For what must have felt like an interminable six months back in 2007, Sabine Begall spent her evenings at her computer, staring at photographs of grazing cattle. She would download a satellite image of a cattle range from Google Earth, tag the cows one by one, then pull up the next image. With the help of her collaborators, Begall, a zoologist at the University of Duisburg-Essen in Germany, ultimately found that the unassuming ruminants were on to something. On average, they appeared to align their bodies with a slight preference toward the north-south axis. But they were not pointing to true north, which they could have located using the sun as reference. Instead they somehow knew how to orient themselves toward the magnetic north pole, which is hundreds of kilometers south of the geographic pole, in northern Canada.
A follow-up study found more evidence that animals as large as cows can react to the earth’s magnetic field: the aligning behavior vanished in the vicinity of high-voltage power lines that drowned out the relatively subtle signals coming from the planet.
Until a few decades ago studies such as Begall’s would have been met with derision. Everyone knew that organic matter does not respond to weak magnetic fields such as the earth’s and that animals do not come equipped with bar magnets to use as compasses. Franz Anton Mesmer’s 18th-century belief in “animal magnetism”—the notion that breathing creatures harbor magnetic fluids in their bodies—had long been relegated to the annals of charlatanism.
Today the scientific community accepts that certain animals do read and respond to magnetic fields and that, for many of them, being able to do so should be helpful to survival—although why cattle would want to align magnetically is still mysterious. A magnetic sense has, in fact, been well documented in dozens of species—from seasonal migrants such as robins and monarch butterflies to expert navigators such as homing pigeons and sea turtles; from invertebrates such as lobsters, honeybees and ants to mammals such as mole rats and elephant seals; and from tiny bacteria to humongous whales.
What no one knows for sure yet is exactly how creatures other than bacteria do it. Magnetism is “the one sense that we know the least about,” notes Steven M. Reppert, a neurobiologist at
the University of Massachusetts Medical School in Worcester.
In the past decade or so, however, collaborations of biologists, earth scientists and physicists have begun to propose plausible mechanisms and to pinpoint candidate anatomical structures where those mechanisms may be at play. None of these ideas has yet gained the acceptance of the full scientific community, but the experimental evidence found so far is truly mesmerizing. Some animals may even harbor more than one type of magnetic organ. And whereas certain biological magnetic field sensors seem to behave much like ordinary bar magnet compasses, others may well be rooted in subtle quantum effects.
The subject continues to have its share of controversy. But increased interest in magnetic reception and rapid improvements in experimental techniques could lead investigators to solve the mystery of this unusual sense in the next few years.
The Urge to Migrate
The first modern hints that animals sometimes use magnetic fields to guide their behavior emerged about half a century ago. Researchers had noticed since the 1950s that in autumn caged European robins seemed to want to escape toward the south—where they would usually migrate—even if they had no visual cues as to where south was. Then, in the mid-1960s, Wolfgang Wiltschko, a student in biology at Goethe University in Frankfurt, demonstrated that electromagnetic coils wrapped around the birds’ cage could trick them into trying to flee in the wrong direction. His was probably the first evidence of a magnetic sense, and the reaction was predictably skeptical. “When I found that the magnetic field plays a role in the orientation of robins, more or less nobody believed that,” says Wiltschko, who recently retired from his professorship at Goethe.
Shortly after the discovery, Wiltschko met his future wife and lifelong scientific collaborator, Roswitha. The couple has studied avian magnetic detection ever since, working mostly with robins they capture with nets near their laboratories. The Wiltschkos began to publish the results of their joint investigations in 1972, when they revealed that robins are sensitive not only to the geographic direction of the magnetic north but also to the inclination of the earth’s magnetic field relative to a horizontal plane.
The geomagnetic field’s inclination varies continuously from pole to pole. At the magnetic south pole it points straight up, whereas at the magnetic north pole it points straight down; roughly halfway, along a “magnetic equator,” it is horizontal. An ordinary compass needs to balance its needle horizontally, and thus it cannot measure the field’s inclination, responding only to its side-to-side component. Birds—and, it turns out, other animals—can do better and probably use inclination to roughly estimate their distance from the magnetic poles.
Variations in the earth’s field are not limited to inclination changes from one pole to the other. Magnetic minerals in the earth’s crust produce local anomalies in both direction and strength. Some animals—notably, sea turtles—seem to have a mental map of those anomalies that helps them know not just where north is but what their position is relative to their destination. Kenneth J. Lohmann of the University of North Carolina at Chapel Hill and his collaborators have found that captured sea turtles tend to respond to artificial magnetic fields that simulate conditions at various locations along their migrating routes. The turtles attempt to swim in the direction that would lead them to their destination starting from those locations. To have such a magnetic map sense, an animal probably needs to detect not only the inclination anomalies of the field but also its varying strength.
