Abstract
Even though previously described iron-containing structures in the upper beak of pigeons were almost certainly macrophages, not magnetosensitive neurons, behavioural and neurobiological evidence still supports the involvement of the ophthalmic branch of the trigeminal nerve (V1) in magnetoreception. In previous behavioural studies, inactivation of putative V1-associated magnetoreceptors involved either application of the surface anaesthetic lidocaine to the upper beak or sectioning of V1. Here, we compared the effects of lidocaine treatment, V1 ablations and sham ablations on magnetic field-driven neuronal activation in V1-recipient brain regions in European robins. V1 sectioning led to significantly fewer Egr-1-expressing neurons in the trigeminal brainstem than in the sham-ablated birds, whereas lidocaine treatment had no effect on neuronal activation. Furthermore, Prussian blue staining showed that nearly all iron-containing cells in the subepidermal layer of the upper beak are nucleated and are thus not part of the trigeminal nerve, and iron-containing cells appeared in highly variable numbers at inconsistent locations between individual robins and showed no systematic colocalization with a neuronal marker. Our data suggest that lidocaine treatment has been a nocebo to the birds and a placebo for the experimenters. Currently, the nature and location of any V1-associated magnetosensor remains elusive.