No abstract

Abstract. 1. The aphid alarm pheromone (E)--farnesene (EBF) is often considered to be used by natural enemies as a prey/host finding kairomone. However, studies show opposing results, some appear to confirm an attraction of aphid natural enemies by EBF whereas others do not provide any evidence for the kairomone function of EBF.2. To clarify if aphid natural enemies are attracted by the amounts of EBF naturally emitted by aphids, the existing literature was reviewed about EBF attractiveness to aphid natural enemies with consideration of the amounts of EBF used in the studies.3. Thirty-one publications that investigated the ability of EBF, aphid cornicle secretion, and attacked aphids, to attract aphid natural enemies were found. Several studies showed an attraction by EBF, but these used much higher amounts of EBF than usually emitted by aphids during a predator attack. Studies investigating EBF amounts similar to what is emitted by aphids are rare and failed to show attraction. Only two studies document an attraction of natural enemies by attacked aphids.4. As EBF is emitted in very low amounts, not very stable, and only present after an attack, we suggest that aphid-derived EBF is not a suitable kairomone for most natural enemy species, especially when they are able to use alternative cues. As EBF, amongst other volatiles, is also emitted by herbivore-induced plants, we propose that natural enemies might use plant-derived EBF as a synomone to identify aphid-infested plants via an altered plant volatile bouquet.

The enormous biodiversity of herbivorous insects may arise from ecological speciation via continuous host-plant switches. Whether such switches are successful depends on the trade-off between different selection pressures that act on herbivores. Decreased herbivore performance due to suboptimal nutrition might be compensated for by a reduced natural enemy pressure. As a consequence, an "enemy-free space" on a certain plant might facilitate host-plant switches and maintain biotypes. To test this hypothesis, we used the pea aphid (Acyrthosiphon pisum) complex, which consists of at least 11 genetically distinct host races that are native to specific legume host plants but can all develop on the universal host plant Vicia faba. Three A. pisum host races native to Trifolium pratense, Pisum sativum, and Medicago sativa were investigated in experiments on their respective host plants and on the universal host plant V. faba. We found that hoverflies preferred to oviposit on P. sativum and the universal host V. faba. Since feeding by hoverfly larvae suppressed aphid population growth on these host plants, the native hosts M. sativa and T. pratense provided enemy-free space for the respective A. pisum races. Mobile predators, such as ants and ladybird beetles, preferred Pisum race aphids on V. faba over P. sativum. Thus, all three of the native host plants studied supply enemy-free space for A. pisum compared to the universal host V. faba. Reducing encounters between aphid races on V. faba would reduce gene flow among them and could contribute to maintaining the host races.

No abstract