Attraction of parasitoids to plant volatiles induced by multiple herbivory depends on the specific combinations of attacking herbivore species, especially when their feeding modes activate different defense signalling pathways as has been reported for phloem feeding aphids and tissue feeding caterpillars. We studied the effects of pre-infestation with non-host aphids (Brevicoryne brassicae) for two different time periods on the ability of two parasitoid species to discriminate between volatiles emitted by plants infested by host caterpillars alone and those emitted by plants infested with host caterpillars plus aphids. Using plants originating from three chemically distinct wild cabbage (Brassica oleracea) populations, Diadegma semiclausum switched preference for dually infested plants to preference for plants infested with Plutella xylostella hosts alone when the duration of pre-aphid infestation doubled from 7 to 14 days. Microplitis mediator, a parasitoid of Mamestra brassicae caterpillars, preferred dually-infested plants irrespective of aphid-infestation duration. Separation of the volatile blends emitted by plants infested with hosts plus aphids or with hosts only was poor, based on multivariate statistics. However, emission rates of individual compounds were often reduced in plants infested with aphids plus hosts compared to those emitted by plants infested with hosts alone. This effect depended on host caterpillar species and plant population and was little affected by aphid infestation duration. Thus, the interactive effect of aphids and hosts on plant volatile production and parasitoid attraction can be dynamic and parasitoid specific. The characteristics of the multi-component volatile blends that determine parasitoid attraction are too complex to be deduced from simple correlative statistical analyses.
Keywords: Brassica oleracea; Diadegma semiclausum; Indirect defense; Mamestra brassicae; Microplitis mediator; Multiple herbivory; Natural enemies; Parasitoid behavior; Plant volatiles; Plutella xylostella.