Dinosaurs Rediscovered

by Michael J Benton

  John Hutchinson came to the field in a newer generation, but he had been influenced by McNeill Alexander’s books. ‘Alexander wrote like a dream, and his insights were so clear you realized there was no limit to how we could explore the locomotion of modern animals and apply the findings to dinosaurs,’ says Hutchinson. Hutchinson is a stockier individual than Alexander, shaven-headed and built like an American football player, but gentle and madly enthusiastic, just as Alexander was. Hutchinson is famed for his TV shows in which he dissects an elephant, a racehorse, or an ostrich. He writes a blog called ‘What’s in John’s Freezer’, and he was characterized in a 2011 National Geographic blog as the man who had the most frozen elephants’ feet. Working as he does at the Royal Veterinary College north of London, he has access to exotic animals that arrive for investigation, many from local zoos. One week he may be studying whether elephants can run (the books say they can’t; John’s videos show they can, and in a very unusual way), and the next advising Hollywood dino-film producers, or helping students with their computer code in locomotion mechanics.

  Professor R. McNeill Alexander explaining how to estimate the original mass of a dinosaur.

  John Hutchinson, in unusually serious mood, surrounded by bones.

  John Hutchinson’s PhD was a classic anatomical study of the muscles of the legs and hips of birds and crocodiles:

  This gave me the groundwork to study dinosaurs. I dissected nine alligators, as well as lizards, snakes, turtles, and dozens of birds, and worked out the common shared patterns of all the muscles that power their locomotion [John says]. When I began my PhD, I was interested in how organisms have evolved. I wasn’t wedded entirely to one idea but had enjoyed the novel and movie of Jurassic Park and saw an opportunity to apply modern biomechanics methods to test how T. rex might have stood and moved. I ended up sticking with that plan throughout my PhD and fortunately it worked out pretty well. Fixing the basics of musculature, that it’s much the same across all vertebrates, was helpful.

  This is the idea of the extant phylogenetic bracket (see p. 17), which allows palaeontologists to determine most details of the leg muscles of dinosaurs with certainty based on their bracketing descendants, the crocodiles and the birds. As Hutchinson found in his dissections, all vertebrates have the same basic limb muscles, so it’s most likely the muscles of dinosaurs were the same too. The palaeontologist then simply has to determine the sizes of those muscles by looking for muscle scars on the bones, and these attachment sites can often be well defined and give a good measure of how broad a muscle was. The width or diameter of a muscle gives its strength, and the positioning of the attachment sites shows its direction, and can be used to calculate forces. But, as John explains:

  …it’s more than just the muscles. We need to understand how an animal holds its legs, how its body mass is distributed through the legs to the ground, and the sequence of movements of the joints and firing of the muscles. There are also all the different ways of moving, the gaits. We know a horse has five gaits, the walk, the pace, the trot, the canter, and the gallop. Now, do all animals show these same gaits? What about dinosaurs?

  Understanding how dinosaurs moved involves a remarkable mix of fashion and science. Until recently, the way we viewed dinosaurs – whether as lumbering, slow reptiles, or active, fast movers – depended on the prejudices of the moment. Evidence from footprints and skeletons forms the basis, and the new computational methods have allowed researchers like McNeill Alexander and John Hutchinson to test what is likely and what is unlikely. Now we know for sure how dinosaurs stood and walked, their speeds, and whether they could fly and swim. We can also then compare the science with the Hollywood glitz – did they get it right in the movies?

  Fashions in dinosaur posture and locomotion

  There have been fashions in how we reconstruct dinosaurs. How we imagine dinosaurs has evolved from giant crocodile to giant rhinoceros to kangaroo to balanced high-speed biped, to somewhat more stately biped today. It’s all about interpreting the skeleton, finding a plausible posture for the limbs, and choosing the appropriate modern examples. We may treat the older images with humour – how could they ever have believed that? – but palaeontologists of the future will likely mock our best efforts. Nevertheless, we might still hope that our hypotheses improve through time: ‘we stand on the shoulders’ of previous researchers, as Isaac Newton said.

