Multiple predators, intraguild interactions and biological control of a single spider mite species

To test whether biological control of spider mites is promoted by the use of multiple instead of a single predator species, experiments have been published that employed either an additive design (initial density of each predator is kept constant) or a replacement design (initial density of all predators together is kept constant). The variable under test is usually some measure of pest abundance. We argue that such tests are not always adequate to infer positive or negative effects among multiple predator species and their impact on the pest. Instead, there is a need for experiments elucidating predator-predator interactions and their impact on the pest at a mechanistic level, using a series of experiments ranging in focus from the individual level to the population level, from the laboratory to the field, from short-term to long-term and from small scale to large scale. To illustrate this, two experiments will be discussed, the first based on a replacement design and the second on an additive design: (1) Combined releases of Phytoseiulus persimilis and Neoseiulus californicus for control of spider mites on strawberries in Turkey, and (2) Predator removal with respect to established populations of Typhlodromalus aripo and Amblydromalus (Typhlodromalus) manihoti, which are supposed to control green mites on cassava in Africa. These examples show that microhabitat specialization of the predators in the field reduces negative effects of intraguild predation among predator species and promotes their synergistic effect on pest suppression.