Reproductive ecology of Chrysophtharta agricola (Chapuis) (Coleoptera: Chrysomelidae)

Chrysophtharta agricola (Chapuis) (Coleoptera: Chrysomelidae) is a pest of eucalypt p~oduction forests in south-eastern Australia, especially Tasmania. My thesis documents the reproductive ecology of C. agricola to identify biological factors influencing its population dynamics, and hence, pest status. Research covers three main areas. First, phenology, natural enemies, reproductive developmental morphology, overwintering, and dispersal are investigated, and day-degree models of population phenology are developed and validated using field data. Second, oviposition and larval behaviours are studied through experiments on intra-plant host selection, and the consequences of larval gregariousness. Finally, the mating system of C. agricola is examined both in an applied context to identify potentially exploitable behaviours for manipulative pest control methods, and in a theoretical context to investigate the evolution of multiple mating, sperm utilisation and mate choice. Adult C. agricola emerge from reproductive diapause in leaf litter, following depletion of fat body, in October and November, and mate and oviposit between October and January, with larval populations peaking in December. New generation adults were present in plantations between January and April, producing eggs in some seasons, and entering reproductive diapause without reproducing in others. Voltinism in C. agricola was found to be a seasonally plastic trait, dependent upon photoperiod at the time of emergence of teneral adults. Day-degree models of population phenology based on laboratory development trials provided a good indicator of predicting first appearance of lifestages in the field. Two of C. agricola's 24 Eucalyptus hosts are important plantation species whose heterophylly (the production of morphologically and chemically distinct foliage) influences the intra-plant distribution of adult and immature stages. Adult beetles feed preferentially on adult foliage, and oviposit preferentially on juvenile foliage, despite neither foliage type conferring increased fecundity or longevity to adults, or gro~ and survival advantages to larvae. Larvae feed gregariously in all instars, and larval density, leaf age and damage to the leaf margin affect larval survival. Immature life stages of C. agricola experience greater than 86 % mortality between egg and final instar larvae: natural enemies and host plant attributes contribute to this low survival rate. In the field, approximately 85 % of all larvae occurred on the first two expanded leaf pairs, and larval mortality was highest between eclosion and establishment of the first instar. The first instances of egg parasitism of C. agricola were recorded in this study, as well as adult parasitism by an undescribed genus of sexually-transmitted parasitic mites. The former, a pteromalid wasp, demonstrated differential parasitism success according to parasitoid geographical origin in reciprocal exposure experiments. The latter have provided insights into the ecology of sexual transmission, and empirical evidence of female-bias of sexually transmitted infection, which has been predicted in animals (Appendix Ill). Other natural enemies were also identified, with significant mortality attributable to predatory mirid bugs and spiders. Two colour morphs of C. agricola adults are present in the field in mainland Australia and Tasmania: a common green-brown form and a rare black form which was found to be genetically dominant over the common form, with no fitness differences or non-random mating between colour morphs. Population models over 100 successive generations confirmed that melanism in this species is controlled by a dominant, neutral allele. Crossing tests using melanism as a paternity marker showed no distinct or consistent pattern in sperm use by dual-mated females. Adults mate many times in their reproductive lifetime, and 20 - 30 % of teneral females mate in late summer and autumn before entering reproductive diapause, and can produce viable eggs in spring without remating. Beetles did not select mates based on size, colour, parasite load or age. Males were more active within plantations than females, representing 90 % of beetles captured on sticky traps, despite no deviation from unity of the operational sex ratio over the same period. Significantly more males were caught on traps baited with conspecific beetles and host foliage than on traps that contained only host foliage, suggesting that volatiles are involved in mate location. Population genetic analysis using allozyme electrophoresis of specimens collected in New South Wales, Victoria and Tasmania suggested that gene flow in C. agricola is limited between populations and between regions (mainland Australia and Tasmania). Therefore, populations of C. agricola would be expected to be spatially predictable and susceptible to attack by natural enemies. Mainland Australian and Tasmanian C. agricola populations also differed in their egg parasitism rates, adult elytral colouration and collection frequency in autumn.