Date: 2021-01-01

Creator: Katie J. Galletta

Access: Open access

Induced defenses following herbivore damage can modify a plant’s chemical or physical characteristics and alter the plant’s interactions with subsequent herbivores. Common milkweed (Asclepias syriaca) provides an excellent system with which to study plant response-mediated interactions given its small but highly specialized herbivorous insect community and its ability to increase toxic cardenolide concentrations and latex production throughout its tissues upon attack. I conducted observational field surveys quantifying leaf damage to examine whether the indirect plant-mediated interactions amongst the milkweed herbivore community as demonstrated in other studies also occur in situ, as well as how foliar herbivory impacts insect flower visitation on A. syriaca. I found that four-eyed milkweed beetle (Tetraopes tetrophthalmus) damage had a negative effect on subsequent monarch (Danaus plexippus) larvae and swamp milkweed leaf beetle (Labidomera clivicollis) damage. I also found that monarchs laid more eggs on milkweed with no herbivore damage. Additionally, I observed a negative relationship between A. syriaca foliar herbivory and flower visitation, which has not been previously demonstrated but illustrates the various potential costs of herbivory to plant fitness. My work’s focus on observing the effects of natural herbivore damage offers insight as to how plant-mediated interactions operate among the milkweed insect community in situ. Furthermore, this study demonstrates how plant responses to herbivory in general can modulate ecological relationships between species that do not directly interact with each other.

Date: 2021-01-01

Creator: Max Thrush Hukill

Access: Open access

The standard statistical methodology for analyzing complex case-control studies in ethology is often limited by approaches that force researchers to model distinct aspects of biological processes in a piecemeal, disjointed fashion. By developing a hierarchical Bayesian model, this work demonstrates that statistical inference in this context can be done using a single coherent framework. To do this, we construct a continuous-time Markov chain (CTMC) to model bumblebee foraging behavior. To connect the experimental design with the CTMC, we employ a mixture model controlled by a logistic regression on the two-factor design matrix. We then show how to infer these model parameters from experimental data using Markov chain Monte Carlo and interpret the results from a motivating experiment.