Avian olfactory navigation: its empirical foundation and conceptual state

Over the last three decades, empirical evidence has been accumulated indicating that displaced homing pigeons, and most likely other birds as well, are able to navigate home by deducing positional information from atmospheric trace gases perceived by olfaction. I briefly describe the most important results leading to this unexpected conclusion and experiments revealing the decisive role of winds at the home site. A hypothesis trying to explain the results proposes that proportional relationships among a number of atmospheric trace compounds (which determine the quality of a resulting odour) show fairly regular spatial distributions with gradient character. Gas-chromatographic analyses of air samples in an area covering 400 km in diameter revealed that differently oriented ratio gradients of volatile hydrocarbons do in fact exist. They are fairly resistant against, but not independent of, changes of wind direction. Using the measured atmospheric data as inputs, a model system is able to simulate homing navigation on a performance level similar to that observed in pigeons. Results of both pigeon experiments and air analyses are incorporated into an earlier ('preolfactory') concept of a hypothetical gradient map linked with a sun compass, which is now modified and furnished with more concrete physical substrates. Winds are identified as the crosslink embedding atmospheric ratio gradients into a directional reference scale provided by the periodically moving sun and the, geomagnetic field. Kramer's 'map-and-compass concept' is reconsidered. As crucial questions are still unsolved, olfactory navigation remains a challenge for future research. (C) 2004 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.