  When dinosaurs were first reconstructed, palaeontologists thought they were looking at outsized lizards or crocodiles. About 1830, Gideon Mantell even reconstructed his Iguanodon as an enormous lizard, over 61 metres (200 feet) long, walking on all fours and with its body close to the ground. Other dinosaurs were seen then as huge crocodiles, also quadrupedal and also low-slung, but presumably sluggish in pursuit of their equally sluggish prey. As we saw earlier in this book, Richard Owen made a radical revision to the image of the dinosaur in the 1840s and 1850s, picturing them as warm-blooded, rhinoceros-like animals, portly and presumably slow-moving, but at the same time warm-blooded.

  Owen’s rhino-dino did not last long, because complete skeletons from North America showed he had got it wrong. The first hadrosaur skeleton reported by Joseph Leidy in 1858 was clearly a bipedal animal, whose legs were three or four times as massive as its arms. Leidy opted for a rather vertical posture, in which the animal sat back on its heels and the tail touched the ground; its torso then stood vertical, like that of an alert kangaroo. This standing posture prevailed until 1970. Other images of bipedal dinosaurs after 1860 were equivocal about the running posture of bipedal dinosaurs – some realized the animal should be flipped to horizontal, with the backbone and tail straight out, and the body in front of the hips balancing the tail behind – more see-saw than kangaroo. Others persisted in trying to show the dinosaur running with its body vertical and the tail trailing along the ground – an impossible contrivance that required several elements of the tail and neck to be broken. At speed, such a dinosaur would have been more Monty Python silly walker than believable moving creature.

  An early vision of Iguanodon, reconstructed as a 61-metre-long (200-ft) lizard by Gideon Mantell.

  Richard Owen’s massive, almost rhinoceros-like, vision of Iguanodon, from 1853.

  Replica cast of Hadrosaurus foulkii, the first complete dinosaur skeleton to be put on display, shown here at Princeton University’s Nassau Hall in 1898.

  The revolution came in 1970, when two young palaeontologists, one British, one American, hit on the same realization. As we saw in Chapter 4, Bob Bakker drew the small theropod Deinonychus as a lively, sleek horizontal runner, and Peter Galton did the same for the hadrosaur Anatosaurus, shown stretched horizontal and ‘in a hurry’, as he stated. The truth of their insight was grasped immediately, and the kangaroo-dinosaurs were never to be seen again, except in the cheapest of kids’ books. For Bakker and Galton, the modern comparator was not the kangaroo, but the road runner. This may be known to most as the cartoon character who tore around the US desert pursued by Wile E. Coyote, but it is also a real bird, slender and famed for running fast with tail and head outstretched.

  The dinosaur skeleton speaks in favour of Bakker’s and Galton’s insights. First, with a decent skeleton, the nature of the joints can be determined. In fact, dinosaur limb joints are as simple as those of birds and mammals, including humans, today. In the leg the knee and ankle are simple hinges, and in quadrupedal dinosaurs, the same is true of the wrist and elbow in the forelimb. Bipedal dinosaurs, like birds and humans, have arms capable of more complex movements, with rotation at wrist and elbow to allow birds to flap their wings, humans to play tennis, and apes to swing hand over hand through the trees. In all cases, the hip and shoulder joints allow rotation as well as back-and-forwards movements.

  Dinosaurs had all adopted an upright or erect posture, which distinguishes them from their sprawling ancestors. At the time of the Permian–Triassic mass extinction, 252 million years ago, all the large sprawling animals died out, and the new groups that
replaced them in the Triassic (see Chapter 1) adopted an erect gait. Sprawlers today include small animals such as salamanders and lizards, in which the arms and legs stick out to the side, and swing widely as the animal moves. The body is held close to the ground, and sprawlers cannot run fast for very long – the effort of holding the belly up is huge, and the stride length is limited. The new erect animals could strut about fast and exert very little effort holding themselves upright.

