The perception that bluetongue virus (BTV), once introduced to a country, would decimate its sheep industry, grew from the acceptance in the late 1950s that it was an emerging virus with Africa as its source. Epidemiological studies in the 1960s and early 1970s confirmed that the geographic distribution of BTV infections included regions of the world, outside the traditionally defined areas where BT was observed. This was interpreted at the time as representing serious and rapid spread of the virus. This review provides evidence to rebut this earlier view. What has emerged through the 1980s is: (a) the recognition that BTV is a common infection of ruminant livestock throughout the tropics and sub-tropics apparently within several separate ecosystems; (b) in most areas of the world, infection is sub-clinical; (c) incursions of virus (with accompanying disease) into temperate climates do occur at certain key locations, but "die out" usually within the same year; (d) recognition of the vector competence of Culicoides spp in the different ecosystems of the world is critical for understanding the epidemiology of disease; (e) persistent infection in ruminants is no longer considered important in the long term perpetuation of the virus.