Many migratory birds show philopatry, i.e. they regularly breed and winter at the same sites. The routes taken by migrants are adjusted to the geographical and ecological conditions between breeding and wintering areas, often resulting in indirect paths. Young birds on their first migration face the task of reaching the as yet unknown population-specific winter quarters with the help of innate information. Large-scale displacement experiments with migrants and cage experiments with hand-raised birds revealed that this innate information is given as direction and distance, with the distance controlled by an endogenous time program that determines amount and temporal distribution of migratory activity. Both, migratory activity and direction — or, in the case of indirect routes, a sequence of directions — are genetically transmitted from one generation to the next.
Birds use two reference systems to convert innate directional information into an actual flying course: celestial rotation and the geomagnetic field. Celestial rotation produces a reference direction opposite from its center, which is obtained by observing the diurnal and/or the nocturnal sky. This reference can be used to establish a star compass, not only utilizing the natural, but also artificial “stars”, provided the birds can observe these “stars” rotating. However, with only stars available, migrants that normally prefer southwesterly courses show southerly tendencies, apparently unable to convert the population-specific components of their migratory direction. Birds raised with only magnetic cues available, in contrast, are well oriented in their population-specific migratory direction, except in areas with steep inclination; here, the magnetic field provides only an axis, and birds also need celestial rotation for unimodal orientation. As the birds' magnetic compass is an inclination compass, migrants of the northern and southern hemisphere may use the same migratory program, starting out “equatorwards” in autumn.
During the premigratory period, both reference systems interact to determine the migratory course. If North indicated by celestial rotation and magnetic North diverge, celestial rotation proves dominant, resulting in a changed magnetic compass course. However, celestial rotation does not simply override the magnetic course. In the natural situation, celestial rotation provides only the reference direction “opposite from the center of rotation”, corresponding to geographic South, which can be substituted by magnetic South if birds have no access to celestial cues. Population-specific deviations from South seem to be coded only with respect to the magnetic field and are then added to the reference direction, resulting in the population-specific migratory course. These processes are interrupted if the sky is made to rotate in the reverse direction. The reasons for using two reference systems may lie in the fact that at higher latitudes, the magnetic field is strongly affected by secular variations, while celestial rotation reliably provides geographic South. At the same time, the magnetic field, being directly perceivable, may be better suited for indicating angular deviations.
During migration itself, the relationship between the two reference systems changes, with the magnetic field becoming dominant. In case of conflict, celestial cues are recalibrated according to magnetic North. The reasons for this shift in dominance may lie in celestial rotation ceasing to play a role. The sky changes its appearance as the birds progress, and the new stars are calibrated with the help of the geomagnetic field which becomes a reliable source of directional information at temperate and lower latitudes.
Many birds change direction during migration. Their second compass course is coded with respect to the magnetic field. The conversion of the respective innate information appears to take place en route; a possible role of celestial rotation has not yet been analysed. In Garden Warblers and Yellow-faced Honeyeaters, the shift in direction can take place under the control of the endogenous time program alone; Pied Flycatchers, in contrast, require magnetic conditions of the region where the shift normally takes place. At the magnetic equator, birds must reverse their course with respect to their magnetic compass from equatorwards to polewards in order to continue southwards. Here, the field of the equator with its horizontal field lines serves as trigger. At the equator itself, where the magnetic compass becomes bimodal, birds may rely on celestial cues.
The innate migratory program enables young birds to reach their general wintering area. The program becomes flexible at the end and allows them to look around for a suitable site to spend the winter. This becomes their winter “home” to which they return upon displacement. For the return migration to the breeding area and any later migrations, migratory birds can make use of experience obtained during earlier travels. The migratory program still provides them with directional information; however, navigational processes based on site-specific information dominate over the innate mechanisms. Many young birds undertake extended exploratory flights before they leave for migration, thus establishing a “map” of their future breeding area. As a consequence, they return to the normal breeding area after displacement. Adult birds must be expected to choose their migration route by mechanisms of true navigation, thus avoiding unfavorable areas and revisiting good refueling sites, at the same time becoming less vulnerable to wind drift and similar phenomena. Details of these navigation processes are not known, as they have escaped experimental analysis so far.
The dominant role of true navigation, which replaces the innate program, represents a parallel to homing, where birds also rely on mechanisms of true navigation as soon as these become available.