Most people are familiar with the sad story of the dodo. This plump, flightless bird was so tasty and so tame that it was hunted to extinction within a century by Dutch sailors arriving on the shores of Mauritius.
Rather fewer people realise that this story is largely incorrect.
Tasty? Apparently not – the waste pits from the early Mauritian settlements, which were excavated for a study published in 2013, are full of animal bones from the Dutch dinner table, but there is not a single dodo bone amongst them.
We have this continuous series of tragedies, forgetting the dodo over and over again
Hunted to extinction? Unlikely – Mauritius was blanketed in thick impenetrable rainforest, and dodos deep in the heartland would have been well beyond the reach of even committed hunters.
Plump? No – the tubby bird depicted in modern reconstructions is based on illustrations probably drawn from overfed captive birds or poorly-stuffed dead specimens. In the wild the dodo was a much leaner bird.
How can such an icon of human-induced extinction be so misunderstood? The answer lies in the shameful way the dodo has been treated since the last bird died about 350 years ago. Arguably, we have lost the dodo at least twice more since then.
"We have this continuous series of tragedies, forgetting the dodo over and over again," says Leon Claessens at the College of the Holy Cross in Worcester, Massachusetts.
But perhaps no longer. Because of the work of Claessens and his colleagues – including Julian Hume at London's Natural History Museum and Kenneth Rijsdijk of the University of Amsterdam – science is finally giving the dodo the attention it deserves.
Dodos were flightless birds, related to pigeons. They evolved on the island of Mauritius in the Indian Ocean.
At most there were a few hundred people living in a coastal settlement
However, it is difficult to trace that evolutionary process. "The island has acidic soil, and is tropical and humid. It's a very unforgiving environment for fossils," says Claessens.
All we can say for sure is that the dodo evolved at some point in the last 8 million years – simply because it was 8 million years ago that Mauritius, a volcanic island rather like Hawaii, first rose above the waves.
The dodo's extinction is easier to pin down than its origins.
Dutch sailors probably first encountered the bird in 1598, marking the beginning of the end for the species.
Europe's 17th-Century scientists did not realise quite how valuable their dodo specimens were
But the sailors themselves did not make much of a contribution to the dodo's extinction, says Claessens. "At most there were a few hundred people living in a coastal settlement," he says.
The problem was more likely the ship rats and other animals they brought with them, which spread across the island, eating dodo eggs and outcompeting the birds for food.
The last confirmed sightings came in the 1660s. The dodo was lost forever.
Or, at least, the living dodo was lost. But specimens of the strange bird had already been sent to Europe for scientific study. In several museums and university collections, skeletons and stuffed dodos survived.
There used to be a complete dodo in Oxford, but they had to discard the majority of the specimen in the 1700s
Unfortunately, Europe's 17th-Century scientists did not realise quite how valuable their dodo specimens were.
The problem was that the dodo had disappeared at the wrong time. Its extinction came long before scientists were willing to accept that species really could vanish forever. Why, they argued, would an all-powerful God doom some of his valuable creations to such a fate?
The great French palaeontologist Georges Cuvier is widely credited with alerting the scientific world to the reality of extinction, but he did not do so until 1796.
This meant that 17th and 18th Century museum curators felt confident that there were more dodos out there to replace any specimens that became damaged. Also, specimen damage or loss was common, especially at a time when taxidermy was in its infancy and museum records were relatively crude. Those surviving dodo specimens gradually dwindled to a mere handful.
"There used to be a complete dodo in Oxford, but they had to discard the majority of the specimen in the 1700s," says Claessens. "They kept just the head and foot."
The British Museum also had a dodo foot. "But they lost it about a century ago," says Claessens. There is also a dodo skull in Copenhagen and part of a beak in Prague. "And that's really it."
Inadvertently, Strickland and Melville had kick-started a wave of dodo-mania
In the early 17th Century, Mauritius had been home to countless dodos. But by the early 19th Century not even one complete bird skeleton survived in a museum. The dodo was lost for a second time.
In fact, the dodo might have fallen into obscurity forever, if not for the work of two Victorian researchers, Hugh Edwin Strickland and Alexander Gordon Melville. In 1848 they published a monograph, The Dodo and its Kindred.
"It's a gorgeous book," says Claessens. "It's based mostly on what they could discover from the Oxford and British Museum specimens, and it helped make dodos rather hot in the Victorian period."
Inadvertently, Strickland and Melville had kick-started a wave of dodo-mania. This arguably reached a peak when the bird featured in Lewis Carroll's Alice's Adventures in Wonderland.
By this time Mauritius had become a British colony, and the island became the focus of intense palaeontological interest. Were there fossil dodos out there that could reveal more about the curious bird?
The marsh, known as the Mare aux Songes, was chock-full of bones
For years, the hunt for more bones was unsuccessful. Things changed in 1865, coincidentally the year Carroll published his book, when labourers began installing a rail line to one of the island's sugar plantations.
"According to stories, it was noticed that some of the labourers were collecting bones out of this marshy area, probably to grind up to use as fertiliser on their own small land plots," says Claessens. "The railway engineer took some of the bones to the local schoolmaster, George Clark, and by comparing the bones with Strickland and Melville's book they realised they had dodo remains."
