Seasonal and Spatial Variation in Juvenile Survival of the Cabbage Butterly Pieris rapae: Evidence for Patchy Density-Dependence

Abstract

(1) The population densities of late-stage larvae of Pieris rapae in the Canberra [Australia] area appear to be determined primarily by the action of natural enemies. Spatial and temporal variation in fifth instar larval numbers were influenced manily by variation in larval survival rates, rather than by variation in the numbers of eggs being laid. (2) Juvenile survival was high in spring and autumn but low in summer, due to increased activity by predatory ants in the summer. (3) Natural enemies were also responsible for high spatial variability in survival. Among ten different sites, even though located in very similar habitats, there were large differences in the impact of predators, virus disease, and braconid parasites. (4) When survival rates of whole cohorts were compared, there was little evidence of density-dependent effects. However, within patches where average mortality rates were high, survival was better on those plants in the patch carrying fewer larvae. That is, where mortality levels were high, they were also density-dependent. Density-dependent mortality (due mainly to predation) was thus spatially and temporally patchy, making it invisible to normal life-table analysis of whole cohorts. (5) Egg numbers were no higher in those patches with high average survival rates than they were elsewhere, indicating that ovipositing females are not concentrated in patches low in natural enemy activity. Because the females are highly mobile, this means that patchy density dependence in juvenile survival may be able to regulate densities in the whole breeding population below the limit set by the larval food supply.