Some researchers believe that birds also have a magnetic map sense in addition to plain magnetic orientation, but Anna Gagliardo, an avian olfaction expert at the University of Pisa in Italy, says the evidence for such a map sense is weak. And birds seem to find their way just fine using other senses. “Forty years of experiments,” she says, “and no amount of magnetic manipulation has ever stopped homing pigeons from coming home.” But she notes that the birds get lost if they have been deprived of their sense of smell by surgically cutting the nerves of their noses. Moreover, she adds, homing pigeons raised in aviaries that only open upward—so that the birds cannot tell which direction environmental scents are coming from—are unable to navigate. So whereas the evidence that birds can tell magnetic north from south is pretty solid, Gagliardo says, she doubts their magnetic sense can do much more than that.
Many other experts, however, now believe that birds have two distinct magnetic senses, each optimized for different uses—a compass sense for the field’s direction and a separate “magnetometer” sense for its strength. Others argue that various lines of evidence suggest the existence of one sense or the other but not both in a species. One reason for the discord is that pinpointing the behavioral effects of magnetism is devilishly difficult, in part because birds and other animals exploit a number of different cues for orientation and navigation—they use the sun, the stars and the moon; they can recognize landmarks on the ground and the prevailing direction of the waves at sea; and they remember smells. Animals always navigate using multiple senses, notes Michael Winklhofer, a geophysicist at Ludwig Maximilian University in Munich. “They use whatever cue is available. Whenever one is dodgy, they use a more reliable one.”
Unfortunately, even the strongest results from well-designed experiments often lend themselves to multiple interpretations. One of the Wiltschkos’ main observations was that robins’ compass sense does not work in the dark: it needs light with a blue, or short-wavelength, component. Their findings were obtained in lab conditions, which help to isolate cues from one another but are also somewhat artificial. Yet in a landmark study in 2004 Henrik Mouritsen of the University of Oldenburg in Germany and his collaborators found compelling evidence of light-compass interaction in the wild. They showed that night-flying thrushes recalibrate their magnetic sense every day at sunset.
For the experiment, Mouritsen’s team captured dozens of thrushes in central Illinois and outfitted them with radio transmitters. At sunset, the researchers exposed 18 of the birds to a magnetic field that simulated the earth’s but pointed east instead of north. After dark, they opened the cages and let the birds go. As the birds flew away, members of the team chased them in a 1982 Oldsmobile with a large antenna sticking out of the roof—which often got them pulled over by the police. While the control group resumed their migration north toward Wisconsin, the 18 birds that had been exposed to the fake geomagnetic field headed west toward Iowa or Missouri. On subsequent nights, however, even those birds corrected their path and headed north again.
Although the results indicated that the birds reset their magnetic north at dusk, interpretations varied on the role of light in that process. One possibility is that the birds have an internal compass that works only in the presence of light, as the Wiltschkos had concluded. Another explanation seems equally plausible: the birds used the sun just as a point of reference for calibrating a compass that did not physically need light to work. In fact, they might have kept using their compass all night.
Clearly, behavioral experiments alone are unlikely to settle such issues one way or another. Eventually one needs to locate and study the sensory organs more directly.
Searching for magnetically sensitive organs is an anatomist’s worst nightmare. The sensors could be single, isolated cells, located anywhere inside the body. They could contain microscopic magnetic particles—serving as the equivalent of a compass needle—that, when analyzed, would be difficult to distinguish from contaminants in tissue specimens. A candidate mechanism also has to meet stringent requirements; in particular, it must be sensitive to fields as weak as the earth’s, and it must separate the magnetic signal from the noise of natural molecular vibrations—something that is especially hard for a microscopic structure to do. So far the only mechanism that has been identified and explained unequivocally occurs in bacteria.
At latitudes where the geomagnetic field’s inclination is sufficiently steep, certain bacteria use it as a proxy for gravity to “know” which way is down so that they can swim toward muddy seafloors—their preferred habitat. In the 1970s researchers demonstrated that these bacteria contain strings of microscopic particles of magnetite—a strongly magnetic form of iron oxide—that align with one another and with the field and in the process orient the entire organisms in the right direction.
The bacteria offered a natural paradigm for trying to understand magnetic reception in general. In the 1980s geobiologist Joseph L. Kirschvink, now at the California Institute of Technology, and others proposed that similar magnetite-based structures might exist across the animal kingdom. Scientists began searching for these particles in magnetically sensitive animals.
In the early 2000s a team that included Winklhofer, Wolfgang Wiltschko, and Gerta and Günther Fleissner—another married-couple team at Goethe—used advanced imaging techniques to reveal intriguing structures lined with magnetite nanoparticles in homing pigeons. They found these structures in the skin of the birds’ upper beaks. The magnetic particles were very small—a few nanometers—and thus their random motion would have been substantial compared with their size. That noise would have been too loud for the particles to read the magnetic field’s strength, but in principle, they could have detected its direction, Winklhofer says: “You wouldn’t have a very strong response, but it would have worked at least as a compass.” Intriguingly, the structures were in regions dense with nerve endings, which is what one would expect of putative detectors because they would need to be integrated into the nervous system.
Only a few of the particles appeared to be magnetite, though; the others were a closely related material called maghemite, which is not as strongly magnetic. Still, researchers thought they might have the smoking gun.
In a follow-up paper, the Fleissners and their co-workers proposed a model for how even a structure composed mainly of maghemite could function as a compass. They suggested that the maghemite structures could temporarily magnetize and thus amplify the geomagnetic field in their vicinity, funneling it into the magnetite particles.
Winklhofer, however, parted ways from his former collaborators and, with Kirschvink, issued a rebuttal. The two researchers cited evidence that the maghemite in the study was “amorphous,” meaning that it lacked a crystal ordering; such amorphous materials make very weak magnets, Winklhofer points out—too weak to do the job being attributed to the particles seen in birds. Others note that whether the nerve endings are located precisely at the magnetic particles is unclear. The candidate structures in homing pigeon beaks may have nothing to do with magnetic reception after all, Winklhofer concludes.
One more reason to be cautious is that magnetite and other magnetic particles are ubiquitous in the environment. “Even dust from the lab contains magnetic materials,” Winklhofer says. Anatomists must use ceramic scalpels to try to avoid introducing metal fragments into tissues they extract from animals. But if the particles enter the body as contaminants, they may get scooped up by white blood cells, which would then show up in the microscope looking like possible sensory cells.
Despite the particular difficulties posed by the putative magnetic receptor in homing pigeons, Winklhofer and Kirschvink remain staunch proponents of the magnetite hypothesis. They point to what they say is the best evidence so far of such an organ: cells lining the nasal opening of rainbow trout. Michael M. Walker of the University of Auckland in New Zealand and his collaborators have been studying the cells since 1997, when they first found them. The researchers were able to demonstrate an electrophysiological response to magnetic fields: the cells actually sent a signal to the brain.
Kirschvink is now leading a multiyear, multilab effort to characterize the structure and behavior of these putative magnetic sensors. He says he suspects that the magnetite particles are contained in organelles that stick directly to the membranes of specialized neurons. Each such cell would constitute a microscopic magnetic-sensing organ. When a magnetic field causes the organelles to spin to a new orientation, they trigger the release of ions that prompt the neurons to fire and thereby “tell” the brain which way the fish should swim [see illustration on preceding page]. Perhaps, Kirschvink says, researchers who have been looking at pigeons’ beak skin should take guidance from the fish and instead search inside the birds’ snout.
Magnetite is not the only leading contender in the race: a quantum physics–based mechanism also seems plausible to many researchers. Klaus Schulten, a theoretical biophysicist now at the University of Illinois at Urbana-Champaign, observed in the 1970s that chemical reactions affected by magnetic fields could provide the physical basis for a magnetic sense. The reactions involved would initiate when photons hit suitable pigment molecules, causing the formation of so-called free radicals. The need for photons would explain the apparent sunlight-compass connection observed by biologists. In those days, however, it just sounded like a wild idea, and Schulten did not explain how the signal would be conveyed to the brain.
Then, in the late 1990s, biochemists discovered a pigment protein called cryptochrome, first in plants and later in the retinas of mammals—including humans—where it was found to occur in several variants and to help the animals adjust their day-night cycles. Schulten, together with his colleagues Salih Adem and Thorsten Ritz, a biophysicist now at the University of California, Irvine, suggested that cryptochrome had just the right properties for a compass sense and that certain cells in the retina might be able to make use of the formation of free radical pairs in it to detect the direction of the earth’s magnetic field.
Lab experiments had shown that when cryptochrome absorbs a photon in the blue part of the spectrum, the energy of the photon kicks an electron from one part of the molecule to another. For a molecule to be chemically stable, its electrons need to share orbits in pairs, but in cryptochrome the displacement leads to two electrons each flying solo. Now the two electrons, termed a radical pair, engage in an elaborate pas de deux dictated by their spins. Spin is the quantum-physics analogue of the magnetic axis of a bar magnet. Every electron’s spin interacts with the geomagnetic field and with the spins of atomic nuclei, and collectively the interactions make the electron’s spin axis precess like that of a spinning top. In a radical pair, the spin of each electron is also influenced by that of its counterpart.
During parts of the paired electrons’ performance, their spins point in roughly the same direction; at other times they point in opposite directions. Crucially, an external magnetic field, such as the earth’s, changes the relative amount of time that the electrons spend in each alignment. That is how an external field can affect cryptochrome’s chemistry: certain chemical reactions can take place only when the spins are parallel. Thus, if a field keeps the spins parallel for a longer time, the reactions will accelerate.
The speed of a spin-sensitive reaction could be the chemical signal for a sensory neuron to fire and thus send a message down a nerve to the brain center in charge of a magnetism-mediated behavior. Unfortunately, although the general principle is well known, in the case of cryptochrome no one knows what the relevant chemical reaction could be, nor how variations in its rate would induce a neuron to fire. Still, in the past decade several lines of circumstantial evidence have appeared.
Spin precession is sensitive not only to static fields such as the geomagnetic one but also to those that change rapidly in time, as in radio waves. In 2004 Ritz teamed up with the Wiltschkos and showed that radio waves disrupt the internal compasses of birds. The disruption occurred only at precise wavelengths, as would be the case if the waves were interfering with the dance of radical pairs. “From a physics perspective, so far that’s the best evidence for the radical pair mechanism,” Ritz says.
Then, in 2009, a team led by Mouritsen found that birds with lesions in a brain center that is related to vision have a hard time with magnetic orientation. And in 2010 a study of European robins and chickens led by Christine Niessner of Goethe found that cryptochrome is copiously produced not just in the birds’ retina but more specifically in their ultraviolet-light-sensitive cone cells—that is, precisely where biologists would expect it to reside, given that radical pair formation requires light.
The case is not closed, however. Most results have yet to be independently replicated. As with the magnetite candidate, some of the evidence seen to date may not be as clear-cut as it sounds. Ritz himself, for instance, cautions that radio waves induce electric fields that could disrupt biological processes in unpredictable ways. For example, the waves are known to interfere with the neurotransmitter receptors that are active in pleasure centers, and thus they could indirectly disorient the animals rather than making them lose the ability to sense magnetic fields.
University of Oxford physicist Peter J. Hore adds that the sensitivity of birds to radio waves seems too good to be true: a field just 1/2,000th the strength of the geomagnetic field is enough to disrupt their magnetic sense.
Similar confusion surrounds cryptochrome studies in fruit flies. In 2008 Reppert and his collaborators showed that fruit flies could be trained to follow magnetic fields to a sugary reward but that mutant flies missing the gene for cryptochrome, and thus unable to produce the protein, could not.
The insects, however, were exposed to fields 10 times stronger than the geomagnetic field. And because the experimenters knew when the artificial fields were turned on or off, they might have cued the insects inadvertently, Kirschvink cautions.
Overall, Hore says, although evidence has been accumulating to support the radical pair idea, “we are not there yet.” Several pieces of the puzzle are missing, starting with the particulars of the mechanism. “I find it very frustrating,” he adds. Ultimately researchers will need to demonstrate an electrophysiological response—neurons firing in response to magnetic fields—to claim they have found the seat of the new sense. Electrophysiology is the golden standard of sensory biology, Ritz notes: “That’s how we learned how vision works.”
Intriguingly, in June 2011 Reppert and his colleagues showed that fruit flies that had their gene for cryptochrome replaced with the one from the human genome still retained the ability to orient magnetically. The discovery rekindled speculation that humans may have the magnetic sense, too, although evidence in that respect is scant. Experiments that Robin R. Baker of the University of Manchester in England conducted in the late 1970s purportedly showed that people have some magnetic homing abilities, but attempts to replicate those results gave negative outcomes.
Putting It All Together
For the most part, experts have abandoned alternative explanations for the magnetic sense, finding at least one of the two leading hypotheses plausible. A possible exception is the magnetic sense of manta rays and sharks, which some say could be a bonus of the animals’ uncanny sensitivity to electric fields. These fishes have microscopic, electrically conducting canals in their skin that they use to sense voltages as weak as five billionths of a volt [see “The Shark’s Electric Sense,” by R. Douglas Fields; Scientific American, August 2007]. Because magnetic fields induce a voltage on conductors in motion, a fish could pick up the geomagnetic field just by moving left and right as it swims.
Even after the controversies are finally settled, feats of navigation by migratory animals such as humpback whales, which can swim for hundreds of kilometers at a time in the open ocean without deviating by more than one degree from the course they initially set, may still remain unexplained.
Yet many researchers are hopeful that the mechanisms of magnetic reception will soon be revealed. Experimental techniques have advanced dramatically: technology now enables researchers to track even small birds, methods for imaging microscopic anatomical structure have become more precise, and scientists from multiple disciplines have joined the effort. Once the mystery is solved, some will look back to these years with longing, Ritz says: “You don’t often have the chance to discover a new sense.”