Finally, a new route for the magnetic-sense field

When newcomers join the band of biologists who study the magnetic sense of animals, they have 60 years of conflict to catch up on. 

They might be warned that it’s an “unhappy area of science,” as Eric Warrant, professor and head of sensory biology at Lund University, describes it. They might hear about the infamous incident from the 2011 Royal Institute of Navigation meeting, where a new member of the field gave a talk that unraveled a finding that had already made its way into textbooks. Or they might learn about the failed attempt to replicate a key finding in fruit flies, and how that derailed 15 years of work.

But they will almost certainly hear about the two main theories of how the magnetic sense works—one based on magnetic crystals, the other on quantum chemistry—and how the two camps have been locked in a stalemate for years.

Now, new data from migratory insects is poised to provide the direct evidence the field has long needed, and a study last year in sea turtles suggests multiple mechanisms could be at play, even in the same species. Those could be enough to break the logjam and open the field back up again.

David Keays, professor of neurobiology at Ludwig-Maximilians-Universität München, who gave the game-changing talk at the 2011 meeting, is one of the many scientists who have spent the majority of their careers wanting to know how this sense works. Finally, he says, “I think we are getting closer.” 

W

Scientists first postulated in the 19^th century that birds could use magnetic fields as a compass, but experimental evidence supporting this assertion didn’t arrive until the 1960s. Even then, the study of magnetoreception was “generally considered to be outside the mainstream of real biology,” says Kenneth Lohmann, distinguished professor of biology at the University of North Carolina at Chapel Hill. “There was intense skepticism in the beginning, and a belief that magnetic effects probably were not real.”

That skepticism began to fade by the 1990s, as more researchers identified animals that could sense magnetic fields: additional species of birds, plus sea turtles, salamanders and even bacteria. As a result, the discourse shifted from whether animals detect magnetic fields to how they do it.

Then the two theories arose. The “magnetite theory” proposes that a change in an animal’s position within the Earth’s magnetic field activates tiny iron oxide crystals called magnetite inside cells. The crystals push to align with the magnetic field, which opens ion channels or triggers other signaling pathways that relay the magnetic field information to the brain. 

The second theory, called the “radical pair hypothesis,” argues that proteins in the retina called cryptochromes absorb blue wavelengths of light. This moves electrons from one molecule in the protein to another and creates a radical pair of molecules that each have an unbalanced number of electrons. In this state, magnetic fields alter how the molecules interact with each other; these alterations can then be converted into neuronal signals.

Determining if either theory was correct, however, turned out to be a herculean endeavor. 

P

“It’s basically a needle-in-a-haystack problem, and we don’t even know what the needle looks like,” Lohmann says.

For about 30 years, some proponents of the magnetite theory thought they were zeroing in on the receptor in pigeons. Two German labs claimed to have found magnetite inside a cluster of neurons near the birds’ beaks and sketched out the nerve pathway that would relay the information. People were so confident in this idea that it appeared in textbooks. 

Then came the 2011 Royal Institute of Navigation meeting.

Keays gave his talk in the last slot on the last day of the gathering. He walked through the painstaking work his lab had undertaken for the past few years. Pigeons have iron-rich cells all over their body, not just a cluster by the beak. And the ones near the beak aren’t neurons; they are macrophages, a type of immune cell. So the proposed mechanism was a bust. 

To date, no one has found a bona fide magnetoreceptor. Most evidence for or against the theories comes from behavioral experiments using a set of “definitive tests” that disrupt an animal’s magnetic sense, Warrant says. These tests generate only indirect evidence, and the behavioral responses are “very subtle,” Lohmann says, likely because magnetoreception is “almost a backup system, or the sense of last resort.” If animals can use the sun or other cues to navigate, they will.

This type of data does not lend itself well to settling arguments. Productive scientific debates rely on sufficient facts to constrain and falsify ideas, Lohmann says, which the field of magnetoreception hasn’t had.