The marsh, known as the Mare aux Songes, was chock-full of bones, which we now know to be about 4,000 years old. Dodo remains soon began winging their way back to Europe's great scientists.
The following year, Richard Owen in London became the first scientist to publish a monograph on the new remains.
One of the leading anatomists of his time, Owen is one of a handful of people with a legitimate claim to have "discovered" the dinosaurs. But he is remembered today as much for his nefarious schemes to undermine his contemporaries as for his brilliant anatomical work.
It's almost as if Thirioux found his dodos at the wrong time
True to form, Owen claimed the honour of describing the Mare aux Songes bones by intercepting and buying a shipment that was intended for another anatomist, Alfred Newton at Cambridge – much to Newton's annoyance.
Owen's dodo is reasonably complete, says Claessens, but there is one important caveat: it was put together using bones from several individual birds. "It's a composite," he says.
Of course, a dodo skeleton cobbled together from the remains of several individuals is better than no dodo skeleton at all, but there is only so much a composite skeleton can teach us, says Claessens. When it comes to understanding the details of how dodos lived and moved, a complete skeleton from a single bird is invaluable.
Enter Mauritian barber and amateur naturalist Louis Etienne Thirioux.
In the years around the dawn of the 20th Century, Thirioux devoted almost all of his spare time to reading natural science books and scouring the Mauritian countryside for biological treasures. Among the most valuable of his finds were a number of dodo bones, probably dating from the last millennia before the bird's extinction.
Thirioux was convinced that his finds were important. He offered to sell them to Newton, who was still working away in Cambridge 40 years after Owen had thwarted his dodo plans.
But Newton did not jump at the opportunity to finally get to grips with some new dodo bones. He valued Thirioux's entire collection at just £20, a figure so low as to be almost insulting.
Disappointed, Thirioux hung on to his dodos until his death. Today, one of his dodo skeletons is on display at the Natural History Museum in Port Louis, Mauritius. A second can be seen in the Durban Natural Science Museum in South Africa.
"It's almost as if Thirioux found his dodos at the wrong time," says Claessens.
The dodo-mania of the 19th Century had ebbed away, and the era of modern palaeobiology – with its emphasis on the importance of complete skeletons from individuals – was still decades away. Newton's offer may have been so low because there was a real sense, in 1901, that the dodo was a solved mystery. There was nothing left to discover about the bird.
But Claessens and his colleagues see things differently today.
They have begun studying Thirioux's dodos. After careful examination of the bones – including a study of their shape, size, and even their colour, which gives an idea of how they have been preserved down the centuries – the researchers have found a remarkable degree of uniformity.
For the first time we have a complete set of information
A century after Thirioux discovered the bones and insisted they were important, the scientific community finally agrees. Claessens and his colleagues can now confirm that Thirioux had discovered the near-complete skeletons of two individual dodos.
They are the only such skeletons known to exist today, and the best specimens we have for understanding the biology of the dodo. "Having these skeletons really does matter," says Claessens.
You can examine Claessens's reconstruction of one of Thirioux's dodos in this interactive graphic:
Durban Dodo Skeleton - Anatomically Correct Pose
by Aves 3D
To the untrained eye they may look similar to Owen's construction, but when it comes to reconstructing the way a dodo actually walked, for example, your starting point really has to be the skeleton of an individual.
"Now we have the upper leg bone that belongs to the lower leg bone that belongs to the middle foot bone and so on," says Claessens. "For the first time we have a complete set of information."
What's more, Thirioux's skeletons are exquisite in their detail.
Previously unknown tiny bones from the wing tips are present, and so are the dodo's kneecaps: small bones that played an important role in how the hind limbs moved.
"Having all of these little bones is important scientifically, but it's also about finally giving the dodo – this icon of human-induced extinction – its due," says Claessens.
The marsh was this big soup, a jumble of bones mostly from giant tortoises
That new scientific respect for the dodo extends to the Mare aux Songes site where so many 4000-year-old dodo bones were discovered in the 1860s.
Sometime in the first half of the 20th Century, the marsh, which was a breeding ground for mosquitoes, had been filled in. Its exact location then faded into memory. It was finally rediscovered and systematically excavated by Rijsdijk, Hume and their colleagues from 2005 to 2007.
"Only then could we say for sure there were no articulated remains there," says Claessens. "The marsh was this big soup, a jumble of bones mostly from giant tortoises." There were plenty of dodo bones too, but most from the bird's sturdy legs, and no complete individual dodo skeletons.
Even with the recent research, though, the dodo has not yet been fully rediscovered.
It remains a mystery exactly where Thirioux found his two near-complete specimens. Claessens and Hume have read all of his records and correspondence looking for clues, but without success.
We may have rediscovered the importance of Thirioux's dodos, but the sites they came from – arguably the most important in the dodo story – remain a mystery. In rediscovering the importance of Thirioux's dodos, the research community is realising that, for a third time, we have lost the dodo.
The hope is that unlike the first and second loss, this third loss is not permanent.
"I have lots of optimism that we will one day discover Thirioux's collection sites," says Claessens. "But it's not going to be easy."