Summary
The Antarctic continent covers an area of nearly 14 million square kilometres, representing one-sixteenth of the Earths land surface. Ninety-eight percent of this area is covered by ice. The remaining 2% comprises widely scattered patches of rock and mineral terrain surrounded by snowfields, glaciers and broad expanses of permanent ice-cap. From an ecological perspective these isolated habitats have essentially insular characteristics: they are analogous to an archipelago of islands protruding from an ocean. In this lecture I will highlight the opportunities that Antarctica offers for testing the principles of island biogeographic theory. Surprisingly, very little research has so far been carried out on this topic. Early studies of the sub-Antarctic islands suggested that Antarctica was anomalous because surface area was only of minor importance in explaining plant and insect species richness. More recently it has been found that variation in species richness can be explained by combinations of island area, temperature, distance from the nearest continent and the presence/absence of humans. One advantage of using Antarctica to test island biogeographic theory is the low complexity of its biotic assemblages: these range from chasmoendolithic algae in the high altitude deserts of southern Victoria Land to the bryophyte-dominated turves that occur in moist, protected sites along the western edge of the Antarctic Peninsula. Unlike lower latitude systems (e.g. Krakatau) these terrestrial assemblages have low vertical stature, small biomass, limited productivity and trophic webs that are restricted to only one or two levels. There is minimal predation, very little competition and virtually no other interspecific interactions. This low complexity means that Antarctica offers great potential as a "natural laboratory" for testing the validity of stochastic and deterministic models of island colonization under circumstances in which biotic factors play only a