Module 2: Magnetoreception — Quantum Biology and Magnetite
Night-migrating songbirds detect the direction of the Earth’s 25–65 μT magnetic field with a sensitivity that approaches the quantum limit. This module derives the two leading biophysical mechanisms for this remarkable sense: the cryptochrome-based radical-pair mechanism of Schulten & Swenberg (1976), updated by Ritz (2000) and anchored in the CRY4 photoreceptor localised to avian retinal cones (Xu et al., 2021), and the magnetite hypothesis of Fleissner et al. (2003), challenged by Treiber et al. (2012). We write down the spin Hamiltonian, simulate the Liouville-von Neumann dynamics producing an angle-dependent singlet yield, and contrast the Wiltschko inclination compass with the polarity compass of magnetite-based detection.
1. The Wiltschko inclination compass
Wolfgang and Roswitha Wiltschko, working at Frankfurt in the late 1960s, made the first definitive demonstration that European robins (Erithacus rubecula) use the Earth’s magnetic field as a compass (Wiltschko 1968; Wiltschko & Wiltschko 1972). Birds in Emlen funnels shielded from celestial cues still oriented toward their population-typical heading; when a Helmholtz coil rotated the ambient horizontal field by 90°, the birds rotated with it by the same amount.
Inclination, not polarity
A key discovery was that birds do not read the polarity of the field. Inverting \(\mathbf B \to -\mathbf B\) does not reverse the bird’s heading, whereas inverting the vertical component \(B_z \to -B_z\) while preserving \(B_x, B_y\) does. The compass therefore reads only the dip angle \(|I|\) between the field vector and the horizontal plane. At the equator where \(I = 0\) the inclination compass is ambiguous, and migrants at low latitudes appear to disengage the compass or switch to other cues.
\[ \tan I = \frac{B_z}{\sqrt{B_x^2 + B_y^2}}\,,\quad |F|=\sqrt{B_x^2+B_y^2+B_z^2} \]
inclination \(I\) and intensity \(|F|\) as geomagnetic coordinates. Earth’s dipole gives \(\tan I \approx 2\tan\varphi_m\) with magnetic latitude \(\varphi_m\).
Inclination vs polarity compass
In contrast, homing pigeons can in some experiments read polarity, pointing to an additional receptor class distinct from the one used by songbirds during migration. Salmon, sea turtles and lobsters exhibit their own variants: sockeye salmon use intensity alone to find natal rivers, while sea turtles use a geomagnetic coordinate vector combining intensity and inclination.
2. The radical-pair mechanism
Schulten, Swenberg & Weller (1978) proposed that a chemical reaction could be sensitive to the Earth’s magnetic field through the spin dynamics of a photogenerated radical pair. Ritz, Adem & Schulten (2000) developed the hypothesis into a quantitative model of the avian compass, with cryptochrome as the candidate photoreceptor. The biochemistry is by now reasonably well established: in vertebrate photoreceptors, blue-light absorption by the FAD chromophore of cryptochrome causes electron transfer along a chain of three tryptophans, generating a geminate radical pair \(\bigl[\text{FAD}^{\bullet-} \cdots \text{Trp}^{\bullet+}\bigr]\) whose spin state evolves coherently before recombination.
Spin Hamiltonian
The two radical electrons, plus surrounding magnetic nuclei (H, N, C), are governed by
\[ \hat H = g\mu_B\,\mathbf B\!\cdot\!(\hat{\mathbf S}_1 + \hat{\mathbf S}_2) + \sum_{k} \hat{\mathbf S}_1\!\cdot\mathbf A_k\!\cdot\hat{\mathbf I}_k + \sum_{l}\hat{\mathbf S}_2\!\cdot\mathbf A_l\!\cdot\hat{\mathbf I}_l \]
first term: Zeeman; remaining terms: anisotropic hyperfine couplings. No electron-electron exchange at the distances typical in cryptochrome.
Starting from a pure singlet state \(|\Psi(0)\rangle = |S\rangle \otimes |\phi_N\rangle\) the coherent evolution under \(\hat H\) mixes singlet and triplet manifolds. Hyperfine couplings break the isotropy of the Zeeman interaction and make the mixing rate depend on the field direction. Recombination through spin-selective channels (kS for singlet, kT for triplet) imprints this direction-dependence onto the chemical yield.
