Summary.The distribution of the Antarctic Fulmar F. glacialoides, the Pacific Fulmar F. glacialis rodgersii, and the Atlantic Fulmar F. g. glacialis is described. In the region Iceland‐Faeroes‐Britain (and Norway) the breeding distribution of the Fulmar has undergone a revolutionary change.In Iceland the Fulmar nested in the seventeenth century perhaps only on Kolbeinsey and Grimsey in the north. In the middle of the eighteenth century a spread and increase started; by 1949 there were at least 155 colonies, many large. The Fulmar spread to the Faeroes in the early nineteenth century and now breeds in great numbers on every island. In 1878 the Fulmar spread from the Faeroes to Shetland; until then its only British station was the very large colony on St. Kilda, known since 1697 and large then. It has now spread round nearly all Scotland, Ireland, England and Wales, and is prospecting south and south‐east England. In 1949 there were at least 365 breeding colonies in Britain, and 212 further stations at which Fulmars were prospecting or occupying nest‐sites but had not been proved to breed. About thirty new stations are now being prospected yearly in Britain. At present over a hundred thousand nest‐sites are occupied at British breeding stations, of which about a third are on the one traditional station of St. Kilda, where regular harvesting ceased by 1921.The world population of the Atlantic Fulmar occupies over a million nest‐sites yearly, but probably not many millions.There is no evidence that the Fulmar's numbers are controlled by either parasites or predators, except sometimes by Man. Its food is described; it eats all manner of plankton and offal and any fatty food avidly.It is suggested that the increase in the Fulmar population was made possible when man produced new and regular supplies of fatty offal, at first by the ship's‐side flensing of whales and, when the supply of whales became exhausted, by the discarding of trawler waste and fish‐guts, and that this increase was, and is, dependent on this food rather than climate, nest‐site supply or predation. In a direct sense the Fulmar appears to be now partly a dependant of Man.The Fulmar's breeding season is strikingly constant over its full breeding range. But there is a marked difference between small and large colonies in breeding success, and a certain difference in season. It is likely that the new colonies in Britain are being prospected and established by young birds, and that their reproductive inefficiency is due to the youth of their inhabitants, not to the “ Fraser Darling Effect”.The suggestion that adult Fulmars breed not every year, but periodically, is rejected. On the contrary, it appears that the Fulmar spends the first years of its life at sea. The length of this period is unknown. In the next period of a Fulmar's life it develops an incubation patch and a drive to sit on a nest‐site, but cannot lay or fertilize an egg. This period may last for four or five years, and one marked Fulmar has been taken at a cliff (and therefore in this period of its life) in the fourth season after that in which it was hatched. When it is perhaps 8 or 9 years old, possibly 7, the Fulmar breeds, and probably continues as a rule to do so annually for the rest of its long life if it is healthy and can find a mate.