More or less pronounced lateralization is a common feature of the avian brain9, but an all-or-nothing lateralization like the one reported by Wiltschko et al. in European Robins1 and Silvereyes4, Zosterops lateralis, would be highly unusual for any sensory system and seems evolutionarily counterproductive. A bird having a magnetic compass located exclusively in its right eye would be more easily affected by eye infection or monocular damage than a bird having functional magnetic compasses in both eyes.

We therefore tested 27 European Robins during autumn migration, when they use simple compass orientation10, and equipped them with light tight8,11 hoods enabling them to see with both eyes, their right eye only, or their left eye only. In all three conditions, the birds oriented in their expected autumn migratory direction towards the South-West in the unchanged geomagnetic field (normal magnetic field, NMF; both eyes open: 236° ± 20° (95% confidence intervals), r = 0.69, N = 27, P < 0.001, Fig. 1a; left eye open: 217° ± 27°, r = 0.57, N = 27, P = 0.001, Fig. 1c; right eye open: 192° ± 24°, r = 0.65, N = 26, P < 0.001, Fig. 1e) and towards the East in a magnetic field turned 120° counter-clockwise (changed magnetic field, CMF; both eyes open: 78° ± 20°, r = 0.72, N = 27, P < 0.001, Fig. 1b; left eye open: 47° ± 45°, r = 0.38, N = 26, P < 0.03, Fig. 1d; right eye open: 112° ± 30°, r = 0.52, N = 27, P = 0.001, Fig. 1f). In all cases, the CMF direction is significantly (no 95% confidence intervals overlap) turned in the expected direction compared to the NMF direction.

Figure 1: European Robins wearing eye covers can use their magnetic compass if light and/or visual input reaches any one eye.
figure 1

af, Each dot at the circle periphery represents the mean orientation of one individual bird tested several times with the given type of hood. mN, magnetic North. The arrows indicate the group mean vectors. The inner and outer dashed circles indicate the radius of the group mean vector needed for significance according to the Rayleigh Test (P < 0.05 and P < 0.01, respectively). The lines flanking the group mean vector indicate the 95% confidence intervals for the group mean direction.

PowerPoint slide

Our results showing that European Robins have a magnetic compass in both eyes are in line with other recent findings, which otherwise would be difficult to explain: (1) garden warblers have a magnetic compass in both eyes11; (2) the putative magnetoreceptive cryptochromes are located in both eyes6; (3) Cluster N7,8, the brain area recently shown to be necessary for magnetic compass orientation in European Robins2, shows similar activation in both brain hemispheres during magnetic compass orientation7,12. In fact, Cluster N activation in European Robins shows a slight but significant dominance of the left eye and right brain hemisphere8, that is, lateralization in the opposite direction to that suggested by Wiltschko et al.1,4; (4) the neuronal pathways between the eye and Cluster N seem to be symmetrical13; (5) magnetic compass orientation is only weakly lateralized in pigeons14,15. We suggest that the Wiltschko et al.1 data may have been artefacts of the unnatural green light conditions under which their birds were tested or of the non-blinded procedures. Alternatively, they might have resulted from the more complicated interaction of map and compass information potentially occurring in spring.

In conclusion, it is very possible that some smaller degree of lateralization of magnetic information processing exists in birds8,14,15. However, our data show that the magnetic compass of night-migratory songbirds is not strongly lateralized and certainly not located in only one of the birds’ eyes.


We tested the birds’ magnetic compass orientation capabilities under broad spectrum white light2 in the normal geomagnetic field (NMF) and in a changed geomagnetic field with magnetic North turned 120° counter-clockwise (CMF). We used a double-blind protocol and large, three-dimensional, double-wrapped, Merritt 4-coils to produce highly homogenous magnetic fields (for details see ref. 2). The same current ran through the coils in both magnetic field conditions. We tested all birds inside aluminium-lined wooden huts, where no cues other than the geomagnetic field were available. The mean directions are based on 4.11 ± 2.76 (s.d.) active and oriented tests per condition (six conditions).