Sunset and the Orientation Behaviour of Migrating Birds

1. Migratory birds integrate information from a wide array of environmental sources. As our knowledge of migratory orientation depends heavily upon the results of cage‐experiments with nocturnal migrants, it is essential that the results of these cage studies be interpreted in the light of field observations of migratory behaviour and experiments with free‐flying migrants. When this is done, the impression emerges that night‐migrating birds integrate directional information prior to departure, probably during the transition between daylight and darkness. At this time, information gained from the sun, in conjunction with other references, becomes especially valuable. 2. Despite intensive work with a few species, how migrants integrate information in the selection and maintenance of a direction is not well understood. The relationship between magnetic stimuli and solar cues at sunset in the selection process, for example, remains to be resolved, as does the contribution of skylight polarization patterns at sunset. Once a migratory heading is selected, birds probably use the stars or winds aloft to maintain that direction. How migrants integrate information is largely a matter of unravelling the complex causal relations among the different environmental stimuli that serve as orientation cues. Imagine a hypothetical migrant that departs on a migratory flight around the time of sunset. Given the uncertain relationship among variables (orientation cues) that might influence her migratory orientation, a path diagram is a useful device for displaying graphically the pattern of causal relations among the set of variables (see Fig. 1). This technique is adopted from path analysis, which is a statistical method developed by Sewall Wright for studying the direct and indirect causal relations among variables (see Kerlinger & Pedhazur, 1973). The pattern depicted in the figure is less a specific model of causal relations than it is a summary of possible relationships among the several cues based on current understanding. Causal flow in this ‘model’ is unidirectional, i.e. at any given point in time a variable cannot be both a cause and an effect of another variable. For example, variable 3 is dependent on variables 1 and/or 2, and is one of the independent variables in relation to variable 5 (orientation of migratory activity). Although the value of path analysis to the study of migratory orientation may be largely heuristic at this point, ‘one virtue of the method is that in order to apply it the researcher is required to make explicit the theoretical framework within which he operates’ (Kerlinger & Pedhazur, 1973). For instance, path diagrams (and path analysis, to the degree that correlations between variables can be specified) would help researchers study (i) the apparent redundancy built into the orientation process (see Fig. 1), (ii) alternative or competing causal models of orientation and navigation, or (iii) the ontogenetic changes that affect the relationship among orientation variables. Imagine, for example, how path coefficients might change in value with migratory experience. 3. Migrants probably redetermine preferred directions soon after landing or shortly before their next departure rather than while aloft. Cage‐orientation results as well as observations of free‐flying migrants suggest that solar‐related information is involved in the morning orientation of ongoing migratory flight and possibly the re‐determination of direction following night‐time displacement. 4. Evidence is not clear on whether migrants respond to sunset by constant‐angle orientation (menotaxis) or constant‐azimuth orientation. 5. How migrants correctly identify sunset as a reference stimulus is an unresolved question. Identification might be based upon the characteristic spectral distribution of sunset, its pattern of illumination, or some other feature, such as the characteristic pattern of skylight polarization at sunset. 6. Several lines of evidence suggest that migrants learn to use the setting sun and associated skylight features as orientation cues. 7. The setting sun functions not only as a source of directional information but also as an environmental stimulus that influences the likelihood of migratory activity