As a result, there were “two schools of thought about how magnetoreception is done, and neither can believe that the other school of thought is correct,” Warrant says—and both believed there could be only one true mechanism. “Which is the most absurd thing, really, if you think about it. Why should evolution be constrained by what we think?”

T

And this, Mouritsen conceded, was much more likely to occur in an insect first, Warrant says. (Mouritsen did not respond to several requests for an interview.)

The bogong moth and monarch butterfly are both migratory insects that use the Earth’s magnetic field as a compass cue. They are smaller than birds, which makes the receptor search easier, but in addition, scientists can more easily manipulate their genes and record from their brains. 

One of the researchers shepherding insects into the field is Christine Merlin, professor of biology at Texas A&M University. She originally developed CRISPR gene-editing tools for the monarch butterfly to study the animal’s circadian clock, but in 2021, Merlin and her team managed to remove a key part of cryptochrome and disrupt magnetoreception. 

When cryptochrome was first raised as a candidate magnetoreceptor, it could only be studied outside an animal, using chemistry techniques. In 2008, Merlin’s former mentor Steven Reppert and his colleagues knocked out one of the proteins from Drosophila and observed that the fruit flies lost the ability to associate a magnetic field with a sugar reward. Reppert and several other labs spent the next 15 years studying how it contributes to the fly’s magnetic sense, but that effort fizzled out when a massive replication effort using more than 100,000 flies found no evidence that they can detect magnetic fields at all—a “great disappointment,” says study investigator Peter Hore, emeritus professor of chemistry at the University of Oxford.

So it was Merlin who had the first great success. With the protein not functioning in her butterflies, she saw that the insects could not detect when the angle of the magnetic field flipped in a flight simulator. Flipping the angle is like teleporting the animals to the opposite side of the world; wildtype butterflies will start flapping their wings as if they are trying to get themselves back where they belong. The cryptochrome mutants, however, act as if nothing has changed.

Now Merlin and Kayla Goforth, a postdoctoral researcher in Merlin’s lab, are removing additional parts of the cryptochrome protein, as well as other light-sensitive proteins present in the monarchs’ eyes, to sketch out the molecular pathway that converts magnetic information into a neuronal signal.

Their collaborator Basil el Jundi, of the University of Oldenburg, wants to use butterflies for a different but complementary project: to find the brain region that encodes magnetic information and trace that pathway back to the magnetic receptor. To do that, he is heading to Texas.

E

Robin Grob, a postdoctoral researcher in both el Jundi’s lab at the University of Oldenburg and at the Norwegian University of Science and Technology, spends his Texas fieldwork days at an electrophysiology rig inside an unassuming shack, recording from butterfly brains. It’s an ideal animal for the magnetoreceptor search, Grob says, because researchers have intricately sketched out and studied the central complex, the navigational hub in the insect brain, over the past 15 years. Here, different streams of sensory information come in, and a steering command comes out. “It’s just logical that it should encode magnetic information too,” Grob says.

Grob starts each experiment by exposing a butterfly’s murky white brain and inserting a few flexible electrodes inside, which snake back to a piece of recording equipment that fits in the palm of his hand. He then carefully carries the setup outside, butterfly in one hand and recorder in the other, while the electrode tails leap in the wind. They are as thin as a strand of hair. “Every wing beat could break them,” Grob says. The wind can break them, too, and the sun can melt the wax that keeps everything attached.

Next he turns to a flight simulator: a large metal cylinder, like a stockpot used for a seafood boil, with a rod resting across the top. Using a metal probe adhered to the butterfly’s thorax, Grob attaches the insect to the underside of the rod. The butterfly can flap its wings and rotate itself as much as it chooses. Now Grob can record from compass neurons in the central complex as he tweaks the magnetic field, which he does via a magnetic coil circling the simulator.

The goal is to find neurons that alter their firing in response to changes in the magnetic field. Grob hypothesizes that these neurons will have a receptive field and fire more in response to certain magnetic field characteristics, just like other sensory neurons do. If he does find some of these neurons, he and his colleagues can follow the path of the magnetic cues and “trace it backwards” to the receptor. It’s too early to make any conclusions, but the team should have a complete dataset after the 2026 field season ends in December, Grob says.

