The Science of Discworld Revised Edition

by Terry Pratchett

  There were several early monitor-lizard-like carnivores in the late Carboniferous and early Permian, and their relatives came down to us as the lizards and the snakes. But to get a better idea of what these early reptiles really looked like, and behaved like, you should look at the tuatara, a genuine ‘living fossil’ from the islands off New Zealand’s coast. You can probably find it in your local zoo. It’s slow, it’s stupid, and any modern iguana or monitor can run rings round it (partly because it’s adapted to colder climes); but it’s a warning not to use any geckos or pythons or goannas you have known to inform yourself about early Permian lizards.

  Slow it may have been, but this lineage became extremely diverse. Its adaptive radiation, its explosive evolutionary diversification, simply swamped the other reptiles, turtles and mammal-like creatures. The early reptiles produced several kinds of marine lizards, of which plesiosaurs and ichthyosaurs were biggest and the most famous. However, another reptile lineage had slid back into the seas in the early Permian, creatures called mesosaurs that were related to turtles and probably lived on plankton that they sieved from the water, like many whales do now. The plesiosaurs, particularly some rather nasty short-necked crocodile-like ones called pliosaurs, were worthy opponents of the big sharks, and probably fed on mesosaurs. But the most successful marine reptiles, as fully adapted to marine life as whales and dolphins are now, were the ichthyosaurs. They flourished long before the famous dinosaurs on land, reaching their peak of size in the Triassic, as far in the past of the tyrannosaur as he was in our past. A length of 30 feet (10m) was common, and occasionally they reached 45 feet (15m).

  They were trumped in the seas by a later branch of the lizard lineage, the so-called fish-lizards or mosasaurs, which took over the seas just when the big brontosaurs and allosaurs were taking over the land. Some of them were only a foot (30cm) long, a few stretched to forty feet (12m). But all those films you’ve seen showing ichthyosaurs in the seas, tyrannosaurs on land are scientifically as (in)accurate as One Million Years BC with Raquel Welch being chased by dinosaurian monsters, the Flintstone family with their tame dinosaurs powering household gadgets, or Hamlet with a PC.

  It’s quite difficult to get the span of geological time to make sense. In his book In Search of Deep Time, Henry Gee does an excellent job of reminding us just how flimsy the ‘fossil record’ is. A few bones here; a few others five thousand miles away and ten million years later; from these we attempt to tell a story of evolutionary ancestry. It’s like claiming to have reconstructed human history from one flint flake and a half-eaten hamburger. Well, not as coherent as that, actually.

  It’s more difficult still to put the range of creatures up against the tapestry of evolutionary time to get some idea of what the dinosaurian Earth might have been like. Take sharks. There have been shark-like sharks since way before there were any reptiles; there have been those odd-looking horseshoe crabs for longer. Coelacanth fishes have been grubbing about in the dark depths off the continental shelves while the dinosaurs came and went. It’s not surprising, perhaps, that yeasts and other fungi, bacteria of several modern kinds, have been around for more than a thousand million years. We don’t expect ‘blobs’ to notice the passage of time. But coelacanths and sharks and tuataras are vertebrates – you’d think they’d be a bit progressive, evolving and changing into … whatever. But they didn’t; they just kept on doing their own thing.

  Sharks have eaten mesosaurs, have been annoyed by plesiosaurs and ichthyosaurs and cautious about pliosaurs, have eaten little mosasaurs and been eaten by big ones. There were ammonites in their seas, and belemnites and all kinds of other shelled octopuses. Then, when the big reptiles went, and all the ammonites and their friends disappeared, the sharks had the top levels of the marine food chains to themselves for tens of thousands of years. Then the mammals produced dolphins, killer whales, big whales … and the sharks just went on being sharks.

  Why didn’t sharks change? They have wonderful immune systems, which we’re just beginning to understand; they don’t get bacterial infections or, apparently, cancers. Perhaps they are not made ill by viruses, either. Though they do have lots of wormy and flukey parasites. Are today’s sharky superfish the newest, latest-model sharks? Or did the ancient ones also have marvellous immunity to disease? Was that their trick? Was that what brought them unchanged, in outward appearance at least, through such a long span of time?