  Peter Galton’s 1970 reconstruction of Anatosaurus in a hurry.

  Indeed, dinosaurs were fundamentally bipedal from the start, and that would have been impossible if they had been sprawlers. It’s not that erect posture evolved to enable bipedalism – the immediate advantage being speed and escape from predators – but bipedalism and huge size, and ultimately flight in advanced theropods and birds, were later enabled by the erect posture. The sprawling posture in lizards means their ankles, wrists, knees, and elbows are much more complex than the hinge joints seen in erect-postured dinosaurs and mammals, as each limb goes through numerous twists and turns during a sprawling stride.

  Some of the first evidence about dinosaur locomotion, however, came from footprints, not skeletons.

  Comparison of sprawling (left) and erect (right) posture.

  What can we learn from tracks and trackways?

  Dinosaur tracks have been known for more than 200 years, even though early researchers were unsure what they were. The first published record dates from 1807, when Edward Hitchcock, a clergyman based in Amherst, Connecticut, was shown some large three-toed footprints in red sandstones from the local Upper Triassic rocks. The specimens had been observed first by a farm boy, Pliny Moody, in 1802, who, it is said, levered them up and made a doorstep out of the slab.

  Hitchcock was hooked. Over the next thirty years, he sought more specimens, and they were readily forthcoming as stonemasons opened quarries around Hartford, Connecticut, seeking red and yellow sandstones for house building. As they levered off the slabs, they saw individual footprints and whole tracks, often containing fifty or more footprints, neatly repeating left-right-left-right across the surface as far as you could trace them. Hitchcock began to collect the slabs and eventually donated them to Amherst College, where they may still be seen. He published several elaborately illustrated monographs describing his finds, most notably his 1858 book, The Ichnology of New England.

  The first dinosaur footprints to be discovered, on show in Connecticut about 1810.

  One of the Hitchcock dinosaur footprint slabs in the Amherst Museum, Connecticut; the two slabs show the underside (above) and mould (right) of the same set of four 3-toed imprints.

  Hitchcock always thought these were bird footprints – and some of them were pretty big birds, admittedly. He could not equate them with any modern reptile such as a crocodile or lizard, which generally have five fingers and toes and put the palms and soles of their hands and feet flat on the ground. The tracks from the Triassic of the Connecticut Valley were three-toed, like modern birds. Theological friends of Hitchcock’s speculated whether they were antediluvian, and had perhaps been made by Noah’s raven as it sought dry land after Noah’s flood. There were so many tracks, though, and many were so huge, that this was hardly likely – unless the raven had been twice the size of an ostrich!

  Now we know that these tracks were made by early dinosaurs, animals like the 2-metre-long (6½-foot) Anchisaurus, known from its long, slender skeletons in rocks of the same age in New England. Anchisaurus had a tiny skull and plant-eating teeth, a long neck and tail, and a slender body supported by arms and legs of similar length – it could probably flip between walking on all fours and running on its hind legs. Larger prints might have been made by relatives like Massospondylus, up to 6 metres (20 feet) long. Both these dinosaurs are sauropodomorphs, ancestors of the long-necked Brontosaurus and the other sauropods.

  Why did these animals leave so many tracks in Connecticut? It’s probable that such animals were running around in all parts of the world, but the conditions of rock deposition happened to be just right in the Connecticut Valley, as well as in footprint localities of similar age in South Africa. In New England, the conditions were unusual – hot climates, with great lakes surrounded by vegetation, and rich faunas of fishes in the lakes, small lizard-like animals snapping at cockroaches and dragonflies at the water’s edge, and the plant-eating dinosaurs crowding on the shore to find the richest plant food.

