Effect of Beauveria bassiana (Balsamo) Vuillemin (Ascomycota: Hypocreales) on Diadegma semiclausum (Hellen) (Hymenoptera: Ichneumonidae), a parasitoid of Plutella xylostella (L.) (Lepidoptera: Yponomeutidae)

Investigation of the interactions among Plutella xylostella (host), Beauveria bassiana (pathogen) and Diadegma semiclausum (parasitoid) was the aim of this thesis. The pathogenicity of B. bassiana to P. xylostella larvae and D. semiclausum females, the direct and indirect effect of the pathogen on the fecundity, longevity, development and survival of the parasitoid, behavioural responses of the parasitoid to fungus-infected and uninfected host larvae and the ability of parasitoids to disseminate fungal conidia among host larvae were investigated under laboratory conditions. The compatibility of B. bassiana with D. semiclausum for the management of P. xylostella larvae was studied under field cage conditions. Nine B. bassiana isolates (three from New Zealand and six acquired from the USA),were tested against P. xylostella larvae under laboratory conditions. All tested isolates were pathogenic. The New Zealand isolates, none of which originated from P. xylostella, were equally effective as the isolate from the USA that originated from P. xylostella. New Zealand isolate F226 had an LT₅₀ value of 3.6 days compared with the USA strains with LT₅₀ values between 4.0 - 5.0 days. When adult female parastoids were directly exposed to B. Bassiana isolates F226, F130, ARSEF 2729 and ARSEF 3288 parasitoids died within four days after treatment whereas untreated parasitoids survived 16.2 ± 1.2 days. The ability of parasitoids to discriminate between infected and uninfected larvae was investigated by using the bacterium Serratia marcescens as an oviposition marker. When D. semiclausum adults contaminated with bacteria were exposed to fungus-treated P. xylostella larvae in a choice test, the mean number of oviposition attempts over three hours made by the parasitoid (determined by the number of dead larvae with red pigmented colonies) was 4.31± 1.19 on infected larvae and 3.56 ± 1.06 on uninfected larvae. The results of the bacterial marker test showed D. semiclausum did not to discriminate between B. bassiana infected and uninfected P. xylostella larvae during oviposition attempts. When oviposition attempts were observed by video in a choice test, the mean number of oviposition attempts (number of stings) in 30 minutes made by a parasitoid on infected hosts was 4.58 ± 0.56 whereas on uninfected hosts it was significantly greater at 7.41 ± 2.64. In no-choice tests no difference in oviposition attempts was observed for both methods (bacteria marker and video observation). Although the bacterium S. marcescens could be used as an indicator of oviposition attempts by D. semiclausum on P. xylostella larvae, it will not reliably indicate the actual number of oviposition attempts. Behavioural responses of female parasitoids such as antennating, encountering and stinging, grooming and walking were video recorded. When D. semiclausum parasitized infected larvae one hour after and one, three and four days after application of fungus to the larvae, no significant difference in parasitoid activities was observed between infected larval groups. Parasitoids spent significantly more time grooming when they parasitized fungus-infected larvae than uninfected larvae, but no differences were observed for other activities such as antennating and walking. Observations showed that grooming is most likely to depend on the contamination of their bodies by conidia. Parasitoids could recognize infected dead larvae, and they did not attempt to oviposit on dead larvae. Parasitoids encountered and attempted to oviposit only on live larvae even when larvae were immobilized, indicating that they do not depend on larval movements for host recognition and oviposition cues. The effect of B. bassiana on the fecundity of parasitoids was tested with isolate F226. All treated parasitoids were dead two days after laying eggs. The number of parasitoid pupae produced by fungus-treated parasitoids in the first two days was not different from untreated parasitoids, but significantly fewer parasitoid pupae were produced by treated parasitoids than untreated parasitoids due to the greater longevity of untreated parasitoids. When female parasitoids were continuously exposed to fungus-treated cabbage leaves a positive linear relationship between the conidia concentration and parasitoid mortality was observed. Parasitoids had shorter longevity when they were exposed to leaves treated with B. bassiana conidia concentrations over 1 x 10⁸ per ml, but no significant difference was observed between untreated larvae and those treated with concentrations below lx10⁷ conidia per ml. The effect of B. bassiana on the development and survival of parasitoids when the host was first treated with pathogen and then exposed to the parasitoids at different times after inoculation was studied. No parasitoids emerged when they were exposed to B. bassiana treated larvae. When different immature stages of parasitoids (eggs, early larval stages, late larval stages and pupae) were exposed to B. bassiana, all parasitoid stages died within six days after exposure to the pathogen, with the exception of parasitoid pupae that successfully developed to adults. The number of dead larvae with mycosis was significantly different among exposed parasitoid development stages. Larger number of mycoses was observed in P. xylostella larvae from one day and three days after parasitization (eggs and early larval stages of parasitoid) than six days after parasitisation (late larval stage of parasitoid). In field cage experiments conducted during summer 2003 and autumn 2004, no significant difference was observed in the survival of P. xylostella larvae or the number of parasitoid pupae between a single application of fungus and two applications at three days apart in either season. In the summer 2003 experiment, fungus and parasitoid treatments alone and the combined treatment of fungus and parasitoid significantly reduced the P. xylostella larval population compared with the untreated cages but no synergistic effect was observed. A similar decrease of P. xylostella was observed in the autumn 2004 experiment but no significant difference was observed between parasitoid and control treatments. This indicates the pathogen was active in both seasons but the parasitoid was most active only in early summer. Even though B. bassiana has adverse effects on parasitoids under optimal conditions in the laboratory, under field cage conditions there was little indication of these effects.