Magnetoreception

A bird that can see north

Drop a migrating European robin into a windowless room on a cloudy autumn night — no stars, no sun, no landmarks — and it still hops insistently toward the southwest, the direction of its wintering grounds. Rotate the magnetic field around its cage with a pair of copper coils, and it rotates its heading to match. The robin is reading something invisible: the geomagnetic field threading through the room. That sense is magnetoreception, and the strangest part is how it almost certainly works — not with a tiny iron needle, but with a quantum chemical reaction unfolding inside the cells of the bird's eye.

Earth behaves like a colossal bar magnet, generating a field that emerges near the southern hemisphere, wraps around the planet, and re-enters near the north. At the surface this field is feeble — between about 25 microtesla near the equator and 65 microtesla near the poles, roughly a thousandth the strength of a cheap refrigerator magnet. It carries two kinds of information an animal can exploit: direction (which way the field lines point, and crucially how steeply they dip into the ground) and intensity (how strong the field is, which varies smoothly with latitude). Direction gives a compass; intensity plus inclination together give something closer to a map.

The radical-pair mechanism, step by step

The dominant model for the light-dependent compass is the radical-pair mechanism, first proposed by Klaus Schulten in 1978 and refined for cryptochrome by Thorsten Ritz in 2000. Here is the chain of events:

1. A photon arrives. A quantum of blue light (around 450 nm) is absorbed by the flavin adenine dinucleotide (FAD) cofactor buried inside a protein called cryptochrome, expressed in photoreceptor cells of the retina.
2. An electron hops. The excited FAD pulls an electron down a conserved chain of three or four tryptophan residues — the "tryptophan tetrad" — that acts like a molecular wire. This electron transfer happens in nanoseconds.
3. A radical pair is born. The result is two molecules each left with a single unpaired electron: a flavin radical (FAD•⁻) and a tryptophan radical (TrpH•⁺), separated by roughly 1–2 nanometers. Crucially, the two unpaired electron spins start out correlated — in a quantum singlet state, with spins anti-parallel.
4. The spins oscillate. Left alone, the pair's combined spin coherently flickers back and forth between the singlet state (anti-parallel) and the triplet state (parallel). The flickering is driven by hyperfine coupling — the magnetic interaction between each electron and the nuclear spins of nearby hydrogen and nitrogen atoms in the molecule.
5. Earth's field tips the balance. The external geomagnetic field adds a tiny Zeeman interaction that shifts the singlet–triplet interconversion rate. Because the field's effect depends on the orientation of the molecule relative to the field lines, the yield of singlet versus triplet products now depends on which way the bird's head is pointing.
6. Chemistry becomes signal. The singlet and triplet radical pairs recombine into different products and have different lifetimes. Cells downstream read out the ratio. Because cryptochrome tiles the retina, the readout varies across the visual field, so the bird perceives the magnetic field as a faint, direction-dependent shading or pattern superimposed on what it sees.

The genius of the scheme is that it never asks the weak field to do real work. The Zeeman energy Earth supplies an electron spin is on the order of 10⁻⁹ to 10⁻⁸ electronvolts — millions of times smaller than the thermal energy kT (~0.025 eV) that would normally drown out any magnetic effect. The radical pair dodges this because it is a non-equilibrium, spin-coherent system: the field only has to compete with the molecule's own internal hyperfine fields and bias a rate, not push the reaction uphill.

The molecules and structures involved

• Cryptochrome (Cry). A blue-light flavoprotein in the same family as DNA-repair photolyase. Vertebrates have several isoforms; Cry4 is the strongest candidate magnetosensor, expressed in the double-cone and other photoreceptors of the avian retina and upregulated during the migratory season. In 2021, Xu et al. purified robin Cry4 and showed its radical-pair chemistry is magnetically sensitive in vitro — and more sensitive than the chicken or pigeon versions.
• FAD (flavin adenine dinucleotide). The light-absorbing cofactor that becomes one half of the radical pair after photoexcitation.
• The tryptophan tetrad. A chain of three or four tryptophans (W1–W4) that shuttles the electron away from FAD by sequential hops, separating the two radicals far enough that they don't instantly recombine.
• Magnetite (Fe₃O₄). The basis of the second, light-independent mechanism: nanocrystals of iron oxide, ~30–120 nm, that physically respond to the field. Found in magnetotactic bacteria, the ethmoid tissue of salmonid fish, and possibly associated with the trigeminal nerve in birds.
• Magnetosomes. In magnetotactic bacteria, membrane-bound magnetite crystals are strung into chains by a dedicated cytoskeleton (MamK protein), forming a passive compass needle ~1 µm long that aligns the whole cell along field lines.
• The trigeminal and ophthalmic nerves. Carry magnetic information from putative magnetite-based receptors (likely the "map" sense) to the brain, distinct from the visual pathway that would carry the cryptochrome compass.

