Magnetoreception is a sense which allows an organism to detect the Earth's magnetic field. Animals with this sense include some arthropods, molluscs, and vertebrates (fish, amphibians, reptiles, birds, and mammals). The sense is mainly used for orientation and navigation, but it may help some animals to form regional maps. Experiments on migratory birds provide evidence that they make use of a cryptochrome protein in the eye, relying on the quantum radical pair mechanism to perceive magnetic fields. This effect is extremely sensitive to weak magnetic fields, and readily disturbed by radio-frequency interference, unlike a conventional iron compass.

Birds have iron-containing materials in their upper beaks. There is some evidence that this provides a magnetic sense, mediated by the trigeminal nerve, but the mechanism is unknown.

Cartilaginous fish including sharks and stingrays can detect small variations in electric potential with their electroreceptive organs, the ampullae of Lorenzini. These appear to be able to detect magnetic fields by induction. There is some evidence that these fish use magnetic fields in navigation.

Biologists have long wondered whether migrating animals such as birds and sea turtles have an inbuilt magnetic compass, enabling them to navigate using the Earth's magnetic field. Until late in the 20th century, evidence for this was essentially only behavioural: many experiments demonstrated that animals could indeed derive information from the magnetic field around them, but gave no indication of the mechanism. In 1972, Roswitha and Wolfgang Wiltschko showed that migratory birds responded to the direction and inclination (dip) of the magnetic field. In 1977, M. M. Walker and colleagues identified iron-based (magnetite) magnetoreceptors in the snouts of rainbow trout. In 2003, G. Fleissner and colleagues found iron-based receptors in the upper beaks of homing pigeons, both seemingly connected to the animal's trigeminal nerve. Research took a different direction in 2000, however, when Thorsten Ritz and colleagues suggested that a photoreceptor protein in the eye, cryptochrome, was a magnetoreceptor, working at a molecular scale by quantum entanglement.

In animals, the mechanism for magnetoreception is still under investigation. Two main hypotheses are currently being discussed: one proposing a quantum compass based on a radical pair mechanism, the other postulating a more conventional iron-based magnetic compass with magnetite particles.

According to the first model, magnetoreception is possible via the radical pair mechanism, which is well-established in spin chemistry. The mechanism requires two molecules, each with unpaired electrons, at a suitable distance from each other. When these can exist in states either with their spin axes in the same direction, or in opposite directions, the molecules oscillate rapidly between the two states. That oscillation is extremely sensitive to magnetic fields. Because the Earth's magnetic field is extremely weak, at 0.5 gauss, the radical pair mechanism is currently the only credible way that the Earth's magnetic field could cause chemical changes (as opposed to the mechanical forces which would be detected via magnetic crystals acting like a compass needle).

In 1978, Schulten and colleagues proposed that this was the mechanism of magnetoreception. In 2000, scientists proposed that cryptochrome – a flavoprotein in the rod cells in the eyes of birds – was the "magnetic molecule" behind this effect. It is the only protein known to form photoinduced radical-pairs in animals. The function of cryptochrome varies by species, but its mechanism is always the same: exposure to blue light excites an electron in a chromophore, which causes the formation of a radical-pair whose electrons are quantum entangled, enabling the precision needed for magnetoreception.

Many lines of evidence point to cryptochrome and radical pairs as the mechanism of magnetoreception in birds: