9629316 Murdoch Control of California red scale, a pest of citrus, by the parasitoid wasp, Aphytis, is one of the most spectacularly successful examples of biological control. Scale are kept at very low densities, and populations are remarkably constant in abundance and dynamically stable. The goal of this project is to explain the co-occurrence of severe pest suppression and dynamic stability, which is contrary to expectations based on simple standard mathematical models of such systems. Prior research has narrowed the potential explanations to only a few, using a combination of observations and experiments in the field and laboratory to investigate mechanisms, and the development of detailed mathematical models that incorporate these mechanisms. The remaining mechanisms include (1) various behavioral responses of Aphytis to individual properties of scale and scale density, (2) a density-dependent increase in the rate at which Aphytis kill previously-laid parasitoid eggs and (3) small-scale spatial heterogeneity in the vulnerability of scale to attack by Aphytis. The primary approach to testing these hypotheses involves a large field experiment. Grapefruit trees will be individually caged and the density of scale on the trees increased 10-fold in one set of trees, 50-fold in a second set and left at ambient density in a third set. It is expected that, through time, the scale densities on higher-density trees will converge to those on ambient-density trees. The hypothesis will be tested by monitoring various system responses as this "density convergence" occurs. Small-scale experiments carried out concurrently inside cages will focus on attack responses of Aphytis. The hypotheses will also be tested by comparing the results of different versions of a detailed mathematical model of the system that include or omit the various mechanisms. This research has wide-ranging implications for ecological theory. It will also contribute to understanding of biological control and to the ability to use it more broadly and effectively.