Plant-Pollinator Interactions in an Ecological and Evolutionary Context: The Promising Role of 3D-Printing Technology and Mathematical Modeling

Thesis (Ph.D.)--University of Washington, 2017-06This dissertation concerns itself with the role of flower shape in affecting the foraging performance of pollinating animals. The pollinator used in this study is a model organism representing crepuscular hawkmoths in research involving the study of flight neuromuscular physiology and plant-pollinator interactions, Manduca sexta (hereafter Manduca). The broader goal of the work is to develop a new experimental framework for investigating the ecological and evolutionary consequences of plant-pollinator interactions. To that end, I have combined 3D-printing technology and mathematical modelling to construct artificial flowers, which can be manufactured with great precision and with objective, quantitatively describable shapes. First, I present a proof-of-concept study to demonstrate the feasibility of collecting foraging data from a real animal pollinator attempting to feed from 3D-printed artificial flowers. I show that Manduca’s foraging performance is extremely sensitive to variation in floral corolla curvature and nectary diameter. These results validate the experimental approach that I describe, justifying a large investment of time and financial resources into a major iterative improvement of my experimental apparatus. Next, I describe the improved experimental apparatus and use it to construct a performance landscape of Manduca’s innate foraging performance as a function of variation in flower morphology. This landscape suggests that Manduca’s foraging performance is contingent on a context-dependent interaction between corolla curvature and nectary diameter. Finally, I use data from infrared sensors attached to each artificial flower and custom computer vision software (both improvements over the proof-of-concept apparatus) to identify putative proxies of fitness for both the pollinator and the plant (artificial flower). The goal is to examine whether the pollination mutualism between Manduca and the flowers that it visits in nature could represent a scenario of evolutionary conflict or harmony. The results are inconclusive due to opposing conclusions that the various indirect proxies of flower fitness point to. As a result, I make suggestions for improving the experimental apparatus even further by including a physical touch-sensitive sensor into the design of the artificial flowers. Such a sensor would act as a physical analogue for a real flower’s reproductive structures, allowing the apparatus to yield a direct rather than indirect measure of flower fitness. Still, the general framework of using 3D-printed flower whose shapes are mathematically specified shows great promise for opening up new areas of experimental inquiry in the field of plant-pollinator interactions