Predators in the Petals: How crab spiders affect plants and their pollinators

Plants are literally rooted; they rely on biotic or abiotic vectors to carry their pollen and enable outcross mating. Most angiosperms partner with insect pollinators. These pollinators, by mediating sexual reproduction, directly impact plant reproduction and affect plant fitness, selection on plant traits, plant mating system, and population genetics. However, because flowers are reliable places to find high numbers of insect visitors, many predators exploit flowers as hunting sites. Predators that specialize on hunting floral visitors, i.e., pollinator predators, are an understudied group of predators and their direct effects on pollinators and indirect effects on flowering plants are poorly understood. I explore the history of research on pollinator predators and summarize what I know in chapter one, highlighting gaps in my understanding, and future research directions. I then present original research aimed at enhancing my understanding of the complex direct and indirect interactions among pollinator predators, floral visitors, and flowering plants. In chapter two, I investigate the impact of crab spiders, common pollinator predators that ambush floral visitors, on the strength and direction of phenotypic selection on floral traits of the yellow mariposa lily, Calochortus luteus (Liliaceae), and seed set as a component of plant female fitness. I find evidence that crab spiders are causal agents of selection on plant traits, and can affect seed set. In chapter three, I explore the effects of crab spider presence on the behavior of floral visitors, and the consequence for C. luteus pollination. I find that the presence of a crab spider significantly reduces pollinator visitation frequency and duration, and decreases stigma and anther contact, yet does not detectably affect pollen limitation or selfing rates. In chapter four, I present the results of a multi-year greenhouse and genetic study to investigate the impact of crab spiders on fitness, inbreeding depression, and population genetics of C. luteus. I find evidence of inbreeding depression within this population of C. luteus. Plants that are the product of outcross pollination have significantly higher fitness metrics than plants that are the product of self-pollination; however, when grown in low-stress greenhouse conditions fitness differences were not apparent until the plants reached adulthood (~3 years). I predicted that progeny from flowers that harbored crab spiders would be more likely to be self-pollinated and therefore would express higher levels of inbreeding depression than progeny from crab spider free plants. I found limited support for this prediction with non-significant trends in the predicted direction. Continued data collection may reveal additional cumulative negative effects of inbreeding depression on the fitness of plants produced via selfing. Together, this body of work suggests that predators in higher trophic levels can have meaningful indirect and direct effects on plants by altering selection on plant traits, pollination, and possibly even population genetics.