On the population dynamics of the Eucalypt-defoliating sawfly Perga affinis affinis Kirby (Hymenoptera)

Abstract

Changes of the abundance of the sawfly P. a. affinis were recorded during eight successive years in a study region extending from the Southern Tablelands of New South Wales to north-eastern Victoria. The influences and processes which affected mortality and reproduction in populations of the sawfly are described, and their contributions to the observed fluctuations of abundance in the study region are discussed. The major causes of mortality were desiccation, fungal disease, and parasitism. The main parasites were tachinids (Froggattimyia spp.) and a trigonalid (Taeniogonalos venatoria Riek). A variable proportion of the prepupae entered a diapause which persisted for 2 or more years. Although extended diapause ensured the survival of the subspecies during seasons unfavourable for its active stages, only a small proportion of the prepupae involved survived to emerge as adults. The consequences of very high numbers were: increased incidence of desiccation mortality and of extended diapause, and a decrease of the fecundity of the subsequent generation. Due to the benefits which accrued from colonial behaviour, survival of the larvae was greater at moderately high than at low densities. These density-regulating mechanisms of the life system of the sawfly operated conspicuously only at very high and very low densities. In the eastern portion of the study region, fluctuations of abundance were closely synchronized in all sites. These fluctuations could be interpreted in terms of changing environmental favourability during successive seasons. Analysis of weather data showed that departures from long-term average temperatures and rainfalls were correlated throughout the region. However, sawfly numbers fluctuated erratically in the western subregion, the changes of density there often being at variance with those consistent throughout the eastern subregion. Parasitism by T. venatoria, which was often very severe in the western subregion, was the only process recognized that was likely to be responsible for the differences observed.