Singlet yield
\[ \phi_S(\theta,\varphi) = k_S \int_0^\infty \!\! \mathrm{Tr}\!\bigl[\hat P_S\,\hat\rho(t)\bigr] e^{-(k_S+k_T)t/2}\,dt \]
angle-dependent chemical compass signal: the fraction of radical pairs that recombine via the singlet channel.
Anisotropic hyperfine tensors are the source of the angular dependence. For a single nitrogen-14 nucleus with hyperfine coupling \(A_{zz} \gg A_{xx}, A_{yy}\), the singlet yield has a characteristic angular contrast of several percent between parallel and perpendicular field orientations, sufficient for behavioural detection when integrated across many pairs. Detailed calculations with realistic hyperfine tensors for FAD and Trp give angular contrasts in the 2–6% range (Hiscock et al., 2016), consistent with behavioural thresholds.
Coherence lifetime
A viable radical-pair compass requires a coherence lifetime long enough that hyperfine- mediated mixing has time to sample the field angle. Theoretical estimates give \(\tau_{\text{coh}} \gtrsim 0.1\) \mu s, with the best current measurements suggesting \(\tau_{\text{coh}}\) on the order of microseconds. This is astonishingly long for a warm, wet protein; it is enabled by the careful tuning of hyperfine magnitudes and by the spatial separation of the two radicals, which suppresses dipolar and exchange couplings that would otherwise scramble the spin state.
3. Cryptochromes and CRY4
Cryptochromes are blue-light-absorbing flavoproteins of the photolyase superfamily, named for their originally cryptic biological function. Mammalian CRY1/CRY2 are core components of the circadian transcriptional feedback loop; invertebrate CRY functions in circadian photoentrainment. Drosophila magnetoreception experiments (Gegear et al., 2008) first linked cryptochromes to magnetosensitivity. For birds, four paralogs are known—CRY1a, CRY1b, CRY2 and CRY4—of which CRY4 has emerged as the prime suspect for the avian compass receptor.
Xu et al. 2021
Xu, Jarocha, Henbest, Lagator, Hore, Mouritsen & co-workers published a landmark paper (Nature, 2021) purifying CRY4 from the European robin (Erithacus rubecula), showing that it forms a long-lived photo-induced radical pair with a coherence lifetime in the microsecond range, and measuring a weak but significant magnetic-field effect on the singlet yield in vitro at physiological fields. The observed magnitude of magnetic-field effect scales as predicted by the radical-pair mechanism. Crucially, CRY4 proteins from migratory Erithacus and resident Gallus gallus (chicken) differ in their magnetic-field response, with the migrant-species variant showing stronger sensitivity.
Localisation in the retina
CRY4 is expressed in the outer segments of double-cone photoreceptors in the avian retina, a geometry consistent with its proposed role: the photoreceptor array allows the bird to read the spatially resolved angular signal by tilting its head, effectively scanning the singlet-yield pattern over visual space (Wiltschko & Wiltschko 2005; Hore & Mouritsen 2016).
Radiofrequency disruption
A powerful experimental test of the radical-pair hypothesis is that a weak (<50 nT) RF field tuned to the frequencies of the hyperfine couplings should disrupt the singlet- triplet mixing and disorient the bird. Ritz, Thalau, Phillips, Wiltschko & Wiltschko (Nature, 2004) and Engels et al. (Nature, 2014) demonstrated this disruption, with a resonance near 1.4 MHz corresponding to the Larmor frequency of free electrons at the Earth’s field. Resident chickens showed no such disruption in parallel experiments, consistent with their reduced need for the compass.
CRY4 radical-pair photocycle
4. The magnetite hypothesis and its controversy
An alternative mechanism invokes biogenic single-domain magnetite (Fe₃O₄) crystals as the transducer. Magnetite-based magnetoreception is robustly established in magnetotactic bacteria, and single-domain magnetite has been isolated from salmon, bees, trout and sea turtles. Fleissner, Stahl, Thalau, Falkenberg & Fleissner (2003) claimed to identify iron-rich structures in the upper-beak skin of homing pigeons containing superparamagnetic magnetite and proposed these as the basis of a trigeminal-nerve-mediated magnetometer.
