PASSERINE MIGRATION IN SOUTH SCANDINAVIA IN THE AUTUMN OF 1954.

Summary.1  An analysis is made of records of passerine bird migration at certain “bird observatories” in the North Sea‐Baltic area in September 1954. Some 1955 records are also used 2  Bird‐observatory records are thought to be only a rough sample of migration over a wider area, and suggestions for improvement in sampling are put forward 3  Extending the work of other writers, the view is adopted that the initiation of migration is controlled principally by an internal rhythm, but that external factors break up the migration into short phases. Because of the variations in the environment the effect of these factors defies detailed analysis 4  The 1954 and 1955 records show a fairly simple relation between the initiation of migration and the clearance of depressions. Day‐migrants (but not night‐migrants) avoided migrating in stable anticyclonic weather 5  Each species of day‐migrant was concentrated markedly into a relatively narrow “channel” across the area. This canalization appears to be achieved by means of accurate navigation 6  Three very different movements of nocturnal migrants are analysed. It is thought that “drift‐” migration plays an important part in the movements of night‐migrants through Scandinavia 7  The reactions of day‐migrants to topography appear to protect the birds against unfavourable weather, and to stabilize their somewhat unreliable powers of orientation. There is no evidence that night‐migrants navigate accurately at night, and they appear instead to derive the maximum benefit from meteorological factors. In either case the uncertainty of the birds' movements contrasts with the ultimate accuracy with which they migrate 8  A theory of navigation is proposed.