The North Atlantic islands, the Shetlands, Faeroes, Iceland and Greenland, have a flora and fauna with no truly endemic species. Their populations are dominated by Eurasian species, particularly conspicuous among which are the bulky flightless insects that should have found difficulty in reaching the islands after the retreat of the glaciers of the last ice age. The puzzling origin of this biota has been the subject of prolonged controversy. Most hypotheses have apparently insurmountable difficulties, largely because they are based on present-day observations augmented by presumed geological inferences for which there is very little hard supporting data. In recent years there has been a dramatic increase in our knowledge of the complexity of Quaternary events and also of the flora and fauna that lived along the southern margins of the ice sheets. It is suggested here that the islands lost almost all their biota at the height of the glacial periods and that they were invaded and colonized almost entirely from northwest Europe during the short phases at the end of the glacial periods, when fresh meltwater and ocean surface currents would have acted as potent aids to dispersal in the North Atlantic. It is suggested that the islands to the south of the major glacial influences, such as the Azores and Madeira, have endemic species because they were not subjected to the frequent exterminations and recolonizations that afflicted the islands further north. In testing this hypothesis, it has been possible to show that the carabid beetle faunas of the North Atlantic islands become more incomplete from east to west in terms of the potential numbers of species that each island could support, given its present-day climate and unimpeded access of invasion and colonization. This suggests a progressive loss of these earthbound insects as they were transported for increasing distances across a hostile ocean.