Treiber 2012 controversy
Treiber, Salzer, Riegler, Edelman & co-workers (Nature, 2012) re-examined Fleissner’s beak structures using higher-resolution histology and electron microscopy and concluded that the iron-rich cells were macrophages, not neurons, and therefore unlikely to be sensory receptors. Subsequent work (Edelman et al., 2015; Lefeldt et al., 2014) has been unable to replicate the original claim. The consensus by 2024 is that the beak magnetite hypothesis for homing pigeons is not supported, although the broader idea that some magnetite-based sense might operate in other species (notably bony fish and sea turtles) remains credible.
Mechanical transduction
Regardless of the anatomical controversy, the biophysics of magnetite magnetoreception is well posed. A single-domain particle of volume \(V\) and saturation magnetisation \(M_s\) carries a moment \(m = M_s V\) and experiences a torque
\[ \vec\tau = \vec m \times \vec B\,,\quad E_{\text{mag}} = -\vec m\!\cdot\!\vec B = -mB\cos\theta \]
The dimensionless alignment parameter \(\xi = mB/k_BT\) governs the thermally averaged alignment via the Langevin function \(\langle\cos\theta\rangle = L(\xi)\). For a 50-nm cubic magnetite particle at body temperature in the Earth’s field, \(\xi \approx 100\), and alignment is essentially complete. A chain of such particles embedded in a cytoskeletal matrix rotates with the field, stretching mechanosensitive ion channels to produce a transduction current.
Polarity vs intensity readout
Unlike the radical-pair compass, magnetite readout naturally reads both polarity and intensity. This is part of the reason why it is thought to underlie the magnetic-map component of navigation (position sensing) rather than the inclination compass (direction sensing). Lohmann et al. (2008) have developed extensive evidence for such a magnetic map in sea turtles based on inherited geomagnetic coordinates.
5. Two senses, two jobs
The consensus emerging from the last fifteen years of work (reviewed by Hore & Mouritsen 2016 in Annual Review of Biophysics) is that many migratory vertebrates possess two magnetic senses serving different functions. The radical-pair / cryptochrome mechanism provides a light-dependent inclination compass used for heading, with angular but not polarity discrimination. A separate magnetite-basedmechanism—possibly in the lagena, the trigeminal ophthalmic branch, or the olfactory epithelium depending on taxon—provides a light-independent magnetometer used for the magnetic map, sensitive to intensity and inclination gradients at the regional scale.
Evolutionary constraints
If CRY4 is the magnetoreceptor, then migratory vs resident phylogeny should leave a selective signature. Einwich et al. (2020) and Bolte et al. (2021) compared CRY4 sequences across Passeriformes and found elevated nonsynonymous-to-synonymous ratios on branches leading to obligate migrants, consistent with positive selection on magnetic sensitivity in those lineages. By contrast, CRY1 and CRY2, which underpin the circadian clock, show pervasive purifying selection.
Open questions
- What is the primary transduction partner of cryptochrome downstream of the radical-pair formation—a G-protein cascade, a kinase, or a direct ion-channel interaction?
- How does the bird integrate the spatially resolved signal across the retina into a single heading estimate?
- Where is the light-independent magnetite receptor (if it exists) in the songbird nervous system?
- Can quantum coherence in CRY4 be demonstrated in vivo, not merely in vitro?
6. The Earth’s magnetic field as a navigation signal
To first approximation the Earth’s field is an axial dipole tilted ~11° from the rotation axis. The IGRF (International Geomagnetic Reference Field) captures the non-dipole and secular-variation terms. At the surface the total intensity ranges from \(\sim 25\) \mu Tat the magnetic equator to \(\sim 65\) \mu Tat the poles, and inclination runs from 0 at the equator to ±90° at the poles approximately as \(\tan I = 2\tan\varphi_m\).
\[ |F(\varphi_m)| = F_0\,\sqrt{1 + 3\sin^2\varphi_m}\,,\quad F_0 \approx 31\,\mu T \]
Because the non-dipole terms give a complex, non-monotonic intensity-inclination landscape, intensity alone is not a single-valued coordinate. Migrants using both inclination and intensity can resolve unique loci; Lohmann’s turtles and Fischer et al.’s loggerhead experiments confirm that this bicoordinate map is operational in vertebrates with magnetite receptors.
