Few and far between: Plant population density and its effects on insect-plant interactions

This dissertation examines the effects of plant population density on insect seed predators and pollinators. In two experiments, harvester ants (Messor ebeninus) were presented with seeds in experimentally controlled densities to test for the possibility of density dependent foraging preferences. Ants significantly favored common seed types when seeds were presented as small, single-species patches (especially among less preferred seed types), but showed no bias in larger mixed-species patches. Rare plant species may consequently be protected somewhat from seed predation. Pollination, however, may be more difficult in rare plants. A computer model of pollinator foraging suggests that pollinators should specialize disproportionately on common plant species in a mixed array, becoming generalists when resources are scarce. Such behavior could cause reproductive difficulties in self-incompatible plants, due to declines in the rate and efficacy of visitation. In field populations of Diplotaxis erucoides, reproductive declines were noted in widely spaced individuals, but no pollinator observations were performed. Field experiments on the annual mustard Brassica kaber were used to test for density effects in pollination and to separate the effects of pollinator quality and quantity. Plants were grown in fan-shaped arrays in different background plantings chosen to manipulate pollinators into behaving as specialists, generalists or some combination of the two. Density effects in reproductive output were seen in all treatments, but were particularly strong where pollinators behaved as generalists, suggesting that pollinator efficiency problems outweigh visitation declines in rare plants. Such reproductive problems may have consequences for evolution and/or extinction among rare species. In a survey of 52 species of annual crucifers, I found regional and local population density to be correlated with a number of reproductive traits including breeding systems and floral display characteristics. The implications of this work for conservation biology and directions for future research are discussed.