  Since we can’t answer those questions, yet, let’s move to one that we can answer. What was happening on land as the Permian became the Triassic? Well, for a start, the biggest mass extinction of all, 248 million years ago. The number of species on Earth dropped by about 93%. Only 7% survived. We’re talking big species here; nobody knows what happened to bacteria or even protozoa, to nematode worms, or to rotifers. Except that a lot of protozoa have mineral shells, and got fossilised as the white cliffs of Dover, and rotifers have tiny hard jaws of characteristic shapes that a few hardy fossil-hunters collect. So we can check these out, and they give a similar picture.

  The precise cause of this mass extinction is debatable. There may have been collusion between a comet impact and massive volcanic activity, as we describe shortly. In any event, the early dinosaurs – and mammal-like reptiles and turtles, even ichthyosaurs and early plesiosaurs (but not mesosaurs) – lived through it. They were among the lucky 7%. Whatever the disaster was, it seemed great for them; it gave them just the open ecological spaces that they needed to radiate wildly. The Triassic seas were just as full of reptiles as today’s seas are full of mammals, and they stayed that way right up to the early Cretaceous. Those ocean reptiles had largely vanished before Tyrannosaurus appeared, though.

  Why were the reptiles so successful as sea creatures? There’s a persuasive biological explanation for this invasion of the seas by land animals. The story starts in the sea, moves on to land, and reverts to the sea again.

  Creatures that live in the seas experience very little gravity, if any; even heavily-armoured creatures like crabs produce forms that swim. In fact their muscles are for swimming, or for closing jaws on other creatures, or for sudden escapes. But as the descendants of these creatures came out on land, they experienced real problems; they sagged and flopped. Compare a salamander with its sprawled-out legs, which cannot support its weight, with a lizard of the same size that has strong enough muscles and bones to run. (The tuatara still sprawls). The frog’s trick, one jump at a time, is much less effective than really well-designed legs, strong limb girdles, an effective bellows system around the lungs to supply the muscles with oxygen, and a four-chambered heart to keep the aerated and depleted blood apart – all the tricks that make a monitor lizard such a good predator.

  When you’ve achieved all these, then feeding on the seashore, or even in the sea like today’s marine iguanas on the Galápagos, becomes really easy. Instead of having muscles and respiratory systems that are just adequate for marine life, you’ve got the supercharged version that terrestrial gravity required of your ancestors. Going back into the sea is now a very viable option; only sharks and octopuses are any match for previously-terrestrial lineages. So the ichthyosaurs and plesiosaurs, like today’s dolphins and whales (but less so because mammals had become warm-blooded too while they were on land) found that living in the sea was easy.

  Until they evolved further and became their own best enemies, of course, after they had radiated into hundreds of species. The reptile-eating pliosaurs were like today’s killer whales, whose main diet is other whales.

  Meanwhile, several rather different lineages of Triassic land reptiles, somewhat prematurely called archosaurs (‘ruling reptiles’), had evolved really good limb girdles, of several patterns. Two very different lineages, ‘bird-hipped’ and ‘lizard-hipped’, went on to produce all those great big dinosaurs. The lizardy lot evolved into gigantic herbivores like Diplodocus and brontosaurs (now, alas, named apatosaurs; ‘thunder-lizard’ was so much more appropriate) and gigantic carnivores like the allosaurs and
the tyrannosaurs. These all arrived at least sixty million years later, though: the early archosaurs were as far in the past of tyrannosaurs as we are in their future.

  The bird-hipped lot eventually produced those spectacularly armoured beasts that make such good film shots when tyrannosaurs are fighting them: ankylosaurs with spikes on a knobby tail, like that spiky ball on a chain that the villain wields in chivalric films; stegosaurs with bony plates, spikes down their backs; triceratops with the bony frill and three horns.

  Filmmakers always seem to make almost-purposeful mistakes: the innumerable eight-year old boys who’ve learned all the lovely names could correct them. It’s a pity that that otherwise lovely fight in Disney’s Fantasia, accompanied by Stravinsky’s Rite of Spring, is between a tyrannosaur and stegosaur. It couldn’t have happened, they weren’t contemporary. And the stegosaur never had the ankylosaur’s tail-armour, either. The late Cretaceous landscape, with the few very big dinosaurs that lived then, was doubtless an impressive scene; but the film producers’ version is no more accurate than Fantasia.

