Birds Lose Their Magnetic Maps as Scientists Reverse Direction

The swallows will still come back to Capistrano, albatrosses will wing their way across vast oceans, and homing pigeons will still arrive at home. But scientists are no longer sure how they do it.

Until last week, some thought they had a pretty good idea. Birds had both a magnetic compass and a map that they followed over impossibly long distances. But research published in the latest issue of Nature shows that map-sensing cells that were supposedly built into a bird’s beak don’t really exist. The cells that researchers thought were there are actually of a completely different kind.

“They are immune cells called macrophages, and not neurons that communicate with the brain,” says David Keays, a neuroscientist at the Institute of Molecular Pathology, in Vienna, and lead author of the new paper. “They can’t detect magnetic fields.”

“I think this knocks the field back 10 to 15 years,” says Henrik Mouritsen, a professor of biology at the University of Oldenburg, in Germany, who studies bird navigation. “That’s a good thing. Usually negative results like this don’t get published by prominent journals. But the original paper claiming these beak cells existed was cited at least 100 times, so this is a very important correction.” Another critic has charged that a separate journal has been sitting on a different debunking paper for four years, neither accepting nor rejecting it, which he says caused many researchers to waste their time.

One of the original beak-sensor researchers has launched a spirited attack on the new work—“The great amount of data conceal the bad quality of the contents,” says Gerta Fleissner, a neurobiologist at the Goethe University of Frankfurt—but the critics appear to be carrying the day. “Birds have magnetic sensors, but these beak cells aren’t part of the picture,” says Joseph Kirschvink, a professor of geobiology at the California Institute of Technology.

Two types of magnetic senses have been found in many animals, from bacteria to sea turtles. Essentially the creatures use a compass and a map to get around. The compass is formed within cells by tiny grains of a mineral called magnetite that reacts, like compass needles, to the north-south direction of earth’s magnetic field. “In bacteria, they really are sensitive, like beautiful little needles,” says Mr. Kirschvink, who helped discover them some 30 years ago. Behavioral experiments with turtles show they will turn to follow artificial magnetic fields, and turn again as those fields are reversed. Pigeons have shown similar behavior.

“It’s not the compass that I’m calling into question. It’s the map,” says Mr. Keays. Biologists have assumed some map sense must exist, because a compass isn’t good enough for finding your way. An animal traveling long distances also needs to know where it is. Magnetism can help here, too. For example, an iron-filled mountain would register as a landmark if a bird could detect magnetic intensity as well as direction.

In two papers, one published in 2003 and another in 2007, Ms. Fleissner and several colleagues suggested they had identified the intensity detector. It was a series of neurons in six locations in a bird’s upper beak, they said. The ends of these nerve cells contained particles of magnetite and maghemite, minerals that could react as a magnetic field got stronger or weaker, and transmit that information along nerve fibers to the brain.

Mr. Keays says he was interested in these pathways, and wanted to quickly identify the six sensor locations so he and his team could move on to investigate how the information was interpreted by the brain. “I chose to work with pigeons because they have this incredible homing ability, and I honestly thought the work would go very quickly,” he says. “But we couldn’t find these six cell areas.” Instead, in about 190 birds, they found a whole bunch of cells with various iron-related minerals in them. And that didn’t make sense, he says, because the abundance would drown out any specific magnetic signal.

“Then we got lucky,” Mr. Keays says. “One of our pigeons had a beak infection. And looking around it we saw lots of little blue cells.” He had used a blue stain, which binds to iron, to identify cells that contain the metal. “That made us wonder if all of these cells were actually immune-system cells.”

After slicing pigeon beaks into 250,000 very thin sections to examine them under a powerful transmission electron microscope, he decided they were immune cells called macrophages, which engulf invaders like bacteria. “We could actually see their little tentacles as they surrounded foreign bodies,” he says. Macrophages often have iron in them because they recycle it from red blood cells.

What they don’t have, and what the researchers didn’t see, was a nucleus. Neurons do. For Mr. Mouritsen, that was the smoking gun. “No nucleus means no neuron,” he says. Even one of Ms. Fleissner’s co-authors on her original paper is won over. “It is exquisitely and convincingly shown that  the cells in question are nothing but macrophages,” writes Michael Winklhofer, a biomagnetism expert in the University of Munich’s department of earth and environmental sciences, in an e-mail. (He had previously expressed doubts that the minerals originally found were magnetically sensitive.)

Pigeon beak

No magnetic-sensing cells in the flesh (purple) or bone (brown).

Ms. Fleissner is not impressed. She responds that Mr. Keays’s methods were poor, and he simply missed the cells that she saw. The sections he sliced were too big, she says, and could have missed the tiny neuron segments. And those areas should contain some neuron terminals, even if they don’t contain magnetic sensing grains, but they don’t show up in Mr. Keays’s slides, leading her to question the quality of his cell preparation. “Not seen does not mean not existing!!” she writes in an e-mail.

But Mr. Kirschvink has long had another objection to her work. Iron particles need to be highly ordered crystals to respond to earth’s magnetic field, and Ms. Fleissner originally described crystals that were oddly shaped and amorphous. “That renders them useless as magnetic detectors,” he says. Mr. Winklhofer agreed with him and in 2008, the two submitted a paper about this problem to the same journal that published Ms. Fleissner’s 2007 work, Naturwissenschaften. And then: nothing.

“We kept contacting them, asking them to accept it or reject it, but we could never get an answer over four years,” Mr. Kirschvink says. “The last time we asked was two weeks ago. For a journal to hold onto a paper that long is really unethical.” If the journal had published his paper, or let him submit it elsewhere by rejecting it, he thinks researchers like Mr. Keays could have saved themselves a lot of time. Other researchers have pointed to different cells, in fish, near their noses, that are stronger candidates for magnetic mappers—they have the right crystal structure—and he thinks Mr. Keays could have been hunting for those.

The journal editor has a different view. Sven Thatje, a senior lecturer in marine evolutionary ecology at the University of Southampton, in England, wrote in an e-mail that he had not heard from Mr. Kirschvink or Mr. Winklhofer in three years, until they reached out to him a few weeks ago because they knew about Mr. Keays’s upcoming Nature paper. ”If there had been a matter of personal urgency, who would ever wait this long?” he wrote. “Pressuring an editor 2-3 weeks in advance of a competitive publication is odd and also a little academic.” He also said that Mr. Kirschvink’s paper fell into a gap as the journal switched editors, and that he had assumed the previous editor had dealt with it.

The person whose time was supposedly wasted isn’t complaining. “What I like about this paper is that it shows science is self-correcting,” Mr. Keays says. “Someone, somewhere, will step up and set you right.”

Now, he says, “we are trying to find the real thing, the real magnetic sense organ. After all, birds must do this somehow.”

[Photo credits. Flying birds: AP IMAGES. Pigeon beak: Courtesy UCL Centre for Advanced Biomedical Imaging]

Return to Top