A lot of animals alive today have long necks. Giraffes, of course, are the best known example. But in the past, animals roamed the Earth that made giraffes look like amateur competitors in the longest neck competition.
The top prize goes to the sauropod dinosaurs, including the Diplodocus, Brachiosaurus and Brontosaurus, of which some members had necks stretching up to 11 metres (36 feet) long. Second are the marine reptiles, plesiosaurs, with necks reaching seven metres, and third is the bizarre-looking extinct reptile Tanystropheus with a three-metre neck.
Perhaps the most unusual of all the long-necked prehistoric animals are the azhdarchid pterosaurs. These gigantic flying reptiles had impressive wingspans of up to 12 metres. The largest known flying animals, these pterosaurs had necks longer than those of giraffes. But with such a large neck, exactly how they kept their heads up and kept flying has been a mystery.
Now, in a new study, we've unlocked the secret of how these large flying reptiles could support such long necks. We found the thin neck vertebrae got their strength from an intricate internal structure, unlike anything seen before.
Towards the end of the Triassic period - around 237 million years ago - a group of reptiles took to the skies. These were the pterosaurs. They were initially small creatures, with wingspans rarely more than two metres across.
But during the Cretaceous period, 145 to 66 million years ago, pterosaurs went through some fundamental skeletal changes that allowed them to achieve enormous wingspans, in some cases in excess of nine metres.
Different long-necked animals achieve their great length in different ways. Some have a greater number of cervical vertebrae - bones that lock together to form the neck. In others, each cervical vertebra is longer - called an elongated vertebra. Some animals, like sauropods, have long vertebrae and have more of them - the best of both worlds.
Propping a head up on such a long neck is a difficult thing to do. In most cases where animals have evolved a particularly long neck, there are usually significant changes in the structure of the individual vertebrae and the complex way they fit with one another. These adaptions are not seen in marine animals such as plesiosaurs, possibly because the water does all the work.
In flying animals, it's really difficult to keep the head up when it's on the end of a long neck. The skull is the largest part of the animal's skeleton, which means the neck has to carry a considerable weight.
But the stresses applied to the vertebrae are more than just the weight of the head. There's the added weight of any prey the animal captures and any stresses applied to the skull, such as from the wind. This means the vertebrae have to be extremely strong.
In land-based animals, the neck bones can be strengthened by being thickened, buttresses can be added and lots of extra tendons and ligaments can envelop the bones. But flying animals don't have the option to add extra weight, they can only have long necks by employing some remarkable engineering.
To keep flying, pterosaurs required extremely light skeletons. Over time, their bones became extremely thin compared with land-based animals. In addition the pterosaurs' bones were filled with air sacs, making them much lighter.
Previous studies have shown the azhdarchid pterosaurs - the largest ever flying animals - had necks with very little flexibility. The neck would have stretched out in front of the animal when it was flying, and likely stuck up like the neck of a giraffe while the animal was grounded.
We wanted to understand more about how much movement there was in between each vertebra. We examined the neck bones of azhdarchid pterosaurs from fossils found in the Cretaceous-age Kem Kem group in eastern Morocco. Thanks to unique preservational circumstances there, fossils have not been crushed. This means their 3D shape is preserved and they are much easier to study.
We examined five vertebrae with CT scanning. To our astonishment, we found a beautifully regular structure inside the bones, which would've helped make them strong while staying light.
Not only could we see the neural tube - the bony tube that carries the spinal cord - within the vertebra, we could also see fine struts that held the neural tube in place, arranged just like the spokes in a bicycle wheel. The spokes were arranged radially but also ran in helices winding their way along the vertebrae internally.
We examined the role of these spokes and found that stress was transferred from the outer wall of the vertebra to the inner neural tube via the spokes, and that there was an optimum number of spokes achieving a maximum increase in strength. Adding more spokes had very little effect on increasing the resistance to (say) bending.
Having a neural tube within the vertebra, instead of over the top of it, as in most animals with a backbone, significantly reduced the mass of the vertebra, making it much lighter and, surprisingly, it made it stronger too. In simple terms, it could catch bigger prey without risking breaking its neck - surely an advantage for any animal.
This is an extraordinary structure that hasn't been seen before in nature, and it changes the way we think about azhdarchid pterosaurs. They aren't just reptiles with long necks, they are enigmatic animals with a phenomenal internal structure that optimises their neck capabilities, such as strength.