  The hazy heat of midday in that era reminds us that New England, like Germany and North Africa, lay close to the equator then, and the Atlantic Ocean was just beginning to unzip between Europe and North America. Volcanic eruptions and tearing movements between the future tectonic plates of Europe, North America, and Africa opened deep rifts, just as in the Great Rift Valley of eastern Africa today. The rifts formed natural depressions in which lakes small and large could form, and these were a magnet for insects, plants, and fishes, and for the dinosaurs that fed on these food supplies.

  After 1850, dinosaur tracks were found in increasing numbers in the Late Triassic, Jurassic, and Cretaceous worldwide. Palaeontologists became adept at matching tracks to likely makers, based on the size and shape of the toe impressions, whether the toes of three-toed prints were tipped with sharp claws (meaning theropod makers) or not (meaning ornithopod makers), or whether the prints were huge circular markings, presumably made by giant dinosaurs such as sauropods.

  As Hitchcock had noted, many of the Connecticut Valley footprints were deeply impressed in the sediment. It must have been very soft, he surmised, as the prints go down as much as 10 or 20 centimetres (4 or 8 inches), penetrating through numerous layers of rock. He could not put this amazing data to use, but locomotion experts Stephen Gatesy and Peter Falkingham were able to apply scanning and digital engineering technology to the specimens. They reconstructed the foot cycle to show the slender left foot entering the silt with all three main toes extended as broadly as possible, sinking down through the sludge until it hit firmer silt or sand, and then taking the animal’s weight as its body continued forwards, and the leg followed. Then, as the right foot grounded and took the weight, it pulled the left foot up, but with the toes now bunched tightly together and curled like an arthritic witch’s hand so they could pull free of the gloop with minimal effort.

  This entire foot cycle can be seen in the multiple layers of sediment, and in the CT scans and models – the foot goes in with toes out, grounds on firm sediment, rolls forwards as the animal keeps moving, bunches the toes tightly and pulls out, as the other foot begins its next stride.

  The slender dinosaur foot enters the sediment with toes splayed, and then withdraws, as the animal moves forward, with the toes clenched.

  How fast did dinosaurs run?

  It might seem impossible to work out how fast dinosaurs walked and ran, but in fact this was one of the most basic of calculations. The method was pointed out in 1976 by McNeill Alexander. He had noticed that there was a ‘rule’ of locomotion, which seemed to hold for bipeds and quadrupeds. It was based on a commonplace observation that as speed increases, so too does stride length – when speeding along very fast, the stride length is twice or three times the length of the walking stride.

  McNeill Alexander knew from basic biomechanics that all animals should show the same relationship between speed and size – in other words, as they become larger, they should move relatively slowly. His inspiration came from a fundamental formula of shipbuilding, of all things. In 1861, the great Victorian engineer William Froude established a fundamental law of mechanics that became essential for ship design: he noted that the resistance in a fluid to the motion of a ship is proportional to the square of its speed relative to the fluid and the area of resistance. McNeill Alexander was the first to realize that this formula could apply to a whale as well as to a ship, but it could also be modified to refer to running as much as to swimming. He found that he could describe the walking and running of a wide variety of living animals as Froude number
= 2.3 (relative stride)0.3. The speed proportional to body size is the so-called ‘dimensionless speed’, or simply ‘relative speed’.

  The relationship between relative stride length and dimensionless speed.

  Estimating speed from stride length and hip height.

  This was the basis for McNeill Alexander’s famous paper of 1976, only two pages in length, but in which he made the proposal that we could calculate dinosaur running speed with confidence from a trackway. His argument was undeniable – we have a formula1 that always works for modern animals. He tried it out on humans, horses, dogs, elephants, birds, rats…it always worked. I remember seeing his wonderful demonstration in a BBC Horizon programme back in the 1970s, when he and his somewhat put-upon family interrupted their holiday to North Norfolk to run along the beach, and then repeat with dogs and horses, while the good professor, vast untrimmed beard flying in the wind, measured their stride lengths with a ruler.