The second component of navigation is a system by which the animal can determine the geographical direction of its goal from its current location. If such a "map" mechanism was based on variations in the geomagnetic field, its receptors would have to be sensitive to very minute changes in intensity and/or angle of the Earth's magnetic field (total intensity 50 mT). The intensity of the natural ambient magnetic field changes about 6–12 nT/km along a north-south axis. Overlying this general gradient are several sources of noise or interference. One source of magnetic noise is the generally uniform variation of about 50–100 nT that results from the daily fluctuations in intensity of the magnetic field caused by the day/night cycle of the sun. A second source of magnetic noise is spatial variations in the magnetic field caused by irregularities in the composition of the Earth's crust. These variations might interfere with a bird's ability to determine the small intensity changes that occur as it moves from location to location.
Behavioral responses of homing pigeons indicate they can distinguish among magnetic fields that differ in intensity by about 10–30 nT (Keeton et al., 1974; Larkin and Keeton, 1978; Wiltschko et al., 1986; Kowalski et al., 1988; Becker et al., 1991). Pigeons whose eyes are covered with frosted lenses are able to return to within 2 km of their home loft, apparently using only nonvisual cues (Schmidt-Koenig and Walcott, 1978). Pigeons released within a magnetic anomaly appear to be trapped and escape from the anomaly only by chance, after which they fly home (Walcott, 1978). Pigeons released repeated from the same location morning and afternoon show differences in azimuth between the two releases that are correlated with the differences in magnetic field intensity (Becker et al., 1991).
The sensitivities to magnetic field variations are similar to those predicted for a magnetite-based receptor (Yorke, 1979, 1981; Kirschvink, 1983). Because it has a permanent magnetic moment, single-domain magnetite is attractive theoretically for use in magnetoreception (Kirschvink and Gould, 1981). Single domain magnetite has the strongest magnetic moment of any naturally occurring compound and is strong enough to overcome thermal agitation despite the small particle sizes ( Banerjee and Moskowitz, 1985). The sensitivity of such a receptor could be enhanced by arranging the particles into a closely spaced chain. Because it has a permanent magnetic moment, the use of SD magnetite can be tested with remagnetization.
Applying a strong magnetic pulse (0.5 T, 5–10 ms) to pigeons and migratory birds usually results in a change in their orientation, in some cases a 90° clockwise or counterclockwise rotation from the original direction (Wiltschko et al., 1994; Beason et al., 1995, 1997; Wiltschko and Wiltschko, 1995). Because the response of birds to different treatments was to select different directions, this treatment might be interpreted as affecting the magnetic compass receptor. However, the response of homing pigeons to some treatments was dependent on how far they were released from the home loft (Beason et al., 1997). Birds released close to the loft were oriented the same as the control birds. The differences between headings of control and magnetized pigeons increased as the distance between the release site and the home loft increased (Fig. 5). Pigeons released about 200 km away were oriented almost directly away from home and few birds returned home (Beason et al., 1997). A compass receptor's response to the treatment should be similar at all locations and should not be influenced by the distance of the release site from home. Such site independent effects is the response that is seen to manipulations of the sun-compass (Schmidt-Koenig, 1965, 1979; Keeton, 1974).
A second set of experiments also indicates that the magnetization treatments do not affect a receptor for the magnetic compass. Young Tasmanian Silvereyes (Zosterops lateralis) show no effect of magnetization treatment during their first migratory journey (Munro et al., 1997) whereas adults do (Wiltschko et al., 1994). Adults are goal oriented in their migration, typically returning to the same locations repeatedly for the nonbreeding season. This ability requires them to use a navigational system analogous to a map. Young birds do not know the nonbreeding grounds because they have not been there. Instead, they use an endogenous program of direction and distance when they leave their breeding grounds. Hence, they do not possess a map, but do possess and use a compass. Although treatment with a magnetic pulse results in directional changes in orientation in adult birds, these changes do not appear to be associated with the magnetic compass. By elimination, the effects must be on a magnetic "map" or some similar location system.
