | Entomology | Integrative Biology | Population Biology


Plant and pollinator interactions are integral to the reproduction of wild and domesticated flowering plants. These interactions are often affected by other members of the community, such as other plants, pollinators, herbivores, and pathogens. The goal of this dissertation is to investigate and understand how the context in which plant-pollinator interactions exist affects their outcomes.

Interactions between individual plants and pollinators are known to be shaped by other plants and pollinators in the community. However, little is known about how community-scale patterns of interactions (i.e., network structure) are affected by the composition of plant and pollinator communities. In chapter 4, we examined the relationships between the taxonomic and functional diversity of plants and pollinators, and plant-pollinator network structure in an elevational gradient in an alpine ecosystem. We recorded plant-pollinator interactions in 24 transects along 8 elevational steps in the Austrian Alps, and for each transect we quantified taxonomic diversity and functional diversity, using morphological traits in flowering plants and their pollinators. For each transect, we also calculated three measures of network structure, complementary specialization (H2’), modularity, and nestedness (weighted NODF). We found that all measures of diversity and network structure were correlated with elevation. We also found that complementary specialization and modularity were mediated primarily by pollinator functional diversity, while nestedness was mediated by plant functional diversity and pollinator phylogenetic diversity. By linking plant and pollinator functional diversity to network structure, this study helps to elucidate the mechanisms by which diversity can impact pollination services at the community scale.

Plants and pollinators exist within the broader context of food webs and often interact with antagonists, which can shape the traits of both plants and pollinators. For example, plants have evolved defenses to reduce damage by herbivores. However, the deployment of anti-herbivore defenses by plants can be accompanied by changes to traits that mediate plant-pollinator interactions. In chapter 5, we assessed how herbivory on common milkweed (Asclepias syriaca) affected the composition of milkweed’s floral volatile organic compounds (VOCs) and the relationships between those compounds. We collected the floral scent of control plants and plants subjected to simulated herbivory in a natural community and analyzed VOC emissions using gas chromatography-mass spectrometry. We found that floral scent composition differed between control and herbivory-induced plants. Furthermore, we found that when VOC compositions differ, the correlations between floral VOCs increased. The effects of herbivory can go beyond the interactions between the damaged plant and its pollinators. Herbivory often has negative impacts on host plant pollination via reduced floral traits, which make flowers less attractive to pollinators and reduce reproductive success. But since multiple flowering plant species often rely on the same pollinators, effects of herbivory on one species may have knock-on effects on undamaged neighbors. In chapter 3, we assessed how herbivory on an attractive and abundant flowering plant species affects pollination services to neighboring plants. We set up plots in natural communities with three or four flowering species. In half of the plots, we simulated damage to common milkweed (Asclepias syriaca) plants and maintained the remaining plots as controls. We recorded the composition of floral visitors in the herbivory and control plots, the proportions of heterospecific pollen deposited on four neighboring plant species, and seed production in one species. We found differences in the composition of floral visitors in damaged compared to undamaged plots and observed species-specific changes in the proportion of heterospecific pollen deposition on undamaged neighboring plants. In two neighboring plant species, heterospecific pollen depositions increased in the simulated herbivory plots, in another species heterospecific pollen deposition declined, and in another species, we saw no change. Along with these changes in pollen deposition, we also observed a marginally significant decline in the average number of seeds per fruit in cow vetch (Vicia cracca). These results suggest that herbivory on one flowering plant species can impact pollinator services for other flowering species at the patch level. These effects highlight the role that herbivory may play in shaping entire flowering plant community dynamics via indirect effects.

While plant antagonists can have dramatic effects on pollination services, pollinator antagonists such as pathogens could have similarly negative effects on plant-pollinator interactions. For example, some pathogens could impact a pollinator’s ability to forage. Furthermore, the effects of pollinator pathogens could change when combined with the effects of plant herbivores. These interactions are important to understand given that these antagonists are ubiquitous. In chapter 2, we investigated the effects of herbivores and pollinator pathogens on the duration of pollinator visits to tomato plants. We created a 10 x 11 array of tomatoes (Solanum lycopersicum), and half the plants were exposed to damage by tobacco hornworms (Manduca sexta). Within this array, we recorded the duration of visits by wild Bombus impatiens workers to herbivore-damaged and control plants, and we collected the B. impatiens individuals after their final visit to assess infection status. We found that while bees free of pathogens spend slightly longer on flowers from herbivore-damaged plants, the duration of flower visits on damaged plants declined with increasing levels of Crithidia bombi infections. These results demonstrate that multiple antagonists can have synergistic negative effects on the duration of pollinator visits, such that the consequences of herbivory may depend on the infection status of pollinators.

The multitrophic interactions between herbivores, flowering plants, their pollinators, and their pathogens can impact more than just pollination services. For instance, defense chemical compounds that evolved to deter herbivores are often present in the floral rewards that pollinators consume. Interestingly, these compounds not only impact the pollinators that consume them, but can also have detrimental effects on pollinator pathogens, thus providing a medicinal benefit to infected pollinators. In chapter 1, we investigated the effects of herbivory on the production of two anti-herbivore, alkaloid compounds in tobacco (Nicotiana tabacum) pollen and the effect of this pollen on pollinator pathogens. We grew tobacco plants in a greenhouse and subjected half of them to a specialist herbivore, tobacco hornworm (Manduca sexta) and analyzed the alkaloid composition of the pollen using high-performance liquid chromatography (HPLC). We then fed the pollen of control and herbivore-induced plants to Bombus impatiens individuals infected with the gut pathogen, Crithidia bombi. We found that herbivory increased the concentrations of the alkaloid anabasine, but not nicotine, and that the concentrations of both compounds fluctuated as a function of time since flowering. Additionally, we found that bees which consumed late-collected pollen from herbivory-induced plants had pathogen counts 15 times higher than bees fed pollen from control plants, but only when pollen was collected after one month of flowering. These results emphasize the role of herbivores in shaping pollen chemistry, with consequences for interactions between pollinators and their pathogens.

Altogether, the work in this dissertation demonstrates that the biotic context around plant-pollinator interactions can shape floral traits, the quality of pollination services, community-wide pollination processes and food web dynamics.


Creative Commons License

Creative Commons Attribution 4.0 License
This work is licensed under a Creative Commons Attribution 4.0 License.

Available for download on Sunday, September 01, 2024