Let's assume that there is an imperceptible gradient in food availability during the non-breeding season, represented below by the green arrow. Individuals that move downwards by chance are more likely to survive until the next breeding season, and ones that move upwards less likely. This has no effect on the future habits of the survivors, however, or of their offspring, which are just as likely to move randomly in any direction at the end of the next breeding season.
The fact that birds move away from their breeding site outside the breeding season means that it's advantageous for them to have a good 'sense of direction' so that they can find their way back the following year. We know that they do have such a sense of direction: recent studies suggest that birds can 'see' the earth's magnetic field, and so effectively have an inbuilt compass.
Suppose, therefore, that some individuals in a population have a genetic 'quirk', which makes them prefer moving in one direction rather than another. This could be caused directly by a random mutation in their DNA, or it could be a side effect of some otherwise advantageous trait. The effect need only be miniscule: For instance, if a bird sitting on a rock can see two similar fruiting shrubs, one to the north and one to the south, and the bird has a genetic predisposition to go to choose the southern option, say 501 times out of 1000.
 |