Third, when the ophthalmic nerve of Bobolinks was blocked with lidocaine, the effect of the magnetization treatment was reversed. The birds were not disoriented but selected the same headings they used before being magnetized (Beason and Semm, 1996). Consequently, the birds had a functional magnetic compass receptor that was not associated with the ophthalmic nerve and was not affected by the magnetization treatment.
The nerve block experiments were conducted on the ophthalmic branch of the trigeminal nerve of the Bobolink because electrophysiological recordings indicated the presence of single units that responded to changes in the intensity of the magnetic field around the bird. The responses can be divided into fast adapting units and slowly or nonadapting units. Fast adapting units responded with a burst of action potentials when the intensity of the ambient magnetic field was changed. In the Bobolink these cells show a minimum sensitivity of 30–50 nT and a logarithmic response to intensities up to 50 mT (Fig. 6). These cells also responded to AC stimuli and to a handheld magnet moved towards the bird. Slowly adapting units responded as amplitude detectors. The change in neural activity persisted for several minutes or the duration of the stimulus. The minimum sensitivity of these cells was not tested. Most of the units tested responded to the change in magnetic field intensity with an increase in the rate of firing. Some responded with depressed activity (Beason and Semm, 1987; Semm and Beason, 1990).
Thus, the sensory capability of the magnetoreceptors associated with the trigeminal nerve can account for the behavioral sensitivity to small changes in the magnetic fields observed in birds and what would be required to use the Earth's magnetic field for a map or some type of geographic location system. The types of responses indicate the nervous system has the capability to measure small changes in the magnetic field (fast adapting units) and to measure the absolute intensity of the magnetic field (slowly adapting units). These capabilities would allow the bird to determine the intensity of the magnetic field at its location in order to compare the intensity with a remembered value from home. This information would tell the bird which direction along the magnetic gradient to fly towards home. The sensitivity would allow the bird to determine the changes in magnetic intensity as it flew.
Blocking or cutting the ophthalmic nerve of pigeons prevented them from detecting large magnetic anamolies during discrimination experiments (Mora et al., 2004). In these experiments the artificial magnetic fields were several times the strength of the natural Geomagnetic field. To date there is no published documentation of units in the pigeon trigeminal nervous system that respond to changes in magnetic stimuli. However, candidate magnetic sensitive structures have been reported from the ethmoidal region, which receives enervation by the ophthalmic nerve (Fleissner et al., 2003; Walcott and Walcott, 1982; Williams and Wild, 2001).
Single domain magnetite has been reported for the pigeon (Walcott et al., 1979) and the Bobolink (Beason and Brennan, 1986) and superparamagnetic magnetite (SPM) has reported for the pigeon (Fleissner et al., 2003; Hanzlik et al., 2000). The use of single domain magnetite is consistent with the results of the pulse magnetization experiments. Because it posses a permanent magnetic moment magnetization could reverse the alignment of the particle's magnetic moment. A particle attached to a membrane or cilia would move in response to changes in the external magnetic field. Such movement would be conducted to the cell membrane, opening or closing ion channels (Kirschvink and Gould, 1981; Semm and Beason, 1990). Remagnetization would produce different responses from the receptor than before treatment. Superparamagnetic particles, on the other hand, would use a different mechanism because the magnetic moments are not permanent, they orient with the external field. One theoretical mechanism has been proposed based on several SPM particles contained within a capsule-like structure (Shcherbakov and Winklhofer, 1999). The size of the capsules would change as the orientation of the surrounding magnetic field changes. The proposed structures closely resemble those reported in association with nervous fibers in the pigeon (Fleissner et al., 2003).
The theoretical sensitivity of a receptor based on single domain magnetite is sufficient to account for both the behavioral and physiological sensitivities observed in pigeons and migratory birds (Yorke, 1979, 1981; Kirschvink and Gould, 1981). The sensitivity of a SPM magnetite-based receptor appears to be adequate to measure the intensity for a compass mechanism (Shcherbakov and Winklhofer, 1999; Hanzlich et al., 2000) but not for a magnetic map receptor.