  This is hardly a surprise; after all, Hollywood has been wrong about so many other things. Scientists don’t always fare better. Right now, palaeontologists believe that tyrannosaurs were scavengers, not predators. We’ll stick our necks out and dispute that conclusion. Yes, tyrannosaurs may not have been fearsome predators, hunters … but if they weren’t, that doesn’t make scavenging the only option. They probably did something that we can’t imagine instead. We simply can’t see these animals as enormous vultures, with their tiny front feet scrabbling at a decaying corpse and that great head hidden in the abdomen of a dead sauropod like Diplodocus. We’d run away from them, whatever the scientists are saying now.

  On the Rincewind principle, you understand. Just in case.

  Other archosaurs went on to produce crocodiles and pterodactyls, and perhaps the birds – or these may have arisen from the same stock as those deinonychids made familiar by Jurassic Park, namely Velociraptor and its ilk. These probably were intelligent creatures, agile carnivores much like their portrayal in Jurassic Park. We’d certainly run away from them.

  There are a few puzzles that biologists keep puzzling at, like whether some of the big dinosaurs were warm-blooded. And why everything was so damned big in the Cretaceous. The biggest teleost (bony) fish that there has ever been lived in Cretaceous seas; it was as big as today’s whale shark. And dinosaurs flew, too. There are lots of fossil pterosaurs like Pteranodon with eight-yard (7m) wingspreads, bigger than any of today’s birds, and there are a couple of fossils of Quetzalcoatlus with wingspreads twice as big. That’s bigger than the fighter-planes of World War II, like spitfires. We have no idea how these creatures lived, but there’s no doubt that they existed.1 Unless you believe in planetary engineers with a sense of humour, as in Strata …

  This is where we should warn you about the most tempting way of thinking about these ancient creatures, and how we get it wrong all the time. Gee’s In Search of Deep Time takes us to task, showing how all of our pretty guesses about evolutionary scenarios – however well they seem to be based in the fossil record – are simply wrong, impossible.

  A classical mistake is the way we habitually think about that fish that came out of the water, whose descendants were the land vertebrates. We imagine it flopping out on the beach (and Rincewind encouraging it to go back into the water) with its developing lungs and its evolving legs … No. It had fairly well-developed legs while it was still living an underwater life, with internal gills like any perfectly respectable fish. It must have done, or it wouldn’t have worked.

  We have little idea what its legs were for at that stage, though: certainly not for walking on land. But there’s no question that these hands that we’re typing with are the descendants of those fishy legs … Just as our cough is a legacy of that fish’s crossover of its airway and foodway. It’s the pictures in our heads, of what we suppose happened, that are mistaken. They are lies-to-children that we can’t correct yet. But humans definitely evolved from a fish, and it had legs. It just didn’t use them to walk along the beach.

  Another interestingly wrong lie-to-children about evolution concerns the origin of birds. When those lovely Archaeopteryx fossils were found in the Solenhofen limestone, so well preserved that the feather imprints could be clearly seen – and the teeth and the claw-digits on the wings – it was clear that we had found the halfway stage between a reptile and a bird. It was a super Missing Link.

  Think about that for a moment: how can you find a Missing Link? Oops.

  Archaeopteryx had a long tail like a lizard, and no keel in its chest for strong flight muscles. Were it not for the feathers, it would have been classified as a small pseudosuchian dinosaur like Ornithomimus (bird-mimic). In the late Jurassic there were many little dinosaurs with bird-like features, and some really well-developed diving-bird fossils had been found in early Cretaceous rocks from about fifteen million years later. These were real birds, called Ichthyornis, and they had already lost their flight abilities, reduced their wings to (very bird-like) vestiges.

  So Archaeopteryx was a bit ‘late’, and in the 1950’s zoologists thought that it probably represented a primitive reptile-bird lineage hanging on, perhaps contemporary with much more bird-like creatures. That’s another scenario that doesn’t make sense now. Many bird-like dinosaurs have been found recently, fossils from South America and, especially, Caudipteryx and Protoarchaeopteryx from China, and these make the problems worse.2 They have feathers, but they didn’t fly. They have no wings, but they have arms with hands, sometimes with only two digits.

