The aim of this project is to investigate the way in which environmental variation cascades up through life history variation to effect species- and assemblage-level responses to habitat fragmentation and climate change. It builds on observations that cli mate change and habitat fragmentation are two major threats to biodiversity, and on recent work demonstrating that predictability and variability are two components of environmental variation that need to be clearly distinguished. The work will focus on a functionally significant group of soil animals, the springtails (Collembola), which are sufficiently circumscribed in their diversity to enable appropriate comparisons between different environments. Moreover, several of the species are common to virtual ly all of the sites owing to biological invasions in the southern hemisphere. The environments will include a southern sub-Antarctic site (Marion Island), a northern Arctic one (Svalbard), and two more temperate sites (in Scandinavia, and the Western Cape of South Africa). These sites differ substantially in the characteristics of their environments. The study will make use of both naturally (Marion, Svalbard) and artificially fragmented (Western Cape) landscapes to investigate the ways in which temperatu re effects on life histories might cascade up to influence responses to fragmentation. Specific goals of the study include investigations of widely held ideas about differences in phenology between the northern and southern hemisphere arthropods, and abou t the likely responses of soil organisms to habitat fragmentation. The project will also develop an inventory of springtails of the Western Cape Fynbos and will use genetic techniques coupled with traditional morphological systematics to identify and desc ribe new species and make this knowledge accessible more broadly to ecologists.