Abstract. Intraguild predation (IGP) and cannibalism occur in the field and could affect the dominance structure of guilds of coccinellid species. The exotic biological control agent Harmonia axyridis (Pallas) (Coleoptera: Coccinellidae) is now well established in most areas of Northern and Central Italy, but it is unclear what effect this species could have on native dominant and non-dominant coccinellids with which it co-occurs. In order to predict the trophic interactions in coccinellid guilds and the likely effect of H. axyridis, the incidence of IGP and cannibalism among the following six species were evaluated under laboratory conditions: H. axyridis, three native dominant species, Adalia bipunctata (L.), Oenopia conglobata (L.) and Coccinella septempunctata L. and two native non-dominant species, Platynaspis luteorubra (Goeze) and Scymnus apetzi (Mulsant). Unfed and fed fourth instar larvae of the above species were paired in an experimental arena and the incidence of predation recorded over a period of 24 h. In absence of aphids, the survival probabilities (SP) of A. bipunctata and O. conglobata were lowest when paired with either C. septempunctata or H. axyridis (< 0.20 SP after 24 h). The SP of C. septempunctata was similar if paired with either a conspecific larva or H. axyridis (< 0.34 SP after 24 h) and that of H. axyridis was reduced similarly if paired with either a conspecific larva or C. septempunctata (> 0.71 SP after 24 h). The SP of P. luteorubra was lower when paired with A. bipunctata and C. septempunctata (< 0.07 SP after 24 h) than with other species and that of S. apetzi was greatly reduced when paired with all the dominant and exotic species (< 0.27 SP after 24 h). In presence of aphids no predatory events occurred in most combinations. H. axyridis acted as a strong predator of native dominant and non-dominant coccinellids when the aphids are scarce. We did not find any evidence, however, that the incidence of IGP among exotic and native species is higher than either IGP or cannibalism in native species. The likelihood of IGP occurring in the field is discussed.
353* GR and CR contributed to all the phases of the work and AO contributed to the statistical analysis.