  So what were feathers ‘for’?

  Feathers are really remarkably complicated. They’re not at all like the scales of lizards and snakes, and it’s not easy to invent an evolutionary route by which feathers (or hairs, come to that) can have developed from scales. Some biologists who haven’t looked at feathers very carefully have imagined scales growing rather like the stage witch’s fingernails, so that they stick out like a pangolin’s scales. (That’s that silly mammalian tree-climbing anteater that looks like a big pine cone.)

  Feathers aren’t like that, though. They develop as cylinders: you can see baby feathers, called pin-feathers, on a plucked chicken from the supermarket. The scales on birds’ legs are respectable reptilian scales; perhaps surprisingly, no modern birds have structures that are halfway between feathers and scales, even though their remote ancestors had scales all over. Probably … we can’t be sure from those very old fossils. Scales are, and probably were in those ancestors, patches of keratin very like fingernails; sometimes they overlap like roof tiles. Feathers are cylinders sitting in follicles, deep pits in the skin. About a millimetre up from their deep end, they have a ring of dividing cells, called the collar, which produce the cylinder by growing it outwards. As the products move up the follicle, they turn all the cells’ productivity into making keratin, the protein of horn, nails, hair and feathers. And the cylinder wall becomes horny in a strange pattern.

  The side of the feather facing backwards on the bird produces creases that pass around on both sides of the cylinder toward the front-facing side, diving into the follicle so that they are almost parallel with the length of the cylinder. They don’t quite meet, and the tissue between their deep ends will become the stalk of the feather. The other side splits open, and the barbs of the feather – between the creases – unfurl to make the feather vane. They are much longer than the cylinder’s circumference, so a narrow pin-feather can generate a feather with a broad vane.

  Not a bit like a scale. And far more complicated. Evolution had to work hard to come up with feathers.

  And they must have evolved for a good reason, because lots of dinosaurs had feathers in various versions. Some were like down-feathers, others more like paintbrushes or feather dusters. They could have evolved for warmth. Adult velociraptors and young tyrannosaurs may have been covered in down, like baby chicks. Palaeontologist Mark Norell says that ‘W
e have as much evidence that velociraptors had feathers as we do that Neanderthals had hair’. But other scientists disagree.

  Perhaps feathers were sexual ornaments. More likely, we haven’t yet thought of whatever function they had.

  So much for stories that ‘some reptiles got feathers and became birds’.

  There’s a very general problem here, and it’s the problem the wizards are having all the time. The overall pattern of evolution, of the birds say, or of tyrannosaurs, looks very sensible. But at a deep level, deeper than the ‘common-sense story’ that the wizards are trying to use for Roundworld, we don’t understand it. We have stories to explain it, all right, but that isn’t what understanding means.

  In fact, we’re not altogether sure what it does mean, in a scientific sense. We know that apparently quite ordinary historical events ‘come to pieces’ when you try to analyse them from several directions. John F. Kennedy’s assassination is a perfect example: the bullets seem to have come from different directions, and there doesn’t appear to be one consistent story that means we can ‘understand’ what happened. We can describe the events, but the underlying causalities, like the physicists’ quantum theory and relativity theory, don’t match up.

  Evolution doesn’t just happen to one creature at a time. The entire ecosystem evolves, and as it does so, new tricks may become worth using, for a limited period of time and in a limited region of geography. A few of those tricks turn out to have useful effects that are quite different from the ‘reason’ why they evolved to begin with, and those effects may continue to be worth having long after the original reason for their appearance has ceased to apply.

  It’s not surprising, in that frame, that historical events as far back as the origin of feathers, or of birds, don’t make detailed sense either. That’s why we can’t imagine what it was like in the Late Cretaceous. Walking with Dinosaurs was beautifully made and based on up-to-date science, but ultimately it was unconvincing. The need to tell a story distorted what was actually known, and mixed it up with guesswork and wishful thinking. (You can’t be sure what colour an animal’s skin was when all you’ve got is fossil bones. And assuming it was a bit like something vaguely similar that’s alive today is cheating, not science.) Anything televisual was automatically preferred to some more mundane scenario. So we got a dinosaur soap, with everything over-dramatised.