Merlin’s cryptochrome mutants provide a “perfect control” for these experiments, Grob says. “Nothing should happen, because the sensor should be knocked out.”

On the other side of the globe, Warrant plans to conduct similar field electrophysiology experiments with bogong moths in collaboration with el Jundi, and he is working on developing genetic tools as well. 

“Those systems look to us to be extremely promising, probably the best bet for breaking open the system using neurobiological techniques,” Lohmann says. 

T

The magnetic field “can potentially provide animals with two different types of information,” Lohmann says. The simplest information is used to maintain a constant heading, akin to a compass. The properties of the magnetic field vary predictably across the Earth, so each location also has its own “magnetic signature,” Lohmann says. If animals can detect these signatures, they can tell in what direction they need to go next, like a magnet-based GPS system.

That’s exactly what loggerhead sea turtles do. After the turtles hatch from their eggs on the beach, they waddle into the ocean, and when it’s time to lay their own eggs, the females return to the same region—and sometimes to nearly the exact same spot on the beach. When Goforth, the postdoc in Merlin’s lab, was an undergraduate student in Florida, she tagged sea turtles for a research project on nesting. She noticed that each egg-laying season, a turtle would emerge from the ocean “in front of the same three houses” on the coastline. “They were so precise.”

Goforth joined the Lohmann Lab in 2016 for her Ph.D. and began trying to untangle the magnetic map and compass senses. This was “groundbreaking,” Lohmann says, because the two senses often function together, so during experiments “it’s really hard to sort out whether whatever you did had an effect on the map sense or on the compass sense.”

Over the course of two months, sea turtles spent 40 minutes a day in pools with artificial magnetic signatures that mimicked those in locations around the Atlantic Ocean, created by magnetic coils. Goforth and her team fed the turtles snacks in one signature but not the other.

To gauge if the turtles could detect the difference between the signatures, the researchers watched them dance. When captive sea turtles expect food, they shimmy from side to side, beat their front flippers, open their mouths and sometimes even spin. After their bout of training, the turtles danced in the magnetic signature they had learned to associate with food; they were less active in the other signature. The turtles could discern between the faux fields of New Hampshire and the Gulf of Mexico, Cuba and Delaware, Florida and Maine, and even Haiti and Turks and Caicos, which are less than 200 miles apart.

This confirmed that sea turtles have a magnetic map sense. Next, Goforth tried to disrupt it.

She generated oscillating magnetic fields in the radiofrequency range, which would support the radical pair theory. It did not affect the turtles’ map sense, but it did impair their compass. 

If the same factor disrupted one magnetic sense but not the other, then the senses must have two different mechanisms, Goforth and her colleagues reasoned—perhaps magnetite for the map sense, and radical pairs for the compass. In other words, both mechanisms could be true, and both could be present in the same animal. 

It’s an “absolutely beautiful study,” Warrant says.

I

The magnetoreceptor field is changing in other ways, too. In the past decade, a new cohort of researchers has entered the field, and older researchers have retired. “There’s more data and less ego in the field now, if I can put it that way,” Keays says.

The idea that there are multiple mechanisms behind magnetoreception, even within the same animal, has also taken some of the pressure off. Keays, for his part, has even resurrected an older, third theory called electromagnetic induction, which suggests that moving through magnetic fields induces a voltage in pigeons’ ears. Last year, he reported that neurons in the pigeon brain respond to magnetic stimuli even in the dark, and the inner ears contain the molecular parts required for induction.

The field’s desire to limit nature to one true solution was “just arrogance, more or less,” Warrant says. “And now we have probably three mechanisms on our books, at least. And why not?”

Ultimately, it’s about simply understanding how a baby sea turtle can enter the ocean alone, make a 10,000-mile migration and return to the beach where it was born, or how swarms of monarch butterflies find their way from Canada to Mexico, generation after generation.

“I fell in love with the question, and I just really want to know the answer,” Keays says. “I’m annoyed that it’s taken this long.”