How do birds know where to migrate?
A Socratic walk-through of how birds know where to migrate — reasoned out one step at a time, not lectured.
The question we started with
THE QUESTION #How do birds know where to migrate?
A young bird that has never made the journey can fly thousands of miles to a place it has never seen, and arrive. We could not do that without a map and a compass. So what serves the bird as map and compass — and how much is learned versus somehow already known?
Reasoning it through
REASONING #Consider what a traveler needs: a sense of direction (which way is south) and a sense of place (am I there yet?). A single clue could fail — the sun hides behind clouds, stars vanish by day. So might a bird carry several clues at once, leaning on whichever is available? What signals are always, or nearly always, present in the sky and the ground?
Before we list them, notice a prior problem: how would anyone find out? A bird in a cage cannot migrate, and a bird in the sky cannot be interrogated. The trick that opened the field exploits a bird's own restlessness — caged migrants in season hop persistently toward the direction they would fly if they could. Put one in a funnel-shaped cage lined with paper, only the sky visible above, and its hops leave a scatter of marks concentrated on one side. Now you have a protractor: change one cue at a time and read off which way the marks move.
Run it and the cues separate cleanly. The sun gives direction only if you know the time, since its bearing changes through the day; so altering a bird's clock should alter its heading predictably. Keep a bird for days under artificial light shifted six hours from the real day and it sets off about ninety degrees off course — exactly what a sun reading taken against a clock six hours wrong would give.
The stars work differently. Raise young buntings in a planetarium under a sky rotating around Betelgeuse rather than Polaris, and come the season they orient away from Betelgeuse. So they are not born knowing constellations: they learn as nestlings where the sky's centre of rotation is — the one point that stands still — and take that as north. Which is elegant, since the pole star itself is not fixed over long spans of time, whereas the fact that the sky turns about something always is.
The third cue is the one with no human counterpart. Birds detect the Earth's magnetic field, but not as a compass needle does: they read the inclination, the angle at which the field lines dip into the ground, which distinguishes poleward from equatorward without telling them north from south at all. Reverse the horizontal component of the field in the laboratory and a robin ignores it; reverse the vertical component and the bird turns around.
One last question separates the two things we started with: is a bird finding a direction or finding a place? Take migrating starlings from the Netherlands, fly them to Switzerland and release them. The adults corrected, making for their usual wintering grounds around England despite starting from somewhere they had never been. The juveniles flew the heading they would have flown at home and wintered in France and Spain — displaced, and never aware of it. So a first-time migrant runs an inherited instruction, a direction and roughly a duration, while an experienced one navigates to a goal it now knows.
The analogy
THE ANALOGY #Imagine a traveler holding several overlapping maps at once: one reads the arc of the sun, another the patterns of the stars, and a third is a compass needle that senses the Earth's magnetic field. When clouds hide the sun, the traveler simply trusts another map. No single guide is the answer — it is the cross-checking that keeps the route true. And on a first journey the traveler has no maps at all, only a sealed envelope reading "south-west, six weeks" — which works splendidly until somebody moves the starting point.
A traveler chooses which map to consult. A bird deliberates over nothing — the cues are weighted by machinery we still only partly understand, and the magnetic sense in particular remains an open research question rather than the settled compass needle the picture implies.
Clarifying the model
THE MODEL #So migration is not one mysterious instinct but a bundle of senses — sun, stars, and a built-in magnetic compass — some inherited and some tuned by early experience. Redundancy is the point: lose one cue and the others still hold the heading.
The misconception worth naming is that these cues sit in a fixed chain of command, with one master sense the others defer to. Cue-conflict experiments were meant to establish that ranking and have instead made it look conditional: set a bird's magnetic cue against its sunset cue and sometimes the magnetic reading wins, sometimes the sky recalibrates the magnetic compass, with results shifting by species, season and latitude. Birds plainly arbitrate; the rules of the arbitration are still argued over.
Second, and more sharply: nobody knows how a bird senses magnetism. Two candidates, neither closed. One is chemical — light-sensitive proteins called cryptochromes in the eye, where an absorbed photon creates a pair of molecules whose unpaired electrons respond to the field's angle, so the bird would in some sense see direction. It fits the odd dependence of the magnetic compass on particular wavelengths of light, and cryptochrome from a migratory songbird has been shown magnetically sensitive in a test tube — which is not the same as showing it does the job in a living bird. The other is mechanical: crystals of magnetite acting as tiny needles, reported in the upper beak. That account took a blow when the iron-rich cells in pigeon beaks turned out to be immune-system macrophages rather than sensory neurons. It is not dead, but it is unresolved, and any source stating the mechanism plainly is telling you more than is known.
A third correction: a compass is not a map. Knowing which way is north does not tell you where you are, and adult birds clearly do know — which requires a positional sense, from magnetic intensity and inclination, from remembered coastlines, possibly from odours. Thinner ground still.
A picture of it
THE PICTURE #How to readThe three slanted boxes at the top are independent cues, each arriving from a different part of the world. They all feed one diamond — the bird is not following a cue, it is comparing them. Take the middle branch to see what happens when cloud removes one: the remaining cues still fix the heading, which is why redundancy, not any single sense, is the real answer. The dotted arrow back to the diamond is the nightly re-check. Read the lowest branch as the honest one: what a bird does when cues genuinely disagree is exactly what is still argued about, so the tidy diamond stands in for a resolution nobody can yet write down.
What became clearer
WHAT CLEARED #Birds navigate by consulting several natural maps at once — the sun's arc, the star field, and the Earth's magnetism — cross-checking them, so the journey survives a cloudy sky or a moonless night. What is inherited is narrower than it first appears: a direction and a duration, enough to get a first-time bird to roughly the right region. The map, in the fuller sense of knowing where you are and how to correct for being moved, is built afterwards, from the journey itself.
Where to go next
ONWARD #- How exactly a bird might "see" or feel a magnetic field.
- Why some species learn the route from elders while others are born knowing it.
- What a bird uses for the map half of the problem, rather than the compass half.
Key terms
TERMS #| Term | What it means |
|---|---|
| Magnetoreception | the ability to sense the Earth's magnetic field for direction. |
| Innate behavior | behavior present from birth rather than learned. |
| Emlen funnel | a funnel-shaped cage with a marking surface, in which a caged migrant's restless hops record the direction it is trying to go. |
| Inclination compass | a compass reading the dip angle of the field lines, distinguishing pole from equator rather than north from south. |
| Vector navigation | following an inherited heading for an inherited length of time, as first-year migrants appear to do. |
Every term the collection defines is gathered in the glossary.