Scientists have sent robot scouts into deep space and unravelled the human genome, yet were forced to admit they were still baffled by how homing pigeons navigate.

It is very clear that birds and a large number of other species detect the earth’s magnetic field

Experts at Vienna’s Institute of Molecular Pathology said they had overturned claims that the birds’ feat is due to iron-rich nerve cells in the beak that are sensitive to earth’s magnetic field.

“It was really disappointing,” molecular biologist David Keays said after years of endeavour. “The mystery of how animals detect magnetic fields has just got more mysterious”.

Dr Keays’s team used 3D scanners to search for the cells and sliced pigeon beaks into 250,000 wafer-thin slivers for analysis.

They found that the particles credited with the pigeon’s homing skills were actually macrophages, a type of white blood cell which protects the birds from infection and had no connection to the brain.

“They are not excitable cells and cannot produce electric signals which could be registered by neurons (brain cells) and therefore influence the pigeon’s behaviour,” the researchers said.

Nor are these cells exclusive to the beak.

Dr Keays described the process as “extremely frustrating” but insisted the findings should not be seen as a setback.

“It puts us on the right path to finding magnetic cells,” said Dr Keays.

“It is very clear that birds and a large number of other species detect the earth’s magnetic field, so they must have a population of cells somewhere that allow them to do this... Hopefully now we can find the real ones.”

Other theories suggest the birds also get a navigational fix from sunlight or from landmarks. As for the other scientists: “I don’t think they are going to like me very much,” Dr Keays admitted.

One of the authors of the original study published in 2000, Guenther Fleissner from the University of Frankfurt, fired an angry rebuttal.

He accused Dr Keays’s team of lacking thoroughness in the search for cells in the upper beak, called sensory dendrites, that − if behavioural research is right −can detect a magnetic field.

“It is in fact a hard job to find the iron-containing dendrites and therefore it is not astonishing at all that these authors obviously have overlooked them,” he said in an e-mail.

“The paper seems to be biased by the authors’ preferred view that these dendrites in reality are macrophages only.”

The new paper appears in the British science journal Nature.

Dr Keays said magnetoreceptors were so hard to find because they were so small, probably about 20 to 40 nanometres (20 to 40 billionths of a metre), and “could be anywhere in the pigeon”.

“Trying to find a magnetoreceptor is not like trying to find a needle in a haystack, it’s like trying to find a needle in a haystack of needles.”