Radical-pair vs magnetite mechanisms

Property	Radical-pair (cryptochrome)	Magnetite (Fe₃O₄ crystals)
Physical basis	Quantum spin chemistry in a protein	Classical ferromagnetic torque/force
Needs light?	Yes — blue light (~450 nm)	No — works in darkness
Information read	Inclination (axis only)	Polarity and intensity
Best for	Compass (which way to go)	Map (where you are)
Sensor location	Photoreceptors of the retina	Ethmoid / trigeminal tissue
Nerve pathway	Optic / visual	Trigeminal (ophthalmic branch)
Disrupted by weak RF fields?	Yes — diagnostic at ~1.3 MHz	No — too slow to follow MHz
Found in	Birds, monarch butterflies, fruit flies	Magnetotactic bacteria, salmon, pigeons

The numbers that make it work

Quantity	Value	Why it matters
Geomagnetic field strength	25–65 µT	The signal being detected; varies with latitude
Zeeman energy on an electron spin	~10⁻⁹–10⁻⁸ eV	Far below kT — so the mechanism can't rely on energy
Thermal energy kT (310 K)	~0.027 eV	The noise floor the field cannot beat directly
Trigger photon	~450 nm (≈2.75 eV)	Blue light excites FAD to start the reaction
Radical separation	~1–2 nm	Far enough to slow recombination, close enough to stay correlated
Required spin coherence time	≳1 µs	Coherence must outlast the field's interconversion effect
Larmor frequency in Earth's field	~1.3 MHz	RF fields here disrupt the compass — the smoking gun
Magnetite crystal size	~30–120 nm	Single-domain crystals act as permanent compass needles
Intensity discrimination	Differences of a few µT	Sea turtles resolve the small intensity gradients that encode latitude on their magnetic map

Where it shows up across the animal kingdom

• European robin (Erithacus rubecula). The model species. Wolfgang Wiltschko's 1966 funnel-cage experiments first proved magnetic orientation; later work established the light-dependent inclination compass and the radiofrequency-disruption signature.
• Homing and racing pigeons. Carry both systems; iron-rich structures near the beak and a trigeminal pathway suggest a magnetite-based map sense layered on a cryptochrome compass.
• Loggerhead sea turtles (Caretta caretta). Hatchlings inherit a "magnetic map," reading both field intensity and inclination as positional coordinates to complete a multi-year trans-Atlantic loop and return to their natal beach to nest.
• Salmon (Oncorhynchus). Chinook and sockeye imprint on the geomagnetic signature of their home river as juveniles and use it years later to find their way back from the open ocean to spawn.
• Monarch butterflies (Danaus plexippus). Use a light-dependent, cryptochrome-based inclination compass to calibrate their multi-generational migration to central Mexico.
• Magnetotactic bacteria (e.g. Magnetospirillum). The simplest case: chains of magnetosomes passively align the whole cell along field lines, helping these microbes swim to their preferred low-oxygen sediment layer — discovered by Richard Blakemore in 1975.
• Mammals. Mole rats, fruit bats, and even cattle and red foxes show magnetic alignment; foxes preferentially pounce toward magnetic north when hunting in deep snow.

Common misconceptions

• "Animals have a tiny iron compass needle, like a Boy Scout's compass." Sometimes — magnetite-based detection is real — but the avian compass is light-dependent quantum chemistry, not a needle. A needle reads polarity; the bird reads inclination, which a needle alone cannot explain.
• "Earth's field is too weak to affect biology — it must be pseudoscience." The field is indeed far too weak to overpower thermal energy. The whole point of the radical-pair mechanism is that it never tries to. It biases the rate of a coherent quantum process, where the relevant comparison is the molecule's hyperfine fields, not kT.
• "The bird feels a force or a pull." There is no force the animal feels. In the radical-pair model the magnetic field changes a chemical product ratio, which the visual system reads as a pattern. The animal more likely sees the field than feels it.
• "Quantum effects can't survive in a warm, wet cell." They usually don't last long — but they don't have to. The compass only needs spin coherence to persist for about a microsecond, and cryptochrome's protein scaffold appears to protect the radical pair just long enough. That's why magnetoreception is a flagship example of quantum biology.
• "It's a polarity compass like ours, so it should work everywhere." Because it reads inclination (dip angle), the radical-pair compass becomes ambiguous or useless near the magnetic equator, where field lines run horizontal and the dip angle approaches zero. Migratory birds crossing the equator must switch reference frames.
• "We've proven cryptochrome is THE sensor." Not quite. The in-vitro magnetic sensitivity of Cry4 and the radiofrequency-disruption behavior are strong circumstantial evidence, but a clean in-vivo demonstration that cryptochrome's radical pairs drive a live bird's compass is still an open frontier.