Secular variation and plasticity
The geomagnetic field is not static. Secular variation shifts the magnetic poles at rates of tens of kilometres per year, and geomagnetic excursions (such as the Laschamp excursion ~41 ka BP) have been implicated in brief periods of enhanced extinction risk. Migrants with innate magnetic headings anchored to a fixed intensity-inclination pair would fail during excursions; selection therefore favours plasticity and reliance on multiple cues, including celestial backup.
7. Cluster N and the neural circuit
The downstream neural pathway that converts the CRY4 signal into a heading estimate is partially mapped. Heyers, Manns, Luksch, Güntürkün & Mouritsen (2007) identified Cluster N, a small forebrain region active only during nocturnal migratory flight in European robins and garden warblers. Lesions of Cluster N disrupt the magnetic compass but leave sun- and star-compass orientation intact (Zapka et al. 2009, Nature). Cluster N receives visual input via thalamofugal relays from the retina, consistent with a light-dependent compass originating in retinal cryptochrome.
The resting-state connectivity of Cluster N is conserved across Passeriformes but its activation during nocturnal flight is absent in non-migratory species. Immediate-early gene expression (ZENK, Fos) patterns during nighttime restlessness in captive migrants trace a circuit from the retina through Cluster N to the entopallium and ultimately to motor pattern generators driving orientation behaviour.
Trigeminal branch and the map sense
A separate pathway through the ophthalmic branch of the trigeminal nerve has been implicated in the magnetite-based map sense. Mora, Davison, Wild & Walker (2004) showed that sectioning the ophthalmic branch impairs pigeon homing over long distances while leaving short-range orientation intact. Whether a functional magnetite receptor lies in the beak, the lagena, or some other location, the dual-pathway architecture (retinal cryptochrome for heading, trigeminal magnetite for map) is now the standard working model.
Simulation 1: Radical-pair compass signal vs field angle
We simulate a minimal two-electron, one-nucleus radical pair with anisotropic hyperfine tensor using the Liouville-von Neumann equation in the full 8-dimensional Hilbert space. Spin-selective recombination kinetics at rates \(k_S = k_T = 10^6\) s^{-1}with coherence lifetime \(\tau_{\text{coh}} \approx 1\,\mu s\) produce an angle-dependent singlet yield \(\phi_S(\theta)\) with the characteristic few-percent contrast of realistic cryptochrome models. We add the field-magnitude tuning and a toy RF-disruption spectrum reproducing the Engels et al. (2014) 1.4 MHz resonance.
Click Run to execute the Python code
Code will be executed with Python 3 on the server
Simulation 2: Magnetite inclinometer under dipole gradient
We compute the latitudinal variation of the Earth’s dipole field—total intensity, inclination, and horizontal/vertical components—and plug these into the Langevin-function alignment of a 50 nm single-domain magnetite particle. The resulting alignment is essentially saturated at all latitudes, confirming that thermal noise is not the limiting factor for magnetite-based transduction. Finally we implement a Wiltschko-style inclination compass comparing the local dip angle to a learned target and highlight the equatorial stopover zone where the compass becomes ambiguous.
Click Run to execute the Python code
Code will be executed with Python 3 on the server
7b. Behavioural threshold of the avian compass
How weak a field can a bird still orient to? Wiltschko & Wiltschko (1978) established that European robins orient only in a narrow intensity window around 46 μT, disorient when the field is reduced to 39 μT or raised to 55 μT, and re-orient after several hours of habituation to a new ambient field. This intensity tuning is one of the strongest phenomenological fingerprints of the compass and is difficult to explain with magnetite-only models but arises naturally from a radical-pair mechanism in which the optimum hyperfine-Zeeman mixing ratio constrains the operating field.
Liedvogel et al. (2007) measured CRY4 retinal expression patterns across the diel cycle and found that expression is elevated at night in migratory species, coincident with the Cluster N activation window. This temporal correlation is absent in domestic chickens. The time-of-day locking of compass biochemistry is another line of evidence for the compass being anchored in CRY4 rather than in a hypothetical magnetite receptor that would operate in constant light.
\[ F_{\text{min}} \approx 39\,\mu\text{T}\,,\; F_{\text{max}} \approx 55\,\mu\text{T}\,,\; \Delta F_{\text{window}} \approx \pm 30\% \]
operating window of the avian inclination compass (Wiltschko & Wiltschko 1978).
