Key points are not available for this paper at this time.
The conventional wisdom of biological control of insect pests, and its related ecological theory, is that successful natural enemies in long-lived ecosystems (1) impose a low, stable pest equilibrium, and (2) share the following properties: (a) host-specific; (b) synchronous with the pest; (c) can increase in density rapidly when the pest does; (d) need only one pest individual to complete their life cycle; (e) have a high search rate for the pest; (f) aggregate at areas of high pest density, which is thought to stabilize the interaction. These features are more characteristic of parasitoids than predators. We suggest that a stable pest equilibrium is neither a necessary nor a sufficient condition for control. We show that satisfactory control in model systems is compatible with both local extinction of the pest and polyphagy in the natural enemy. Only one of nine real examples of successful control is convincingly a stable interaction; the remainder show either strong evidence for instability and local extinction of the pest or are consistent with this interpretation. Successful natural enemies have collectively violated all of features 1 and 2 above, and violations of features 1 and 2a, 2b, and 2d appear to have been central to success in several situations. Two strategies by which a natural enemy may control a pest in a nonequilibrium state, termed here "lying-in-wait" and "search-and-destroy," are distinguished.
Murdoch et al. (Fri,) studied this question.