8. Synthesis: a quantum sense at ambient temperature
The radical-pair hypothesis remains one of the most striking claims in biology: that a functionally important sensory signal depends on nanosecond-to-microsecond quantum coherence in a warm, wet protein. The last twenty years have assembled substantial indirect evidence—magnetic-field effects in purified cryptochromes, RF disruption of orientation in vivo, a co-segregation of CRY4 sequence with migratory behaviour, and a neuroanatomical pathway connecting retinal cryptochrome to a dedicated forebrain processing station. No single experiment is decisive, but the cumulative case is strong.
The convergence of magnetite-based and radical-pair-based mechanisms in a single animal suggests that what we call “magnetoreception” is a confederation of sensory modalities, each optimised for a different computational task: inclination compass, intensity map, polarity detector. Module 6 will take up the parallel story in monarch butterflies, where the compass biochemistry has the additional twist of being embedded in an animal whose full migration spans four generations.
Key references
• Bolte, P., Einwich, A., Seth, P. K. et al. (2021). Cryptochrome 1a localisation in photo- and non-photoreceptor cells in the retina of nocturnally migrating night-migratory songbirds. PLoS ONE, 16, e0246021.
• Edelman, N. B. et al. (2015). No evidence for intracellular magnetite in putative vertebrate magnetoreceptors identified by magnetic screening. PNAS, 112, 262–267.
• Einwich, A., Dedek, K., Seth, P. K., Laubinger, S. & Mouritsen, H. (2020). A novel isoform of cryptochrome 4 (Cry4b) is expressed in the retina of a night-migratory songbird. Scientific Reports, 10, 15794.
• Engels, S. et al. (2014). Anthropogenic electromagnetic noise disrupts magnetic compass orientation in a migratory bird. Nature, 509, 353–356.
• Fleissner, G. et al. (2003). Ultrastructural analysis of a putative magnetoreceptor in the beak of homing pigeons. J. Comp. Neurol., 458, 350–360.
• Gegear, R. J., Casselman, A., Waddell, S. & Reppert, S. M. (2008). Cryptochrome mediates light-dependent magnetosensitivity in Drosophila. Nature, 454, 1014–1018.
• Hiscock, H. G. et al. (2016). The quantum needle of the avian magnetic compass. PNAS, 113, 4634–4639.
• Hore, P. J. & Mouritsen, H. (2016). The radical-pair mechanism of magnetoreception. Annu. Rev. Biophys., 45, 299–344.
• Lefeldt, N., Heyers, D., Schneider, N. L., Engels, S., Elbers, D. & Mouritsen, H. (2014). Magnetic field-driven induction of ZENK in the trigeminal system of pigeons. J. R. Soc. Interface, 11, 20140777.
• Lohmann, K. J., Putman, N. F. & Lohmann, C. M. F. (2008). Geomagnetic imprinting: a unifying hypothesis of long-distance natal homing. PNAS, 105, 19096–19101.
• Ritz, T., Adem, S. & Schulten, K. (2000). A model for photoreceptor-based magnetoreception in birds. Biophysical Journal, 78, 707–718.
• Ritz, T., Thalau, P., Phillips, J. B., Wiltschko, R. & Wiltschko, W. (2004). Resonance effects indicate a radical-pair mechanism for avian magnetic compass. Nature, 429, 177–180.
• Schulten, K., Swenberg, C. E. & Weller, A. (1978). A biomagnetic sensory mechanism based on magnetic field modulated coherent electron spin motion. Z. Phys. Chem. NF, 111, 1–5.
• Treiber, C. D. et al. (2012). Clusters of iron-rich cells in the upper beak of pigeons are macrophages not magnetosensitive neurons. Nature, 484, 367–370.
• Wiltschko, W. & Wiltschko, R. (1972). Magnetic compass of European Robins. Science, 176, 62–64.
• Wiltschko, W. & Wiltschko, R. (2005). Magnetic orientation and magnetoreception in birds and other animals. J. Comp. Physiol. A, 191, 675–693.
• Xu, J., Jarocha, L. E., Zollitsch, T. et al. (2021). Magnetic sensitivity of cryptochrome 4 from a migratory songbird. Nature, 594, 535–540.